Handbook of Precast Segmental Tunnel

Lining Systems

This comprehensive handbook covers all aspects of design, production and construction of
precast concrete tunnel segmental lining, with the best practices in the field included in one
book for the first time.
New and current design methods and quantitative analyses are considered in line with
ACI and ASTM codes, as well as a full selection of global standards for the reliable design
of the product and all components. Also incorporated are new applications of science and
technology, such as new admixtures, and the latest manufacturing processes and precisions,
such as tight dimensional controls and high repeatability cycles.
With detailed guidance from world-​ leading practitioners, this is the definitive inter-
national technical and practical manual on these linings, forming a one-​stop reference for
tunnel engineers and an invaluable resource for advanced students in civil, mechanical and
mining engineering.
Handbook of Precast Segmental
Tunnel Lining Systems

Edited by Verya Nasri, David Klug,

Brian Fulcher, and James A. Morrison
Cover image: Verya Nasri, David Klug, Brian Fulcher and James A. Morrison
First edition published 2025
by CRC Press
2385 NW Executive Center Drive, Suite 320, Boca Raton FL 33431
and by CRC Press
4 Park Square, Milton Park, Abingdon, Oxon, OX14 4RN
CRC Press is an imprint of Taylor & Francis Group, LLC
© 2025 selection and editorial matter, Verya Nasri, David Klug, Brian Fulcher and James A. Morrison;
individual chapters, the contributors
Reasonable efforts have been made to publish reliable data and information, but the author and publisher
cannot assume responsibility for the validity of all materials or the consequences of their use. The authors
and publishers have attempted to trace the copyright holders of all material reproduced in this publication
and apologize to copyright holders if permission to publish in this form has not been obtained. If any
copyright material has not been acknowledged please write and let us know so we may rectify in any future
reprint.
Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced,
transmitted, or utilized in any form by any electronic, mechanical, or other means, now known or hereafter
invented, including photocopying, microfilming, and recording, or in any information storage or retrieval
system, without written permission from the publishers.
For permission to photocopy or use material electronically from this work, access www.copyri​g ht.com or
contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923, 978-​7 50-​
8400. For works that are not available on CCC please contact [email protected]
Trademark notice: Product or corporate names may be trademarks or registered trademarks and are used
only for identification and explanation without intent to infringe.
ISBN: 978-​1 -​0 32-​4 5330-​9 (hbk)
ISBN: 978-​1 -​0 32-​6 7553-​4 (pbk)
ISBN: 978-​1 -​0 32-​6 7554-​1 (ebk)
DOI: 10.1201/​9 781032675541
Typeset in Sabon
by Newgen Publishing UK
Contents

Preface vii
List of contributors ix
About the editors xiii

1 Precast segmental tunnel lining history 1

LEAD AUTHOR: BRIAN FULCHER
CONTRIBUTING AUTHORS: MARTIN KNIGHTS, JOHN HART, RICK P. LOVAT, AND
JOEL BOT

2 Geotechnical investigations for segmental tunnel lining design 53

LEAD AUTHORS: FRANCIS J. ARLAND AND JAMES A. MORRISON
CONTRIBUTING AUTHORS: MICHAEL LAW, ANDREW KLAETSCH, GARY KRAMER, AND
RITIKA KUNDU

3 Analysis and design of precast segmental linings 97

LEAD AUTHORS: MEHDI BAKHSHI AND VERYA NASRI
CONTRIBUTING AUTHORS: MAHDI SOUDKHAH, MONTAZAR RABIEI, PEGAH JARAST,
AND TOUFIC MAALOUF

4 Precast concrete technologies 254

LEAD AUTHORS: BARZIN MOBASHER, JIM LINDSAY, AND LOUIS FALCO
CONTRIBUTING AUTHORS: STEPHEN FARRINGTON, STEVE SCHAEF, AND
OKAN DUYAR

5 Fiber reinforcement in precast concrete segments 293

BILL GEERS AND BENOIT DE RIVAZ

6 Connections and accessories for segmental tunnel linings 356

LEAD AUTHORS: CHRISTOPHE DELUS, SANDRINE ORDENER, AND IVICA DUZIC
CONTRIBUTING AUTHOR: MATHIEU JANES

7 Gasket systems for sealing segmental tunnel linings 407

LEAD AUTHOR: ANDREAS DIENER
CONTRIBUTING AUTHOR: PETER TIEDERMANN
vi Contents

8 Formwork systems for precast segmental tunnel linings 457

LEAD AUTHOR: STEFAN MEDEL
CONTRIBUTING AUTHOR: FLORIAN WERRES

9 Fabrication of precast segmental tunnel linings 511

LEAD AUTHOR: CARLO CATTELAN
CONTRIBUTING AUTHORS: MARC DROUIN, FRED ESCH, AND VAN TUAN CHAU

10 Transportation, handling and installation of precast segmental

tunnel linings 588
LEAD AUTHOR: SHANE YANAGISAWA
CONTRIBUTING AUTHOR: WERNER BURGER

11 Annular backfill grouting of precast segmental tunnel linings 629

LEAD AUTHORS: SHANE YANAGISAWA, NESTOR GARAVELLI, AND JOEL JOAQUIN
CONTRIBUTING AUTHORS: JEFFREY CHAMPA, RICHARD MCLANE, MIRKO MARTINI,
BRADLEY KRUMEL, CHRISTOPHE BRAGARD, AND DENNIS ARBOUR

12 Durability and service life of precast segmental linings 709

MEHDI BAKHSHI AND VERYA NASRI

13 Innovative products and applications 767

LEAD AUTHOR: DAVID KLUG
CONTRIBUTING AUTHORS: JON KANESHIRO, RICHARD CUBETA, DAVID GREEN, AND
RALF WINTERBERG

Index 825
Preface

After several decades of continuous progress, mechanized tunneling has now become the
main method applied in almost all types of geology, all sizes, and in increasingly shorter
lengths. The predominant excavation support system for TBM tunnels consists of a one-​pass
precast concrete segmental liner, which in addition to typical use in soft ground contexts, is
also increasingly applied in competent rock settings to better manage the geotechnical and
hydrogeological risks.
The design, production and construction of the tunnel precast concrete segmental liners
is now one of the most complex and advanced fields in civil engineering. However, there is
currently no publication available that covers all these aspects, and this was the main reason
I decided to write this book. During more than three decades work on major tunneling
projects around the world, I gained a wide range of experience on various methods of seg-
mental liner design, production and construction. The difference in practice in various parts
of the world and the lack of one document covering all basic aspects and best practices were
among the main reasons I felt the need for writing this book.
I started first by suggesting and authoring the guidelines for the American Concrete
Institute (ACI) and for the International Tunneling Association (ITA). Then I started think-
ing about authoring a textbook for university students at undergraduate and graduate levels
and young engineers to allow them to learn all necessary details from a single document
instead of spending multiple years in the tunneling industry to become familiar with the
basic principles of segmental tunnel linings. Such a book could obviously provide young
engineers with a significant advantage when they start working in the tunneling industry
and in addition, allows the standardization and the use of best practices in various parts of
the world.
To achieve this goal, I asked David Klug, a leading figure in the tunneling industry with
four decades of experience with all aspects of tunnel segmental liners in North America and
Europe to join me in this effort. Then David and I asked Brian Fulcher and James Morrison
to join our team in writing and editing the book chapters, making full use of their decades-​
long experience in this field. Brian and Jim are very well known in the tunneling industry in
North America and have worked on many large and challenging TBM projects with differ-
ent types of segmental liners on both design and construction sides. Finally, to author the
state-​of-​the-​art and the most comprehensive document on this topic, we decided to involve
the very best international experts in various aspects of design, production and construc-
tion of tunnel segmental liner from consulting, manufacturing and contracting industries to
write the chapters and sections of this book.
The book is structured in 13 chapters describing practically all details related to pre-
cast concrete segmental tunnel linings. Chapter 1 discusses the history and evolution of
viii Preface

segmental tunnel liners and explains different steps and breakthroughs in their technical
and commercial developments, which provides a complete background context for all read-
ers. In Chapter 2, fundamental geotechnical aspects relevant to the design of tunnel seg­
mental linings are briefly explained, which allows understanding of basic design concepts.
Chapter 3 covers all necessary details for analysis and design and provides in-​depth infor­
mation allowing a young engineer to start designing tunnel segmental linings. All required
elements of concrete technology are presented in Chapter 4 by one of the best specialists
in this field. Chapter 5 describes the use of fiber for reinforcing segmental linings, which
has become the preferred solution for most cases. Highly engineered connections play an
important role for this type of precast structure and that is why their state-​of-​the-​practice
data is presented in Chapter 6. The latest developments for segment waterproofing gaskets
are described in Chapter 7. The gasket is a critical element in success of a tunnel project.
Chapter 8 details all aspects of segment formwork manufacturing while Chapter 9 discusses
all practical details of segment production by the very top specialists in the field. All essen-
tial information related to segment transportation in different production and construction
phases are described in Chapter 10. Chapter 11 provides the required practical details for
backfill grouting, which is a critical part of the tunnel lining performance. Durability of
the permanent tunnel lining, which is increasingly becoming an important element of the
tunnel design is discussed in Chapter 12. Additionally, this chapter describes the practical
methods needed to design a segmental tunnel lining based on the project required design
life. Finally, Chapter 13 presents the latest developments and innovations in various aspects
of segmental tunnel linings, which results in further expanding its use in global tunneling
industry.
I believe the authors and contributors to this book have achieved their goal of covering
for the first time, all aspects of design, production and construction of precast concrete tun-
nel segmental lining and the best practices in this field in one consolidated document. I am
confident this book will become a significant educational resource for tunnel engineering
students and practitioners all over the world.

Lead Author and Editor

Verya Nasri, PhD, PE
Chief Tunnel Engineer, AECOM
Contributors

Dennis Arbour Jeffrey Champa

AMIX Systems Master Builders Solutions
Abbotsford, BC Beachwood, OH
Canada USA

Francis J. Arland Van Tuan Chau

Mueser Rutledge Consulting Technopref Industries
Engineers, PLLC Montreal, Quebec
New York, NY Canada
USA
Richard Cubeta
Mehdi Bakhshi SolidCast Polymer
AECOM Technology
New York, NY Houston, TX
USA USA

Joel Bot Christophe Delus

Delve Underground Optimas OE Solutions
Vancouver, BC Sarreguemines
Canada France

Christophe Bragard Andreas Diener

Traylor Brothers, Inc. Cordes GmbH and Company
Long Beach, CA Boizenburg /​Elbe
USA Germany

Werner Burger Marc Drouin

Herrenknecht AG Technopref Industries, Inc.
Schwanau Alexandria, VA
Germany USA

Carlo Cattelan Okan Duyar

Technopref Industries, Inc. Precast and MCP Specialist
Montreal, Quebec Stoney Creek, Ontario
Canada Canada
x List of contributors

Ivica Duzic Pegah Jarast

Leviat GmbH AECOM
Langenfeld New York, NY
Germany USA

Fred Esch Joel Joaquin

Technopref Industries, Inc. Master Builders Solutions
Alexandria, VA Mount Sinai, NY
USA USA

Louis Falco Jon Kaneshiro

United Underground Precast & Private Consultant
Infrastructure, LLC San Diego, CA
Salem, NH USA
USA
Andrew Klaetsch
Stephen Farrington Mueser Rutledge Consulting
Master Builders Solutions Engineers, PLLC
Beachwood, OH New York, NY
USA USA

Brian Fulcher David Klug

Delve Underground David R. Klug & Associates
Pasadena, CA Wheeling, WV
USA USA

Nestor Garavelli Martin Knights

Frontier-​Kemper Constructors Inc London Bridge Associates Ltd
North Vancouver, BC London
Canada United Kingdom

Bill Geers Gary Kramer

Bekaert Corporation Hatch
Valrico, FL Mississauga, Ontario
USA Canada

David Green Bradley Krumel

Master Builders Solutions Kelley Engineered Equipment, LLC
Beachwood, OH Gretan, NE
USA USA

John Hart Ritika Kundu

CSI Tunnel Systems McMillan Jacobs Associates
Vienna, VA Washington DC
USA USA

Mathieu Janes Michael Law

Optimas OE Solutions Mueser Rutledge Consulting Engineers, PLLC
Sarreguemines New York, NY
France USA
 List of contributors xi

Jim Lindsay Sandrine Ordener

Master Builders Solutions Optimas OE Solutions
Portland, OR Sarreguemines
USA France

Rick P. Lovat Montazar Rabiei

Lovat Tunnelling Solutions AECOM
Vaughan, Ontario New York, NY
Canada USA

Toufic Maalouf Benoit De Rivaz

AECOM Bekaert Underground Solutions
New York, NY Zwevegem
USA Belgium

Mirko Martini Steve Schaef

Lane Construction Company Master Builders Solutions
Chantilly, VA Beachwood, OH
USA USA

Richard McLane Mahdi Soudkhah

Traylor Bros. Construction AECOM
Long Beach, CA New York, NY
USA USA

Stefan Medel Peter Tiedemann

Herrenknecht Formwork SEALABLE Solutions GmbH
Technology GmbH Waltershausen
Schwanau Germany
Germany
Florian Werres
Barzin Mobasher, Ph.D., P.E., FACI VMT GmbH
Arizona State University Bruchsal
Tempe, AZ Germany
USA
Ralf Winterberg
James A. Morrison BarChip Inc.
STV Inc. Sungai Buloh, Selangor
Omena, MI Malaysia
USA
Shane Yanagisawa
Verya Nasri Osprey Engineers LLC
AECOM Edgewater, MD
New York, NY USA
USA
newgenprepdf

About the editors

Verya Nasri is Senior Vice President and Chief Tunnel Engineer at AECOM. He was for-
merly Tunnel Practice Leader for URS Corporation, and research scientist at MIT. He has
published more than 200 journal and conference papers mainly in tunneling.

David Klug is President of David R. Klug and Associates, Inc. and Klug Construction
Systems, LLC. He is the Past Chairman of the Underground Construction Association
of SME.

Brian Fulcher is Principal and Tunnel Practice Leader at Delve Underground. He is the
Past Animator of the International Tunnelling Association’s Mechanized Tunnelling
Working Group.

James A. Morrison is Chief Technical Officer of Gateway Development Commission and

former Vice President at STV. He is the Past President of the Deep Foundations Institute.
Chapter 1

Precast segmental tunnel lining history

Lead author: Brian Fulcher
Contributing authors: Martin Knights, John Hart,
Rick P. Lovat, and Joel Bot

1.1 INTRODUCTION
Precast concrete segmental tunnel linings have been in existence for in excess of 100 years
but have only gained popular acceptance and use in the past 40 years or so. During these
most recent years, precast segmental tunnel liners have undergone considerable changes
and improvements in design and materials to the point that they are presently a highly engi-
neered, precision manufactured concrete product with very reliable physical and structural
properties. Additionally, the design and fabrication of precast concrete segmental tunnel lin-
ings are intimately linked to tunnel boring machine (TBM) design and operations. As will be
discussed in greater detail in the following chapters, modern day precast concrete segmental
tunnel linings have substantially benefited from high-​quality manufacturing processes and
equipment in conjunction with highly refined materials that did not exist decades ago.
The following describes the history and technical developments for precast concrete seg-
mental tunnel linings from the earliest concepts including materials and manufacturing
methods, through its progressive preference over the use of cast iron and fabricated steel
segments. During this period, many innovations were achieved related to design, materials
and manufacturing processes.

1.2 TUNNEL SEGMENT DEVELOPMENT TIMELINE

While the original patent for precast concrete segmental tunnel lining was first granted to
James Henry Greathead in 1874, little technical or commercial development was done until
about 1900. At about the same time in 1902 (and subsequent years), Walter C. Parmley
patented his “Concrete Blocks, The Parmley System” precast tunnel liners. In the same
period, in the United States, Albin F. Mattson patented his design for “Tunnel-​Lining of
Reinforcing Concrete Blocks”, Figure 1.1 above. In New York, in 1912, American John
F. O’Rourke patented his “Interlocking Tunnel Blocks” for use on large and small tunnels
for different applications.
Later, additional designs for precast concrete segmental tunnel liners were made in the
United Kingdom by C. V. Buchan, Charcon and Kinnear Moodie. Widespread use of precast
concrete segmental liners eventually outgrew in use over both cast iron and fabricated steel
in the 1970’s. Still, there were local preferences influenced by regional supply of materials
and skilled labor. Additionally, precise molds with very tight tolerances and integrated cast-
ing operations have been developed. While some cast iron is still in use (i.e. mining shafts

DOI: 10.1201/9781032675541-1
2 Handbook of Precast Segmental Tunnel Lining Systems

Figure 1.1 A .F. Mattson patent for precast tunnel lining segments.
Note: This patent describes one of the earliest precast concrete segmental lining designs devel-
oped in the United States for shield-​d riven tunnels in soft soils. (US Patent No.830,345 drawing
dated 1906).

and some transit tunnels in Russia), precast concrete segmental tunnel linings are univer-
sally accepted for both rock and soft ground tunneling conditions, for both initial and final
lining requirements.
Figure 1.2 illustrates the overall timeline for development of precast concrete segmental
tunnel liners. While not all activities and tunnels are listed, it provides a summary of the
key developments and application stages from the first patents, through to modern day
use on a wide variety of tunnels for municipal utility systems and transportation needs.
It should be noted that whereas precast concrete eventually overtook the use of cast iron
 Precast segmental tunnel lining history 3

Figure 1.2 G eneralized timeline for precast segment tunnel lining.

Note: Summary of the principle events for the design, testing and implementation of precast
segmental tunnel lining systems. Significant technical achievements have been made to improve
durability as well as economical fabrication, installation and materials. During the same period,
both cast iron and fabricated steel segmental tunnel linings have declined in use.

as the preferred material in soft ground tunneling applications, it has also gained popular
acceptance for ground support needs in rock tunnel applications and more recently for shaft
linings. Figure 1.3 illustrates one of the earliest precast segmental tunnel lining installations
in a small diameter hand shield-​driven tunnel.
The following summary was excerpted from “Use of Precast Segmented Concrete Lining
for a Tunnel in Soft Clay”, by J. D. Morton, D. D. Dunbar and J. H. L. Palmer (1978).

Tunnel lining in the form of cast iron segments expanded and bolted together were
in use as early as 1869. By 1920, 50 years later, concrete segments had been intro-
duced (Donovan, 1974) but it was not until cast iron was in short supply about 1936
that concrete segments became more widely used (McBean and Harries, 1970). This
increased use of concrete was paralleled by an increase in the number of segment design
alternatives.

In Europe and in the United Kingdom, precast concrete segments have been widely used
by tunnels in so-​called “soft ground” where, usually the ground was self-​supporting for
sufficient time to erect each ring. Use of segments, particularly unbolted segments, in
ground that is not self-​supporting without resorting to compressed air is a very recent
innovation.

In North America, the traditional tunnelling format even for ground that is self-​supporting
for a limited period as comprised a two-​stage lining system in which steel liner plates
4 Handbook of Precast Segmental Tunnel Lining Systems

Figure 1.3 E arly precast segmental tunnel lining.

Note: Early precast segmental tunnel lining used in a soft ground tunnel with a shield. No appar-
ent bolts, dowels or gaskets used.

or steel ribs and timber lagging are erected as the primary lining, followed by a cast-​in-​
place concrete secondary liner. Both the primary and the secondary linings are independ-
ently designed to be self-​supporting to comply with commercial and safety legislation.
The responsibility for design of the primary lining rests with the contractor, whereas
the responsibility for design of the final liner rests with the Owner (and the Owner’s
Engineer). This procedure results in over-​design and is an increasingly significant cost
escalator for the estimated overall project cost.

1.3 TUNNEL-​S HIELD-​DEVELOPMENT-​R ELATED SEGMENTAL TUNNEL

LININGS

1.3.1 Peter W. Barlow –​ tunnel shield design and patents

Peter W. Barlow was the first to patent in 1864, a shield capable of motion in one piece,
and surrounded by a thin cylinder of iron to build in successive rings, a full-​circular cast
iron tunnel lining. Barlow’s key innovations were making the shield circular, increasing its
strength, and lining the tunnel behind the shield with iron rings instead of Brunel’s brick-
work, which was stronger and faster.

1.3.2 James Henry Greathead –​ tunnel shield design and patents

Greathead worked closely with the engineering consultants and brothers Peter and William
Barlow. Greathead’s great accomplishments were considered the cutting-​edge technology of
 Precast segmental tunnel lining history 5

Figure 1.4 Image from Greathead’s Patent GB 1738 of 1874.

Note: Cross-​s ection view of the Greathead tunnelling shield of 1874 illustrating the use of seg-
mental tunnel lining for ground support.
Source: British Patent No.1738 drawing dated 1874 from the British Library. 

his day. Greathead’s achievements went further than any other engineers’ tunneling with
additional patented processes and materials. Greathead’s patent of 1874 (see Figure 1.4)
further improved the shield by using water pressure or compressed air to force debris back
from the shield face into the tunnel, using hydraulic rams to force the shield forwards, and
introducing the use of compressed air in the tunnel to further reduce the risk of collapse in
very soft soil conditions. Another patent by Greathead was the addition of the grouting pan
at ceiling height that allowed cement grout to be injected hydraulically behind the cast iron
segments to stabilize the tunnel wall outside the shield itself. (This appears to be the initial
application of annular grouting behind assembled segments.)
The shield was used for the construction of the corridor between the Waterloo and City
Line and the Central Line (Figures 1.5 and 1.6). The red structure is the remains of a
Greathead style shield used at the tunnel head when the Waterloo and City Line was origin-
ally built. It was left at this point when the railway was completed in 1898 and rediscovered
in 1987 during construction of the DLR extension to the Bank Station.
Greathead was accredited with the first shield construction for what is now known as
the Tower Gateway complex in 1869. According to William C. Copperthwaite, who once
worked under Greathead, both Greathead in England and Alfred Ely Beach in New York
invented and constructed their own versions of tunneling shields simultaneously and inde-
pendently of each other. Copperthwaite’s comprehensive publication, Tunnel Shields and
the Use of Compressed Air in Subaqueous Works (1912) recorded many of the innovative
works by Greathead.
6 Handbook of Precast Segmental Tunnel Lining Systems

Figure 1.5 Greathead tunneling shield.

Note: Tube excavation “by means of a Greathead Boring Shield”, circa 1930.

Figure 1.6 G reathead tunneling shield –​ remnant.

Note: Tunnel shield and cast iron tunnel lining remnant found and preserved within a pedestrian
passageway tunnel in London.
 Precast segmental tunnel lining history 7

Figure 1.7 Parmley system for precast tunnel construction –​ circular cross-​s ection.
Note: Ten-​foot diameter water tunnel in Cleveland, Ohio. Precast segment erector machine
was mounted in a separate track and was operated by compressed air and electric motors. Unit
weight of precast concrete block were approximately 485 kg (1,075 lbs).
Source: The Parmley System of Arch Construction; Catalogue E (1927). 

1.4 WALTER C. PARMLEY –​THE PARMLEY SYSTEM FOR PRECAST

TUNNEL LININGS
Walter C. Parmley was an American engineer who developed and patented a design for pre-
cast concrete segmental tunnel liners in 1902, which Parmley believed was one of the most
efficient methods by which sewers and pipelines could be built. The Parmley system was a
comprehensive system of reinforced concrete constructions (monolithic, segmental and unit
pipes). These were used for sewer and water tunnels up to approximately 3.0-​m (10-​foot)
diameter. These tunnels, however, did not gain wide acceptance for larger tunnels including
transportation tunnels. Just the same, they provided a very attractive alternative solution
when compared to other materials such as wood and cast iron linings for specific applica-
tions (Figures 1.7 to 1.11).
Over the course of several decades, Walter C. Parmley patented many precast concrete
products having special emphasis on sewer and tunnel construction. None of these, how-
ever, were related to tunnel excavation using shield or tunnel boring machines. Parmley’s
patents are listed in Table 1.1.

1.5 JOHN FRANCIS O’ROURKE –​ INTERLOCKING TUNNEL BLOCKS

John Francis O’Rourke was an American engineer born in Tipperary, Ireland. He emigrated
to the United States with his family in 1856. He later attended Cooper Union in
New York City and obtained a Civil Engineering degree. Throughout his career, O’Rourke
was heavily involved with the tunneling design and construction business, initially with John
F. O’Rourke Engineering and later with O’Rourke Tunnelling Contractors. He developed
8 Handbook of Precast Segmental Tunnel Lining Systems

Figure 1.8 Parmley system for precast tunnel construction –​ oval cross-​s ection.
Note: 24” × 36” (60 cm × 90 cm) Standard English egg-​s haped sewer sections. Parmley Reinforced
Concrete Segmental linings were available in all sizes from 24” × 36” (60 cm × 90 cm) and larger.
For a given wall thickness, they had greatly increased strength against collapse than other circular
sewer designs and materials.
Source: The Parmley System of Arch Construction; Catalogue E (1927). 

patented designs for precast concrete segmental tunnel liners starting with his US Patent No.
1,043,348 in 1912. Additional patents included those listed in Table 1.2.
Over the course of several decades, John Francis O’Rourke patented many precast con-
crete products having special emphasis on sewer and tunnel construction. None of these,
however, were related to tunnel excavation using a shield or tunnel boring machines.

1.6 EARLY APPLICATIONS OF PRECAST SEGMENTAL

TUNNEL LINERS
Although not well documented for sewer tunnels until approximately 1930 (Kinnear,
Moodie & Co.), the use of precast segmental tunnel linings did receive some notoriety
in the early 1900’s. The following describes two tunnels; one built and the other pro-
posed, that used precast segmental tunnel linings in difficult ground conditions and well
before modern TBMs had been invented that integrated modern precast segment building
technologies.
 Precast segmental tunnel lining history 9

Figure 1.9 Parmley system factory mock-​u p for precast concrete tunnel segment assembly.
Note: Segments for a 3.05 m (10-​foot) diameter water tunnel in Cleveland, Ohio. Precast seg-
ment erector machine was mounted in a separate track and was operated by compressed air and
electric motors.
Source: The Parmley System of Arch Construction; Catalogue E (1927). 

Figure 1.10 Isometric view of the fully assembled parmley precast tunnel system.
Note: “Parmley reinforced concrete segment plans, using solid concrete walls with all steel
embedded in individual segments as they are molded. All transvers joints are broken; longitudinal
joints being continuous”.
Source: The Parmley System of Arch Construction; Catalogue E (1927). 
10 Handbook of Precast Segmental Tunnel Lining Systems

Figure 1.11 C ross-​s ectional view of the fully assembled parmley precast tunnel system.
Note: “Cup Block” plan of the Parmley System reinforced concrete segmental construction for
conduit sizes 0.760m (30”) to 1.676m (66”) diameter” 
Source: The Parmley System of Arch Construction; Catalogue E (1927).

Table 1.1 Parmley’s patents for precast concrete products for tunnels.

Summary of Parmley’s United States patents for precast concrete products used for
sewer and tunnel construction projects. Many of these were considered as precast
segmental tunnel linings used in shallow trench excavations

Patent document data

Item Patent holder Number Date Patent title
1 Walter C. Parmley 1,774,664 02 Sep 1902 Joint for Conduits
2 Walter C. Parmley 1,601,411 28 Sep 1902 Concrete Block
Construction
3 Walter C. Parmley 712,841 04 Nov 1902 Process of Building
Concrete and Masonry
Structures
4 Walter C. Parmley 709,794 23 Sep 1902 Concrete Arch
Construction
5 Walter C. Parmley 696,838 01 Apr 1902 Concrete Arch
Construction
6 Walter C. Parmley 764 302 05 Jul 1904 Concrete Arch
Construction
7 Walter C. Parmley 764 303 05 Jul 1904 Concrete Arch
Construction
8 Walter C. Parmley 740,039 29 Sep 1909 Concrete Arch
Construction
 Precast segmental tunnel lining history 11

Table 1.1 (Continued)

Patent document data

Item Patent holder Number Date Patent title
9 Walter C. Parmley 1,131,646 09 Mar 1915 Concrete and Steel
Construction
10 Walter C. Parmley 1,133,313 30 Mar 1915 Concrete Pipe
11 Walter C. Parmley 1,258,805 12 Mar 1918 Concrete Block
Construction
12 Walter C. Parmley 1,293,898 11 Feb 1919 Concrete Block
Construction
13 Walter C. Parmley 1,435,174 14 Nov 1922 Concrete Block
Construction
14 Walter C. Parmley 1,781,699 18 Nov 1930 Reinforced Concrete Pipe

Table 1.2 O ’Rourke’s patents for precast concrete products for tunnels.
Summary of O’Rourke’s United States patents for precast concrete products used for
tunnel construction projects. Many of these were considered as precast segmental
tunnel linings used in both shallow and deep tunneling applications

Patent document data

Item Patent holder Number Date Patent title
1 J. F O’Rourke 1043348 05 Nov 1912 Tunnel or Conduit Lining
2 J. F O’Rourke 1277107 27 Aug 1918 Tunnelling (for very large single bores)
3 J. F O’Rourke 1296312 01 Mar 1919 Tunnel Shield
4 J. F O’Rourke 1382668 28 Jun 1921 Means for Supporting Tunnel Lining
Segments
5 J. F O’Rourke 1484188 19 Feb 1924 Method of and Means for Erecting
Tunnels
6 J. F O’Rourke 1788248 06 Jan 1931 Tunnel or Conduit Lining
7 J. F O’Rourke 1235233 31 July 1931 Method of Tunnel Construction (for
backfill grouting)
8 J. F O’Rourke 1889563 29 Nov 1932 Tunnel or Conduit Lining
9 J. F O’Rourke 1976627 09 Oct 1934 Mold for the Manufacture of
Concrete Blocks
10 J. F O’Rourke 1976628 09 Oct 1934 Curved Tunnel Construction and
Method of Producing Same

1.6.1 Mount Royal Tunnel –​ Montreal, Quebec –​1912 to 1916

A twin track railway tunnel was constructed in Montreal from 1912 to 1916 to provide
better access for cross-​country passengers and freight transportation. While the tunnel was
located under Mount Royal in the middle of the city, in competent granitic and limestone
strata, approximately 500 m (1,650 feet) was excavated in mixed-​face conditions. The
12 Handbook of Precast Segmental Tunnel Lining Systems

Figure 1.12 M ount Royal Tunnel –​ mixed-​f ace tunneling using precast segmental tunnel lining.
Note: 1912 isometric view of excavation and lining operations in the Mount Royal Tunnel in
Montreal, Quebec. O’Rourke Interlocking Tunnel Blocks were erected for both parallel tunnel
arches only.
Source: “Tunnelling–​S hort and Long Tunnels of Small and Large Sections Driven Through Hard
and Soft Materials”, by Lauchli, 1915. 

mixed-​face tunnel was located near the inter-​city terminus of the tunnel and proved to be
very challenging. The O’Rourke Interlocking Tunnel Block system was successfully used
in conjunction with a tunnel shield and segment erector device, Figure 1.12 below (from
Lauchli, 1915).
Precast segmental tunnel lining (arch only) were constructed using “Interlocking Concrete
Blocks”, patented by John F. O’Rourke, New York City. Reference US Patent No. 1,043,348
(1912); Interlocking Concrete Blocks”. Details of the precast concrete segmental tunnel
liner used in the mixed face tunnel section included the following:

As-​constructed Mount Royal Tunnel –​precast segmental tunnel liner

• Segment thickness 610 mm (24”) • On-​site precast plant
• Segment length 685 mm (27”) • Cast iron segment molds
• Segment arc length 1.50 m (60”) • Reinforcing (not used)
• Segment radius 3.05 m (10’) • Center cast-​in-​place lifting bolt

Key details of the fixed-​face portion of the Mount Royal Tunnel are shown below in
Figures 1.13 and 1.14.
Excerpts from The Journal for the Engineering Institute of Canada, April 1919, No.4
include the following.
 Precast segmental tunnel lining history 13

Figure 1.13 M ont Royal Tunnel –​ cross-​s ection showing precast segments in arch.
Note: Twin tunnel cross-​s ection in soft ground requiring excavation by use of the shield and
“concrete block” lining.
Source: The Mount Royal Tunnel Report, by Mackenzie, Mann and Company,
1913. 

Figure 1.14 M ont Royal Tunnel –​ tunnel lining demonstration in the precast shop.
Note: Full size shop demonstration of the O’Rourke Concrete Block Arch precast segmental
tunnel lining in Montreal, Quebec in 1912.
Source: The Mount Royal Tunnel Report, by Mackenzie, Mann and Company,
1913. 
14 Handbook of Precast Segmental Tunnel Lining Systems

Tunnel
The concrete blocks [segments] were 610mm (2 feet) thick radially and 810mm (2 feet,
8 inches) long in the direction of the tunnel. The oval tenons 300 x 790mm x 180mm
(12 x 20 x 7 inches) deep were cast on one side of each block. While corresponding
depressions were cast on the opposite side to engage the tenons on the blocks in the adja-
cent ring of the arch. At the side of the tunnel the arch blocks rested on concrete walls
built into the solid rock.
Segment Precast Plant
A special plant was installed between Dorchester and Cathcart Streets for the casting of
the concrete blocks [segments]. The plant consisted of travelling cranes and hand hoists
for handling the forms and blocks, with an elevated runway for the concrete mixing
plant, from which the concrete was fed to the forms by gravity. The forms themselves
were built-​up of flanged cast iron sections mounted on a steel plate. The tenons were
filled to overflowing and smoothed-​off to correct level when the concrete was commen-
cing to set. A tapered steel mandrel through the center of the form held a special 50 mm
(2-​inch) nut in such position that it was cast into the center of the block and could be
reached through the hole left by the mandrel. This nut was used for attaching the block
to the erector on the shield.
Segment Erector
At the rear of the shield [operation in the mixed face portion of the twin tunnels], two
mechanical erectors were provided for placing the arch blocks in position. Each erector
was pivoted under the working gallery and by means of hydraulic rams, racks and pin-
ions, was able to make four distinct motions, namely, rotation about the central pivot,
longitudinal motion along the axis of the pivot, radial motion, and rotation on its own
axis. By this means, the erector could pick up a block from a car in the heading, revolve
it, and swing it until it was opposite its position in the ring, then push it into the line so
that the tenons fitted into the depressions in the blocks already in place.

Segment Sealing
After erection of the blocks, the joints, the bolts holes and also the space over the blocks
left by the skin of the shield were all closed up with a cement gun and grouting machine.

Additional information on this successful early application of precast concrete segmental

tunnel liners was recorded in a publication titled “The Mount Royal Tunnel, Montreal
Quebec, Canada being built by Mackenzie, Mann and Company, Ltd. For The Canadian
Northern Montreal Tunnel and Terminal Co. Limited, December 1913”.

1.6.2 Holland Tunnel –​ New York City –​1919 concept proposal

The “vehicle tunnel” under the Hudson River that connected New York City (Lower
Manhattan) to Jersey City, New Jersey was initially conceived in the early 1900’s, but
was challenged over its size and the available tunneling methods and materials in that era.
Previous tunnels under the Hudson River (modern-​day PATH and Amtrak tunnel), took
decades from start to finish to complete (Burr, 1885 and others).
The Holland Tunnel would be the largest and most ambitious of all tunnels and be located
in both hard rock and soft ground conditions. Indeed, the early proposals for the tunnel
included both twin one-​way tunnels and an innovative single 11-​m (36-​foot) internal diam-
eter tunnel having two levels for bi-​directional vehicle traffic, Figures 1.15 and 1.16.
 Precast segmental tunnel lining history 15

Figure 1.15 H olland Tunnel concept proposal in 1918 using precast segmental tunnel lining.
Note: Proposed 3D concept rendering for a twin level tunnel Holland Tunnel by George Goethals
and John F. O’Rourke in 1919.
Source: The Illustrated London News, Sunday, 05 Apr 1919. 

The larger diameter tunnel was supported with precast concrete segmental tunnel liners,
most probably using the patented O’Rourke Interlocking Tunnel Blocks. (Reference US
Patent No. 1,043,348 (1912). Ultimately, twin parallel tunnels were constructed using gray
cast iron segmental tunnel liners. While this proved successful, O’Rourke argued that the
precast concrete tunnel liners were less expensive and more durable.

Double-​deck Holland Tunnel –​precast segmental tunnel liner (proposed)

• Segment thickness 900 mm (36”) • On-​site precast plant
• Segment length 900 mm (36”) (assumed) • Cast iron segment molds
• Segment arc length 1.50 m (60”) (assumed) • Reinforcing (undisclosed)
• Segment radius 5.49 m (18’–​0”) • Center cast-​in-​place lifting bolt

Ultimately, the Holland Tunnel was constructed as a twin tunnel facility, completed
in 1927. It is still in service today. The twin tunnels were constructed 9.0-​m (29.5-​foot)
16 Handbook of Precast Segmental Tunnel Lining Systems

Figure 1.16 H olland Tunnel concept proposal in 1919 using precast segmental tunnel lining, cross
section.
Note: A 12.8-​m (42-​foot) outer diameter twin level tunnel was proposed by George Goethals
and John F. O’Rourke in 1919 for the Holland Tunnel. Three-​foot thick precast “interlocking
tunnel blocks” were proposed as the initial tunnel lining, for ground support and groundwater
control.
Source: The Sun (newspaper), Sunday, 26 Jan 1919. 

diameter using four separate soft ground compressed air shields that concurrently installed
a bolted and sealed cast iron segmental tunnel liner (comprised of 14 curved segments that
are each 1.83 m (6-​feet) long as the excavation advanced.
The twin tunnel initial concept arrangement in comparison to the single-​bore tunnel con-
cept is illustrated in Figure 1.17. The north tube is 2,608-​m (8,558-​feet) long and the south
tube 2,551-​m (8,371-​feet) long. The roadway is 6.1-​m (20-​feet) wide and reaches a max-
imum depth below mean high water of 28.5 m (93.4 feet).
Excerpts from The Sun (newspaper), Sunday, 26 Jan 1919:

“….The part of the tunnel extending under the river, on inshore as far as the subsurface
and street surface conditions permit, will be 42 feet in diameter at the outer rim and built
of concrete blocks three feet in thickness. Steel has been rejected, the tendency being
nowadays in such work to be back to concrete and stone. Construction will be by the
shield method….”
“….Interlocking Concrete Blocks–​as the basis of construction will be thoroughly
waterproofed at the outer side of the joints by asphalt, while the face of the joints on the
inner side of the tunnel will be treated with tarred oakum and cement mortar. The whole
 Precast segmental tunnel lining history 17

Figure 1.17 H olland Tunnel concept proposal –​ 1918 using precast segmental tunnel lining.
Note: A 12.8-​m (42-​foot) outer diameter twin level tunnel was proposed by George Goethals
and John F. O’Rourke in 1918 for the Holland Tunnel in comparison to a twin tube concept using
traditional cast iron segments. The single-​b ore tunnel used three-​f oot thick precast “interlocking
tunnel blocks” as the initial tunnel lining.
Source: The Sun (newspaper), Sunday, 26 Jan 1919. 

interior filled with mortar introduced under compression and then coated with white
cement….”

1.7 MID-​C ENTURY ASSESSMENTS AND TUNNEL PLANNING REPORTS

In the mid-​1960’s and with the increasing demand for more mass transit systems, additional
attention was paid to research and development to tunneling systems and contracting prac-
tices by funding agencies in many countries. The following is a brief summary of some
focused research activities and reports related to the use of precast segmental tunnel linings
for transportation tunnels as an alternative to other “traditional” materials and tunneling
methods.

1.7.1 Tunnel lining research activities

During the 1960’s and 70’s in the United States, Japan, Canada and many European coun-
tries including the United Kingdom, numerous university and tunnel industry studies were
conducted to focus on means and methods, machinery and materials needed for improving
tunnel design and construction. Many of the studies (and tunnels) were for high-​speed
ground transposition and in particular, urban transit tunnels.
Agencies such as the United States Department of Transportation and Federal Railway
Administration were eager to advance the state-​of-​the-​art and the United States tunneling
industry to accommodate proposed ambitious transportation schemes in major cities and
in select areas for long distance, i.e. high-​speed rail systems. This approach involved a
18 Handbook of Precast Segmental Tunnel Lining Systems

composite solution (systems) involving highly mechanized tunneling and ground support
systems and innovative materials. A partial summary listing of reports prepared for the
United States Department of Transportation (DOT) and Federal Railway Administration
(FRA) included the following, for example.

• 1968 Tunneling –​State of the Art (USA)

• 1969 Design of Tunnel Liners and Support Systems (USA)
Some Design Considerations in Selection of Underground Support
• 1969 
Systems (USA)
Concrete for Tunnel Liners: Behavior of Steel Fiber Reinforced Concrete
• 1974 
Under Combined Loads –​Final Report (USA)
• 1975 Concrete for Tunnel Liners: Structural Testing of Segmented Liners (USA)
Concrete for Tunnel Liners: Mix Design Recommendations or Prototype
• 1975 
Extruded Liner System (USA)
• 1976 Tunneling: The State of the Industry (USA)
Tunnelling Technology –​An Appraisal of the State of the Art for Application
• 1976 
to Transit Systems (Canada)
Tunnel Construction –​State of the Art and Research Needs –​Special
• 1977 
Report (USA)
• 1978 A Review of Tunnel Lining Practice in the United Kingdom (UK)
• 1980 Segmented Concrete Tunnel Liner and Sealant Systems (USA)

Many of the above-​noted reports were developed with the intent to advance the state of
the domestic and international tunnelling industry, materials and engineering. Many other
reports and technical publications are listed in the Bibliography and Additional Reading
Materials.

1.8 FABRICATED STEEL TUNNEL LININGS

There was a period several decades ago when, for economic and technical reasons, fabri-
cated steel tunnel linings were popular. In the United Stare for example, precast segmental
tunnel lining and manufacturing technologies had not gained popular acceptance, mean-
while fabricated steel tunnel lining segments were considered as an economical solution and
compatible with the anticipated subsurface conditions. The use of fabricated steel tunnel
linings was for a short period of time, more economical than cast iron and could be manu-
factured with more flexibility and more locations without the heavy capital investment
needed for efficient casting operations.
Some of the more notable mid-​twentieth century and modern-​era tunnels utilizing fabri-
cated steel segmental tunnel liners included those listed in Table 1.3.

1.8.1 Bay Area Rapid Transit System –​San Francisco, California

While much of the San Francisco Bay Area Rapid Transit (BART) system tunnels were
lined with single-​pass precast concrete segmental tunnel liners, a significant portion was
not. Instead, specific tunnels were lined with fabricated steel segmental tunnel liners as
illustrated above in Figures 1.18–​1.20. There were many reasons for this choice, most not-
ably cost and local preferences (and industrial capacity) for fabrication of welded steel
plate tunnel liners. Other tunnels built in the same area in California, for example, also
 Precast segmental tunnel lining history 19

Table 1.3 Fabricated steel segmental tunnel linings –​ project summary.

Summary of completed tunnels in rock and soil conditions wherein fabricated steel
segmental tunnel linings were selected as the final lining. Subsurface conditions as
well as other decision factors are noted

Dimensions Ground conditions Other issues

Project
Item Project name location Dia. Length Rock Soil Water Gas Seismic CP/​S ta
1 BART System a USA 6.70m Various     
San Francisco Metro (22’)
2 MARTA System a USA 6.70m Various   
Atlanta, GA Metro (22’)
3 LA Metro USA 6.40m Various     
Tunnels b Metro (21’)
Los Angeles, CA
4 Hiawatha USA 6.40 m 4.570 m   
Tunnels c Metro (21’) (15,000
Minneapolis, MN LF)
5 WMATA USA 6.40 m Various     
Tunnels d Metro (21’)
Washington, DC
Notes:
a
Design life cycle (+​/ ​- ​1 00 years), subsurface conditions and estimated capital cost of alternate tunnel linings
including cast iron, precast segmental and two-​p ass tunnel lining systems were considered with fabricated
steel being preferred.
b
Fabricated steel segmental, bolted gasketed tunnel lining selected to address seismic conditions and poten-
tial differential movement at the bored running tunnel-​t o-​s tation interface only.
c
Fabricated steel, bolted gasketed segmental tunnel lining selected to provide an efficient break-​o ut panels)
at the designated locations for cross-​p assages from the precast segmentally lined bored running tunnels.
d
Gray cast iron was originally specified but became in short (and erratic) supply. Alternative fabricated steel
(Figure 1.20) was considered. Precast segmental tunnel linings were ultimately adopted for the remaining
underground portion. It should also be noted that large, curved architectural precast panels were used in
station caverns initially excavated by drill and blast methods.

Table Legend:
• Primary condition or majority of the work
 Secondary (infrequent) condition or occasional encounter to the work
CP/​S ta Cross-​p assage break-​o ut location. Station-​t o-​t unnel lining transition due to seismic design criteria.

used fabricated steel plate tunnel liners. It was only in subsequent BART tunnels, that pre-
cast concrete segmental tunnel liners became the preferred material choice as well as the
most economical solution.

1.8.2 Washington Metropolitan Area Transit Authority (WMATA)

The initial tunneled portion of the WMATA system utilized cast iron segments for support.
In the early 1980’s, this material came into short supply. While cast iron was considered
as a single-​pass tunnel lining system, alternative materials such as fabricated steel as illus-
trated in Figure 1.21 and precast concrete as illustrated in Figure 1.24 were evaluated for
economics and supply chain basis. Other tunnels built in the same area and time frame (i.e.
Baltimore Rapid Transit System), for example, also used precast concrete segmental tunnel
liners due to the shortage of alternative materials.
20 Handbook of Precast Segmental Tunnel Lining Systems

Figure 1.18 P refabricated steel segmental tunnel lining –​ BART system. Example 1.
Note: Transit tunnel construction using fabricated steel segmental tunnel liners.
Source: BART Archives. 

By the mid-​1980’s and continuing to the present, all WMATA tunnels were designed for
precast concrete segmental tunnel linings, with tunnel excavation equipment adjusted to
suit. Fabricated steel tunnel linings were not used. Precast concrete segmental tunnel liners
become the preferred material choice as well as the most economical solution. There were
many reasons for this choice, most notably cost and local preferences (and industrial cap-
acity) for fabrication of precast concrete.

1.9 MODERN ERA PRECAST SEGMENTAL TUNNEL LINER

DEVELOPMENTS
There are only a very few applications of precast concrete segmental tunnel liner installa-
tions before the 1960’s apart from some metro tunnels in London and incidental sewer tun-
nels. Progressively, sewer and water tunnels adopted the use of precast concrete segmental
tunnel linings, but this was done in conjunction with improvements to both manual and
mechanized tunnel excavation operations, segment material supply (availability) and eco-
nomic factors.

1.9.1 Kinnear Moodie & Company (Scotland and Great Britain)

Excerpts from Kinnear Moodie & Company; Civil Engineering, Contractors and
Manufacturers of Precast Concrete lining for Shafts and Tunnels, Corporate Brochure, 1956:
 Precast segmental tunnel lining history 21

Figure 1.19 P refabricated steel segmental tunnel lining –​ BART system. Example 2.
Note: Transit tunnel construction using fabricated steel segmental tunnel liners.
Source: BART Archives. 

Kinnear, Moodie & Co., Ltd. Made the first precast reinforced concrete bolted segmental
lining for shafts and tunnels in 1936. At that time, the Company was asked
by the Consulting Engineers on the then London Passenger Transport Boars, Messrs.
Mott, Hay, & Anderson, to carry out experiments in the manufacture and use of precast
reinforced concrete bolted segments, with a view to meeting the need, which began to
develop at that time, for economizing on cast iron. Before this, various form for concrete
block lining had been devised but none with outstanding success. The underlying prin-
ciple of Messrs. Mott, Hay, & Anderson’s design was to retain the bolted segmental ring,
and to replace cast iron by reinforced concrete in the manufacture.
The distribution and type of loading likely to be applied to tunnel segments do not
lend themselves to laboratory tests, so that in order to settle the structural merits of these
segments, the Engineers carried-​out full-​scale tests on specially constructed experimental
lengths of tunnel. The results of these tests have been published and have established def-
initely that concrete segments compare favorably with cast iron in strength and stiffness,
whilst their cost is substantially lower.
The making of precast concrete segments at the Company’s works was controlled
at every stage, strict attention being paid to all details of the manufacturing processes
involved. Clean aggregates of proper mechanical analysis are used, and the mixing is
done by gravimetric methods, the material being tested regularly as a matter of routine
and any necessary alteration made forthwith. Strict attention is paid to the accuracy of the
22 Handbook of Precast Segmental Tunnel Lining Systems

Figure 1.20 P refabricated steel segmental tunnel lining –​ BART system tunnels.
Note: San Francisco Bay Area Rapid Transit (BART) system tunnels were constructed using fab-
ricated steel segment tunnel liners.
Source: BART Archives. 

molds and all segments are properly consolidated on heavy vibrating table. The method
of manufacture ensures that there can be no difference of dimensions of practical import-
ance between different units of the same pattern, all segments being interchangeable.

1.9.2 C. V. Buchan (Great Britain)

C. V. Buchan Concrete was formed in 1966 to manufacture precast concrete tunnel lin-
ings. In 1996, C. V. Buchan of Cheshire, England was granted United States Patent No.
5,549,416 for joining two tunnel lining rings together known as Buc-​Lock.

1.9.3 Charcon Tunnels (Great Britain)

Charcon Tunnels was formed in 1974 following the purchase of Kinnear Moodie Concrete
who had been manufacturing and developing precast concrete tunnel linings since 1936.
Subsequent to its formation, Charcon became involved in manufacturing tunnel liners in
France, Egypt, Australia, Thailand, Hong Kong, Canada and the USA. Figures 1.22 and
1.23 illustrate some of the early technical developments by Charcon Tunnels.
 Precast segmental tunnel lining history 23

Figure 1.21 P refabricated steel segmental tunnel lining –​ WMATA contract design.
Note: 1980-​e ra Contract Drawing illustrating the Owner’s design of fabricated steel segmental
bolted and gasketed tunnel lining for a metro tunnel in Washington, DC, United States.
Source: WMATA Archives. 

Figure 1.22 C harcon patent for segment joint detail.

Note: Segment joint detail enclosed in a US Patent document.
Source: US Patent No 4,969,769 dated 13 Nov 1990. 
24 Handbook of Precast Segmental Tunnel Lining Systems

Figure 1.23 C harcon patent for spiral tunnel lining design.

Note: Spiral tunnel lining design enclosed in a US Patent document.
Source: US Patent No 3,969,906 dated 20 Jul 1976. 

Figure 1.24 P recast concrete segmental tunnel lining –​ WMATA contract design.
Note: 1980-​e ra Contract Drawing illustrating the Owner’s design of precast segmental bolted
and gasketed tunnel lining for a metro tunnel in Washington, DC, United States.
Source: WMATA Archives. 
 Precast segmental tunnel lining history 25

1.9.4 Lovat Tunnelling Equipment and segment erector (Canada)

In 1978 Lovat Tunnel patented a unique segment erector device (aka, the “one-​armed
bandit”) for handling precast segment tunnel liner elements. Refer to Figures 1.25 and
1.26, extracted from US Patent No. 4,067,201 dated January 1978. A new Lovat Tunnel
Boring Machine equipped with this segment erector was constructed in 1976 and used
on the City of Thunder Bay, Ontario (Canada) Neebing-​McIntyre Sanitary Trunk Sewer
Tunnel in 1977. This project consisted of approximately 3,170 m (10,400 LF) of 2.16 m
(85”) internal diameter tunnel and located partially in soft clay at a depth of 12 m (40 feet).
Another portion of the tunnel was constructed in water-​bearing silts and sands at shallower
depth. This was the first tunnel in North America constructed using precast concrete seg-
mental tunnel lining as the combination primary and secondary lining.
A description of the segmental tunnel liner as well as the tunnelling equipment were pub-
lished by Morton, Dunbar and Palmer, 1978. A few excerpts are recorded below.

Tunnelling is an ancient art which, until recent times, has used a combination of manual
labor for excavation and readily available raw materials (usually timber) for support
as required. Although tunnelling is basically an art, scientific influence is becoming a
common place due to the combination of improved technology, increasingly difficult
ground conditions, an increased cost of labor, and other economic pressures. Such factors
have provided impetus to the development and use of Tunnel Boring Machines instead
of manual excavation, and a concomitant reassessment or development of compatible
tunnel lining systems. One such novel tunnelling procedure which has been developed
in Canada employs full-​face Tunnel Boring Machine together with an unbolted precast
segmented concrete tunnel lining. The initial contract use of the system was in a soft to
firm clay deposit of lacustro-​deltaic origin at Thunder Bay, Ontario; however, the system

Figure 1.25 S ection through a TBM adapted to handle and erect precast segmental tunnel linings.
Note: Excerpt from a 1978 US Patent drawing for a Lovat TBM adapted to transport and erect
precast segments as tunnel liners in the first ever application in Canada.
Source: US Patent No.4,067,201 dated January 1978. 
26 Handbook of Precast Segmental Tunnel Lining Systems

Figure 1.26 T BM adapted to handle and erect precast segmental tunnel linings.
Note: Excerpt from a 1978 US Patent drawing for a Lovat TBM adapted to transport and erect
precast segments as tunnel liners in the first ever application in Canada.
Source: US Patent No.4,067,201 dated January 1978. 

is highly adaptable to other ground conditions and has been specified as a design alterna-
tive for tunnels in more competent soils and sedimentary rock.

Tunnel Boring Machine

The Tunnel Boring Machine manufacturing company undertook the task of designing
and incorporating into the new machine, a grouting system as well as a means of hand-
ling and erecting the heavy unbolted precast lining segments. Also, the gaskets and pro-
tective skirt had to be developed to prevent soil and/​or grout from squeezing back into
the machine between the lining and the tailpiece as the machine advanced.
The ease with which each segmental ring was erected was a key element in the suc-
cess of the project. The erector arm permitted rotation of each segment through 360° in
a vertical plane as well as complete rotation horizontally. The tool to grip each segment
was designed to utilize two of the recesses cast in the segment intrados for mechanical
securing clamps.

Tunnel Lining
Since the TBM would advance by thrusting on the completed tunnel lining, each thrust
had to be as long as possible, and the erection of each ring had to proceed rapidly within
 Precast segmental tunnel lining history 27

the protection of the tail shield. To achieve this, unbolted segments were preferred. Each
assembly ring would come in contact with the ground only as is emerged from the rear
of the tail shield during the machine advance. As the lining was to serve both as a pri-
mary and secondary lining, it had to be designed to accommodate all anticipated ground
loads, as well as erection and handling stresses and the longitudinal stresses developed
as a consequence of the TBM thrust. The design would rely on the segmental ring hav-
ing sufficient flexibility to deform to the degree needed to reduce the bending stresses
to acceptable limits. Finally, all joints have to be adequately sealed to prevent inflow of
grout or remolded soil and water when the segmental lining was thrust out of the pro-
tection of the tail shield.

1.9.5 Development and use of “junk segments” for tunnels

(United States)
The so-​called “junk segments” were initially developed and used in tunnels (soft ground
and hard rock) where a final secondary tunnel liner was specified in the final design. “Junk
segments” were used in conjunction with a TBM (for rapid tunnel excavation) for imme-
diate ground support and preferred over other materials in suitable ground conditions.
They provided little resistance to groundwater inflows since they were not supplied with
gaskets. A typical “junk segmental” ring was comprised of the following features and
appurtenances.

• Four-​piece precast concrete ring

• Keyway wet cast (or shotcreted) in place
• Backfill grouting (with or without pea gravel) through the segments
• Segment length consistent with the stroke length of the TBM
• No gaskets installed
• No connection bolts installed
• Modest ring-​to-​ring alignment devices used

Figure 1.27 shows a representative image of “junk segments” installed in soft ground and
hard rock tunnels as the “primary lining”.

1.9.6 Development and use of precast segments for shaft linings

Precast segmental shaft linings were initially developed and used in shafts having satisfac-
tory ground stability and little water inflows. They could be considered as one-​pass system
or as the primary lining in a two-​pass shaft lining design.
Precast segments used in suitable soft ground shaft excavations were generally han-
dled and placed using the available excavation equipment (i.e. excavator and cranes) for
rapid excavation cycles and immediate ground support and preferred over other materials.
Completed rings would be grouted in-​place and waterproofed with gaskets, they provided
substantial resistance to groundwater inflows. A shaft lining segmental ring was comprised
of the following features and appurtenances.
A notable feature of the this approach, starting with design, manufacture, specification
and use of precast segment shaft linings, was to standardize the dimensions in the same
manner used for competing pressed steel liner plates. This resulted in considerable reuse of
casting molds and a reliable supply chain consisting of normally stocked “standardized and
dimensioned” products.
28 Handbook of Precast Segmental Tunnel Lining Systems

Figure 1.27 N arragansett Bay CSO –​ main spine tunnel.

Note: Interior of the Narragansett Bay CSO tunnel in the segment build area of the 9.1m
(30-​foot) diameter TBM where “junk segments” were used as the primary tunnel support.
Source: Brian Fulcher, personal photo collection. 

1.10 PRECAST CONCRETE TUNNEL LINERS USED IN BORED HARD

ROCK TUNNELS
In recent years, several notable hard rock tunnels have been excavated and lined with
single-​pass precast concrete segmental linings. The reasons have included water control and
ground support, but in at least one instance, the estimated cost of a secondary cast-​in-​place
reinforced concrete liner was a deciding factor. The following tunnels as listed in Table 1.4
received precast segmental tunnel linings as the final tunnel lining, some without a per-
manent waterproof sealing system. Figures 1.28 and 1.29 illustrate the Buckskin Mountains
Tunnel in Arizona, United States.
As noted in Table 1.4, the rationale for the use of precast segmental tunnel lining was
highly variable and adapted to the anticipated ground conditions. In at least one case, Deer
Island Effluent Outfall Tunnel Project (Boston, MA) shown in Figure 1.30, the ground con­
ditions were largely unknown; therefore, the selection of precast segmental tunnel lining
was also considered as a risk mitigation measure as well as for groundwater inflow control
needed to fulfill construction permit requirements. The two Arrowhead Tunnels located in
San Bernardino, California utilized a very robust precast segmental tunnel lining for both
ground, water and hazardous gas control, in what would generally be characterized as
highly variable tunneling conditions (Figure 1.41).

1.11 TRANSITION TO PRECAST CONCRETE SEGMENTAL TUNNEL AND

SHAFT LININGS
Several related events occurred in the early 1970’s (in the United States) that changed
preferences for initial tunnel lining materials in soft ground conditions (and some hard
 newgenrtpdf
Table 1.4 B ored hard rock tunnels with precast segmental tunnel linings installed. Summary of recently completed hard rock tunnels where
precast segmental tunnel lining was selected as the final lining. General subsurface conditions as well as other decision factors
are noted

Dimensions Ground conditions Other issues

Item Project name Project location Outside diameter Overall length Rock Soil Water Gas Risks Costs
1 Park River Tunnel USA 6.70 m 2,775 m   
Hartford, CT 1982 (22’–​0 ”) (9,100 LF)
2 Buckskin Mtn Tunnel USA 7.2 m 10,900 m   
Centra Arizona 1985 (23’–​6 ”) (35,770 LF)
3 San Antonio River Tnl USA 7.40 m (4,940 m    
El Paso, TX 1995 (24’–​4 ’) (16,200 LF)
4 Deer Island Outfall a USA 8.8 m 14,935 m    
Boston, MA 2002 (29’–​0 ”) (49,000 LF)
5 DRO Tunnels b USA 6.40 m 1,900 m    
Detroit, MI 2005 (21’–​0 ”) (6,235 LF)
6 Narragansett CSO USA 9.1 m 4.8 km   
Providence, RI 2006 (30’–​0 ”) (16,000LF)

Precast segmental tunnel lining history

7 Guadarrama Tunnels, Spain 9.5 m 28.4 km   
Spain 2007 (31’–​0 ”) (17.3 m)
8 Arrowhead Tunnels c USA 5.8 m 12,800 m    
San Bernardino, CA 2009 (19’–​1 ”) (42,000 LF
9 Lake Mead Intake 3 USA 7.9 m 4.6 km    
Henderson, NV 2017 (26’–​0 ”) 15,000 LF
10 No.7 Line Subwayd USA 6.2 m 2,500 m    
New York City 2010 (21’–​0 ”) (7,500 LF)
11 OARS Tunnel e USA 7.0 m 7,010 m    
Columbus, OH 2017 (23’–​0 ”) (23,000 LF)
(Continued)

29
 newgenrtpdf
Table 1.4 (Continued)

30
Handbook of Precast Segmental Tunnel Lining Systems
Dimensions Ground conditions Other issues
Item Project name Project location Outside diameter Overall length Rock Soil Water Gas Risks Costs
12 South Hartford Tnl e
USA 5.5 m 4,600 m  
Hartford, CT 2021 (18’) (15,000 LF)
13 Three Rivers CSO 5 USA 5.5 m 7,475 m    
Ft Wayne, IN 2021 (18’) (24,520 LF)
14 Kemano Power Tunnel CAN 5.72 m 7,606 m   
Kitimat, BC 2021 (18’–​9 ”) (24,954 LF)
Notes:
a
PSTL (ungasketed) selected to control excess groundwater infiltration and to support potentially variable ground conditions. Additional benefit was to provide
a predictable and beneficial hydraulic surface for effluent flows.
b
PSTL selected to satisfy Tunnel Health and Safety requirements, control excess groundwater and dissolved H 2S gas infiltration.
c
PSTL selected to satisfy construction permit requirements, control excess groundwater infiltration, stabilize highly variable ground conditions and potential
gas pockets.
d
Design life cycle (100 years) and estimated capital costs of the CIP concrete liner was more than the single-​p ass precast segmental tunnel liner; with predict-
able waterproof performance and maintenance costs.
e
PSTL selected to cope with excess groundwater infiltration and planned water control measures

Table Legend:
• Primary condition or majority of the work
 
Secondary (infrequent) condition or occasional encounter to the work
 Precast segmental tunnel lining history 31

Figure 1.28 B uckskin Mountains Tunnel –​ contractor’s precast segmental tunnel lining approach.
Note: Contractor’s drawings for the first ever precast segmental lining used in the bored hard
rock tunnel in the United States Groseclose and Schoeman).
Source: RETC 1976, Chapter 26, Society and Mining Engineers. 

Figure 1.29 B uckskin Mountain Tunnel –​ as completed in 1985.

Note: Completion of the 7.2m (23’-​6 ”) diameter Buckskin Mountains rock tunnel using precast
segmental tunnel liner.
Source: ASCE and USBR–​C entral Arizona Project. 
32 Handbook of Precast Segmental Tunnel Lining Systems

Figure 1.30 D eer Island Effluent Outfall Tunnel –​ under construction in 1995.
Note: Interior of the Deer Island Effluent Outfall Tunnel where 8.1m (26’-​6 ”) diameter precast
segmental tunnel liner was used in an argillite formation.
Source: Brian Fulcher, personal photo collection. 

rock tunnels) and use of mechanized excavation equipment. Research by the United States
Department of Transportation (US DOT) and acute steel shortage precipitated the evalu-
ation of alternate ground support materials and especially for the growing quantity of
transportation tunnels anticipated in urban areas. Whereas fabricated steel tunnel lining
segments had been developed and used in several transportation projects (i.e. BART transit
tunnels in San Francisco), precast concrete segments were initially used in the Baltimore
Metro tunnels in the mid-​1980’s.

1.11.1 Sehulster Tunnels –​ precast segment developments for soft

ground tunneling
Sehulster Tunnels was started in 1980 by Joseph P. Sehulster of Milwaukee, Wisconsin,
United States. Joseph was a visionary who at the time had over 25 years of experience in
the design and supply of precast concrete pipe in the United States Upper Midwest heavy
construction market. He saw the evolutionary need for precast tunnel linings that could
be installed in soft ground conditions where the tunnel exceeded 3.0-​m (10-​foot) diameter
that precluded the use of jacking pipe that had become popular at that time in his market.
Precast tunnel linings were more adaptable to the new developments in shields and TBMs,
Richard Lovat was also then introducing his new EPB (Earth pressure balance) TBMs that
increased the speed and safety in which tunnels could be driven.
In 1980 Sehulster Tunnels received its first contract for the supply of precast tunnel lin-
ings from the Milwaukee Metropolitan Sewage District (MMSD) for 3,353 m (11,000 feet)
of 2.30-​m (90-​inch) internal diameter tunnel lining. As noted earlier in this chapter, precast
 Precast segmental tunnel lining history 33

Figure 1.31 S ehulster Tunnels –​ segment reinforcing.

Note: Shop assembly of bar reinforcement for small diameter precast segmental tunnel liners.
Source: Sehulster Tunnels, Inc. 

tunnel linings were used previously in North America and other parts of the world, but this
was their first application in the United States upper Midwest and would go on to become
a very integral part of the extensive Great Lakes Clean-​Up Program. Sehulster brought
together with his own engineers, local engineering firms; i.e. Freeze and Nichols as well
as international experts from Charcon Tunnels Ltd. in the United Kingdom to develop a
precast segmental tunnel lining design that was submitted and approved by the MMSD
(Figures 1.31 and 1.32).
Note the segments resemble cast iron tubbings that were the international standard at
the time with pockets cast in the lining to facilitate weight savings and ease of bolting with
straight bolts as these segment types were typically erected manually in the tunnel in the
United Kingdom. After construction of the precast tunnel lining, a welded steel pipe was
installed and the annular space between the steel pipe and the precast lining was grouted.
A lining of this design was then manufactured and supplied to the Northeast Ohio Regional
34 Handbook of Precast Segmental Tunnel Lining Systems

Figure 1.32 S ehulster Tunnels –​ segment casting.

Note: Static mold casting of small diameter precast segmental tunnel liners by Sehulster
Tunnel crews.
Source: Sehulster Tunnels, Inc. 

Sewer District (NEORSD) in Cleveland, OH for 5,000 feet (1,524 m) of 10.3-​foot (3.14-​m)
internal diameter from 1985 to 1986.
As Sehulster Tunnels was gaining experience in precast segment manufacturing the team
were working on a new innovation in conjunction with the MMSD that is now known as
“one-​pass precast tunnel linings”; meaning the internal face of the precast segments were
smooth that facilitated the removal of a secondary carrier pipe of steel or concrete thus
saving the MMSD time from the construction schedule and an overall lower project cost,
Figures 1.33 and 1.34.
To achieve this concept, Sehulster developed, tested and implemented proprietary innova-
tions in precast segment design that were later used on other tunnel projects throughout
the world by Charcon Tunnels Ltd. In order to facilitate a true “one-​pass” segmental lining,
the internal face of the segments had to be completely smooth without pockets as the
 Precast segmental tunnel lining history 35

Figure 1.33 One-​p ass tunnel lining mock-​u p assembly.

Note: Full-​s cale assembly of a 5+​1 precast one-​p ass tunnel lining to illustrate proof-​o f-​c oncept.
Source: Sehulster Tunnels, Inc. 

cast-​in pockets would create turbulence and decrease the hydraulic efficiency of the lining.
To obtain the “segmental lining without pockets”, a new innovation was created for con-
necting the circumferential joints of the segmental ring while at the same time allowing for
segment joint stagger that is important for ring stability. This new innovation was called the
“Fast Lock Dowel and Socket Assembly”, Figure 1.35. This connection device is covered
under US Patent 4,477,204 dated 16 Oct 1984.
The “Fast Lock Dowel System” aligned the adjoining segments and associated gaskets
while the special internal locking system prevented the rings from separating during ring
installation, with the steel center axis providing substantial shear resistance from external
forces acting on the lining. Past connection systems were dependent upon manual installa-
tion and tightening bolts, but the real innovation of this system was that it used the massive
power of the TBM thrust jacks to set the system into place thus minimizing human involve-
ment and effort, resulting in improved ring build quality and reduced installation time.
This, combined with the Charcon Tunnels “Camlock Radial Joint Connection System”,
dramatically changed how tunnels were built in the Milwaukee area and other parts of
North America.
36 Handbook of Precast Segmental Tunnel Lining Systems

Figure 1.34 Fast lock dowel and socket assembly.

Note: Full-​s cale assembly of a 5+​1 precast one-​p ass tunnel lining using the Sehulster “Fast Lock
Dowel and Socket Assembly”.
Source: Sehulster Tunnels, Inc. 

Sehulster’s one-​pass tunnel lining innovations were used on two sewer tunnel projects in
Milwaukee, Wisonsin and later incorporated into a large storm water diversion tunnel in
San Antonio, Texas where Sehulster Tunnels successfully built 6,706 m (22,000 feet) of
7.4-​m (24.3-​foot) internal diameter one-​pass precast segmental tunnel lining. This precast
segment manufacturing technology was then used to manufacture the tunnel lining for three
world class underwater tunnel projects at the time.

• MWRA Boston Outfall Tunnel in Boston, MA, United States (1991 to 1994)
• 13,715-​m (45,000-​feet) long
• 7.4-​m (24.3-​foot) internal diameter tunnel lining
 newgenrtpdf
Precast segmental tunnel lining history
Figure 1.35 P recast segmental tunnel lining –​ fast lock dowel and socket assembly drawing.
Note: Patent application drawing for the Sehulster “Fast Lock Dowel and Socket Assembly” to illustrate a section through the installation and
all component parts.
Source: Sehulster Tunnels, Inc. 

37
38 Handbook of Precast Segmental Tunnel Lining Systems

• St. Clair River Railroad Tunnel in Sarnia, Ontario, Canada (1992 to 1994)
• 1,854-​m (6,083-​feet) long
• 8.4-​m (27.5-​foot) internal diameter tunnel lining
• San Diego Southbay Ocean Outfall in San Diego, CA, United States (1995 to 1998)
• 6,096 m (20,000 feet)
• 2.74-​m (9.0-​foot) internal diameter tunnel lining
Sehulster and his team of engineers were tunnel industry leaders and innovators who built
11 major tunnel projects from Boston to Seattle while always striving to improve the quality
of the product being delivered. Sehulster promoted the beneficial uses of one-​pass precast
segmental tunnel linings to owners, design engineers and tunnel contractors. The work
Sehulster did to promote the use of one-​pass precast segment tunnel linings was instru-
mental in one-​pass precast segmental tunnel linings having the industry acceptance they
enjoy today.

1.12 TECHNICAL DEVELOPMENTS IN THE MODERN ERA

1.12.1 Development and use of concrete mix designs

Continuous development of improved concrete mix designs (i.e. strength, durability, and
porosity, etc) and other properties, have greatly benefited technical development and
improved the quality of precast segmental tunnel linings for major infrastructure projects.
Whereas 20 to 27.5 MPa (3,000 to 4,000 psi) concrete was commonplace in the 1930’s with
initial development of segment tunnel lining element, it is now not unusual to see 40 to 55
MPA (6,000 to 8,000 psi) Steel Fiber Reinforced Concrete (SFRC) widely used.
Additionally, concrete with increased durability for longer service lives are well estab-
lished. Issues with curing, handling and storage have been overcome resulting in precast
products with over 100 year service life expectancy. This is matched to modern sealing sys-
tems and connection devices needed for rapid and secure installation of precast segmental
tunnel linings.
A more thorough discussion on the development and use of reliable concrete mix designs
is enclosed in Chapter 4.

1.12.2 Development and use of reinforcing materials

Early precast segmental tunnel innings showed elementary steel reinforcing as noted in
the Parmley and O’Rourke patented designs. More recent developments for the use of
bar reinforcement have included increasing yield strengths of steels (and GFRP), weld-
able steel (for cages) and automated methods for cutting, bending, welding and assembly.
Additionally, reinforcement (using bar) has made use of coiled supplies in lieu of straight
bundled bars for greater handling and fabrication efficiencies and safety considerations
(Figure 1.36).
In recent decades, use for steel fibers solely and/​or in combination with bar reinforce-
ment has been developed and widely accepted. Indeed, there are considerable advantages
with handling and erection methodologies adapted to suit reinforcement distribution in
individual segments. Design codes have been introduced (e.g. ACI 544.7-​16) to reflect and
endorse the benefits of the use of steel fiber reinforcing for precast segmental tunnel lining
elements.
 Precast segmental tunnel lining history 39

Figure 1.36 Traditional bar reinforcing cage for precast segment.

Note: Traditions deformed reinforcing steel bar cage for PSTL segments, with allowance for con-
nection accessories.
Source: David Klug, personal drawing collection. 

Innovative reinforcing solutions have also included the use of polypropylene fibers, par-
ticularly for transportation tunnels where fire (and severely elevated temperatures) pose
threats to the structural integrity of the segmental tunnel lining.
A more thorough discussion on the application and use of both steel and plastic fibers
reinforcing is enclosed in Chapter 5 .

1.12.3 Segment molds and production systems for casting operations

The segment casting and handling process has also evolved substantially in much the same
manner as segment materials, reinforcing and connection devices. Based on reports from
the early concrete segment manufacturers, the following segment mold materials listed in
Table 1.5 have been used with success, but as the demands of the tunneling equipment
increased, so did the requirements for more and more precision and repeatability from the
segment forming equipment for the precast segments.
O’Rourke’s Interlocking Concrete Blocks were used for the Mount Royal Tunnel, in
Montreal, Quebec, Canada in 1912. A contemporaneous project description of these pre-
cast concrete elements comes from the Railway Age Gazette (November 1915);

The tunnel sections consist of two separate tubes with a semi-​circular top made of con-
crete arch voussoirs cast separately, set into position and supported on either side by
concrete walls ad in the middle by a dividing wall made up of 10.5 inch, 65 lbs/​foot
Bethlehem H-​columns space 2’-​3” center to center and spanned by structural steel built-​
up lintels, all encased in concrete. The [interlocking concrete] blocks are 2 feet thick
radially and 27 inches in the direction of the tunnel axis and were made in length of 5
 newgenrtpdf
40
Handbook of Precast Segmental Tunnel Lining Systems
Table 1.5 M olds and production systems for segment casting operations
Summary of mold types (and materials used) and production casting systems for recently completed tunnels where precast
segmental tunnel lining was selected as the final lining. Note the mold materials changes

Dimensions Precast segment molds Other items

Item Project name Project location Dia. Length Cast iron Wood Steel Conc. Gasket Bolts
1 Mount Royal Tunnel a CAN 6.1 m 500 m   
Montreal, Quebec 1912 (20’) (1,650LF)
2 Parmley System b USA Various Various   
Various locations >1902
3 Japanese tunnels c Japan Various Various   
Various locations >1970
4 Buchan WMATA d USA Various Various    
Washington, DC >1980
5 Arrowhead Tunnels e USA 4.9 m 12.8 km   
San Bernardino, CA >2002 (16’) (42,000 LF)
6 OARS Tunnel e USA 7.0 m 7.01 km   
Columbus, OH 2013 (23’) (23,00 LF)
Notes:
a
According to contemporaneous construction reports published in 1912, segment molds were fabricated from cast iron, steel and wood. A threaded central
lifting bolt was used for handling and placement by an erector arm.
b
Molds for the Parmley system were made from fabricated steel. It appears that low slump concrete was poured (or pressed) into place. Molds may have also
had incidental wooden inserts for bolt pocket block-​o uts
c
1980-​e ra Japanese segment precasting utilized fabricated steel molds and steel block-​o uts. None of the contact surfaces were machined. Dimensional control
was achieved with internal measurement devices (calipers). Operation of the molds was very labor-​intensive and without any automation.
d
The Buchan casting system was based on the use of precast concrete bottom molds and locking wooden side walls (with gasket grooves). Additional metal and
wood inserts were used as block-​o uts for bolt pockets, etc.
e
Arrowhead Tunnels segment precasting utilized fabricated steel molds and steel block-​o uts. Segments were 330 mm thick × 1.50 m long. All contact sur-
faces were machined and dimensional control was achieved with internal measurement devices (calipers) and high precision electronic measurement devices.
Operation of the molds was very labor-​intensive but benefited from the use of an automated carousel casting and curing production line.
 Precast segmental tunnel lining history 41

Figure 1.37 C ast iron segment forming system.

Note: Cast iron form used to fabricate the O’Rourke Concrete Block Arch precast segmental
tunnel lining elements for the Mount Royal Tunnel.
Source: Railway Age Gazette, page 858, Vol 59, No.19, 1915. 

feet circumferentially. They have plane faces on the intrados and on the radial joints but
are provided on one vertical face with concrete keys or tenons 12 x 20 inches in section
and 7 inches high, while of the opposite face contains depressions of the same shape to
engage the projections of the blocks in the adjacent ring of the arch.

Figures 1.37 and 1.38 illustrate the steel and cast iron molds used to form Interlocking
Concrete Blocks for the Mount Royal Tunnel project (circa 1912).
Figure 1.39 below illustrates a patented 1930-​era precast segment casting operation for
multiple (side-​by-​side) tunnel project.
Modern precast segmental tunnel liners are commonly fabricated from 0.10 to 0.05 mm
(0.004 to 0.002 inch) tolerance. This precision and repeatability greatly contribute to the
safe and reliable handling at the plant, site and in the tunnel. Individual segment unit weights
are also carefully controlled resulting in repetitive, predictable handling through all phases
of fabrication and tunnelling operations.
A more thorough discussion on modern segment molds and casting facilities is enclosed
in Chapter 8 and Chapter 9. Additional information on segment handling and installation
is enclosed in Chapter 10.
42 Handbook of Precast Segmental Tunnel Lining Systems

Figure 1.38 O ’Rourke segment casting form.

Note: Patent drawing for an O’Rourke Interlocking Tunnel Block cast iron form.
Source: US Patent No.1,976,627 drawing dated 09 Oct 1934. 

Figure 1.39 O ’Rourke interlocking tunnel block casting system.

Note: Patent drawing for segment casting operations (and molds) for the O’Rourke Interlocking
Tunnel Blocks.
Source: US Patent No. 1,976,627 drawing dated 09 Oct 1934. 
 Precast segmental tunnel lining history 43

1.13 LOGISTICS OF SUPPLY FOR PRECAST SEGMENTAL TUNNEL

LINING MATERIALS
Historical and technological development of tunnel linings including precast concrete seg-
ments have been heavily influenced by logistical and material supply challenges including
local area transportation and manufacturing capabilities. Indeed, raw material sourcing on
the front end and finished material handling afterwards has greatly influenced the selec-
tion (and preference) of the tunnel lining system and materials. In some cases, the selection
of tunnel lining approach was heavily biased towards locally available fabrication facil-
ities and the corresponding local skilled labor capabilities, and not the most economical or
technologically optimized solution.
Consider the following as listed in Table 1.6 as a concise summary of the logistical and
material supply challenges encountered during the decades old historical and technological
development of precast segmental tunnel linings.
A more thorough discussion on modern segment molds and casting facilities is enclosed
in Chapter 8 and Chapter 9.

1.14 HEALTH AND SAFETY ASPECTS OF PRECAST SEGMENTAL

TUNNEL LININGS
Health and Safety considerations and worker training have been assertively addressed in
the progressive development of precast segmental tunnel linings. In that many of the same
handling and hoisting methods exist for cast iron and later for fabricated steel segments,
Health and Safety provisions for precast concrete segments had to be carefully considered
starting at the fabrication plant and extending through all phases (and locations) until
final installation in the tunnel. Table 1.7 summarizes many of the hazards, mitigations and
safety systems developed for precast segment tunnel liners since the earliest segments were
installed in the 1870’s.

1.14.1 Lifting and handling precast segmental tunnel liner elements

Whereas precast segmental tunnel lining elements were commonly lifted and positioned
using the “screwed-​in ball device” modern developments include mechanical grapples,
clamps and vacuum lifting devices for all segments in the plant, site and tunnel environ-
ments. Concrete segments using steel fibers (SFRC) require multiple-​point support or dis-
tributed lifting (i.e. vacuum) to avoid concentrated stress points. Additionally, vacuum
lifting devices are adapted to the finished segment surfaces as well as all appurtenances and,
vice versa; to achieve 100% reliable intimate contact with all segments comprising the ring.
The TBM segment erection system has evolved from robust mechanisms to highly con-
trolled devices capable of very precise positioning and adjustment of individual segments
(all axes of movement) and all appurtenant features (i.e. gaskets, alignment and connection
devices). Some levels of automation and laser guidance are under development to repeatedly
achieve precise segment positioning and assembly. Refer to Figure 1.40 that illustrates the
installation of precast segmental tunnel liner elements using a hydraulic hoisting and placing
boom long after the completion of tunnel excavation.
A more thorough discussion on modern segment handling and installation operations is
enclosed in Chapter 10.
 newgenrtpdf
Table 1.6 Segmental tunnel lining logistics and material supply.

44
Generalized summary of the primary logistics, materials, facilities trade skills and transportation modes needed for fabrication of
segmental tunnel liners since first conceived in the 1870s and later refined and consolidated to precast concrete segmental tunnel

Handbook of Precast Segmental Tunnel Lining Systems

linings in the 1970s

Fabrication operations Materials needed Transport’n

Item Description Method Labor Supplies Fuel/power Rail Ship Road
A Prior to precast concrete segmental lining development
1 Cast iron: gray, Foundry casting Skilled labor Bulk supplies of Depending on   1
malleable/ductile, operations: classes: the following: manufacturing
and spheroidal • Shop and plant • Foundry • Gray cast iron era, the
graphite facilities workers (all • Malleable and following:
• Furnaces skill sets) ductile cast iron • Coal
• M olds • Heavy • Spheroidal • Fuel oil
• H eavy hoisting equipment graphite iron • Natural gas
• Machining facilities operators (SGI) • High voltage
• Machinists electrical power
• Millwrights
2 Fabricated Plate rolling, Skilled labor Bulk supplies of Depending on  2 
steel: malleable/ cutting and welding classes: the following: manufacturing
ductile, highly operations: • Fabrication • Plate sections era, the
weldable and readily • Shop and plant workers (all • Rolled sections following:
machinable facilities skill sets • Pipe • Coal (for steam
• Calendar rollers • Welders • Welding supplies boilers)
• Stamping, punching • Heavy • Welding gases • Fuel oil
and shearing equipment • Natural gas
• Welding operators • High-voltage
• Heavy hoisting • Machinists electrical power
• Machining facilities • Millwrights
B Precast concrete segmental linings – modern-era developments
1 Precast concrete – Precast concrete Skilled labor Bulk supplies of Depending on   3
early stage and not operations: classes: the following: manufacturing
fully mechanized; • Shop and plant • Precast plant • Straight and era, the
ready-mix concrete, facilities workers (all bent deformed following:
static molds and • Overhead cranes skill sets rebar • Fuel oil for
deformed bar • Rebar fabrication • Heavy • Welding supplies steam boilers
reinforcing • Segment molds equipment • Ready-mix • Natural gas
• Concrete placing operators concrete • High-voltage
• Segment finishing • Electrician • Segment electrical power
• Gasket installation • Quality accessories • Steam heat
• Handling & storage yard Control • Gaskets
• QC lab facilities Technician
2 Precast concrete – Precast concrete Skilled labor Bulk supplies of Depending on   3
early stage operations: classes: the following: manufacturing
with some • Shop and plant • Precast plant • Straight and era, the
mechanization; facilities workers (all bent deformed following:
ready-mix concrete, • Overhead cranes skill sets rebar • Fuel oil for
carousel production • M echanize and • Heavy • Welded wire steam boilers
line, machined molds, automated rebar and equipment mesh • Natural gas

Precast segmental tunnel lining history

deformed bar and WWM fabrication operators • Welding supplies • High-voltage
WWM reinforcing • S egment molds with • Electrician • Ready-mix electrical power
machined surfaces • Millwrights concrete • Steam heat
• C arousel production (for machined • Segment
line system molds) accessories
• O n-site concrete batch • Quality • Gaskets
plant Control
• Concrete placing & Technician
finishing
• Gasket installation
• Handling & storage
yard
• QC lab facilities
(Continued)

45
 newgenrtpdf
Table 1.6 (Continued)

46
Handbook of Precast Segmental Tunnel Lining Systems
Fabrication operations Materials needed Transport’n
Item Description Method Labor Supplies Fuel/power Rail Ship Road
3 Precast concrete – Precast concrete Skilled labor Bulk supplies of Depending on   3
advanced operations: classes: the following: manufacturing
stage with full • Shop and plant • Precast plant • Steel fibers era, the
mechanization; facilities workers (all • Welding supplies following:
ready-mix concrete, • O verhead cranes with skill sets • Ready-mix • Natural gas
carousel production GPS controls; repetitive • Heavy concrete • High-voltage
line, machined and handling equipment • Segment electrical power
power-assisted • M echanize and operators accessories • Steam heat
molds, steel fiber automated steel fiber • Electrician • Anchored
reinforcing handling • E lectronic gaskets
• S egment molds with • P LC
machined surfaces, Programmers
power assist and • Millwrights
anchored gaskets (for machined
• C arousel production molds)
line system • Quality
• O n-site concrete batch Control
plant Technician
• Concrete placing &
finishing
• Handling & storage
yard
• QC lab facilities
Notes:
1
In the late 1800’s and early 1900’s, highway systems were very limited and primitive and definitely not suitable for heavy load transport. Rail and marine (ship)
modes were preferred. Fabrication facilities were, therefore, located at transportation hubs.
2
As fabricated steel segments were popular for a short period (mid-1990’s), the use of marine (ship) transport was very limited with the preference being rail
and road. Logistically, fabrication facilities were, therefore, located at transportation hubs.
3
As precast concrete segmental tunnel liners gained in popularity (in the 1970’s and to the present), and the location of suitable plant facilities were more
flexible, the use of rail and marine (ship) transport became very limited with the preference being road (highways). Logistically, precast segment fabrication
facilities were, therefore, located at highway transportation hubs.
Table Legend:
• Primary condition or majority of the work
 
Secondary (infrequent) condition or occasional encounter to the work
 Precast segmental tunnel lining history 47

Table 1.7 H ealth and Safety considerations for precast segmental tunnel linings.
Summary of the primary Health and Safety considerations for the manufacturing,
handling, storage, transport and installation of the precast segmental tunnel liners

Manufacturing and construction hazards

Item Operational description Pinch Fall Hit by Electr’l Chem’l
A Segment fabrication plant
1 Segment casting operations A1    
2 Demolding and curing segments A2   
3 Segment hoisting and storage A3    
B Tunnel handling
1 Segment storage on-​s ite B1  
2 Segment handling and sorting B2  
3 Segment prepping (e.g. gaskets) B3   
C Tunnel installation
1 Segment transport into tunnel C1   
2 Segment handling in the tunnel +​ TBM C2   
3 Segment erection in TBM tail shield C3   
Special Table Notes and Mitigations to Listed Hazards:
A1 Segment fabrication plants are highly mechanized and are progressively moving toward more automa-
tion for repetitive operations. Additionally, extensive crew training, active safety systems are normal
practices.
A2 Demolding and curing operations often include rolling and repositioning segments quickly and efficiently.
This has required more mechanization, controls and safety devices to avoid hazards.
A3 Final segment handling, hoisting and storage has become far more mechanized, programmable and auto-
mated, using positioning devices and proximality controls in limited access, dedicated storage and hand-
ling (i.e. load-​o ut) facilities.
B1 On-​s ite storage generally in dedicated areas with inventory and handling controls
B2 Use of well-​p ositioned high capacity hoisting equipment is the norm. Also custom, power actuated lifting
cradles (clamps) have largely replaced ordinary slings and chains
B3 Temperature controls applied to segment intrados and extrados surface before hoisting. Gaskets and
glue rarely applied in the field (preference to factory installation).
C1 S pecialized rail cars and multi-​s ervice vehicles transporters are commonly used to transport precast
segment in the tunnels to the TBM.
C2 Segment feeder system located within the TBM trailing gear avoids low head room hoisting and handling
of individual segments.
C3 Use of vacuum lifting devices is well accepted. Use of high precision positioning devices avoids segment
damage and increases ease of installation.
Table Legend:
• Primary condition or majority of the work
 
Secondary (infrequent) condition or occasional encounter to the work

1.14.2 Backfill grouting the segmental tunnel liner

Backfill grouting of the installed precast segmental tunnel liner requires cementitious mate-
rials, sand and gravel on occasion, and construction chemicals, all pumped and placed at
elevated pressures. Considerable safeguards are in place for handling, mixing, pumping and
placing backfill grout such as the following.
48 Handbook of Precast Segmental Tunnel Lining Systems

Figure 1.40 Overhead precast segment handling – mechanized system.

Note: Lifting and placing precast segment cavern liners using a vacuum lifting device and hydraul-
ically powered boom. Note that a waterproof membrane has been previously installed.
Source: Verya Nasri, personal photo collection.

• Personnel protective equipment (PPE)

• Material safety data sheets (MSDS)
• Pressure relieving devices
• Controls and interlocks with concurrent TBM operations
• Real time voice and data communications
• Segment build quality: i.e. tensioned erection bolts, fully engaged sockets and inter-
locking devices

A more thorough discussion on modern segment backfilling is enclosed in Chapter 11.

1.15 CONCLUSIONS AND RECOMMENDATIONS

The history and technical development of precast segmental tunnel linings has undergone a
series of memorable events and important technical milestones since it was first conceived
in 1870. It was not until the 1930’s and a shortage of cast iron material in the United
Kingdom, that concrete was considered as an alternative material. This transition initiated
further development and acceptance into a highly merchantable product and readily adapt-
able to small tunnel requirements. Further significant developments in materials, connec-
tions and sealing systems occurred in the 1970’s to 1990’s and onto the present that have
enhanced precast segmental tunnel lining elements as the preferred tunnel lining system for
all sizes and configurations of transportation, sewer and flood control tunnels.
Movement away from fabricated steel and cast iron was gradual as the economics and
sustainability of precast segmental tunnel lining took precedence in the design of tunnels that
were progressively located in more egregious ground conditions and depths. Additionally,
precast segmental tunnel linings were considered for shafts, declines and caverns too in add-
ition to non-​circular applications.
 Precast segmental tunnel lining history 49

Figure 1.41 A rrowhead Tunnels Project, San Bernardino, California, United States.
Note: A very robust precast segmental tunnel lining system was designed and installed on this
project. In this case there were two different concrete design strengths needed to suit subsur-
face conditions and a single gasket sealing system designed for 27 bar groundwater pressure.
Source: Brian Fulcher, personal photo collection. 

Precast segmental tunnel linings also benefited from changes (enhanced) in material
properties focused on concrete, reinforcement and sealing systems that occurred in recent
decades. These changes increased durability and the estimated service lives of the installa-
tions. Sealing systems and corresponding connection devices have advanced to the point
where 30 bar groundwater pressures are readily achievable. The use of a very robust precast
segmental tunnel lining for the Arrowhead Tunnels Project located in San Bernardino,
California was instrumental to the success of the project. The use of state-​of-​the-​art TBMs
to penetrate very egregious ground and groundwater conditions could not have been safely
and successfully accomplished without the concurrent installation of a precast segmental
tunnel lining specifically designed for highly variable ground conditions including squeezing
ground occurrences and very high groundwater pressures, Figure 1.41.
At the manufacturing stage, efficiency, accuracy and reliable repeatability have greatly
improved. High precision and low tolerance molds and carousel casting and curing pro-
cesses for increasing daily segment production rates are also the State-​of-​the-​Art resulting in
considerable fabrication economies over alternate materials. Manufacturing improvements
currently underway include increased automation and extensive use of robotics. It should
be noted that the “build tolerance” and watertightness of segmental tunnel linings is a direct
reflection on design and fabrication of segment molds as well as strict protocols on segment
casting, curing, storage and handling. Whereas, the original precast segmental tunnel lining
molds were built from cast iron, steel, concrete and wood, they are now universally fabri-
cated from very high precision machined and polished rigid steel frames.
50 Handbook of Precast Segmental Tunnel Lining Systems

Future developments for precast segmental tunnel linings will include additional auto-
mation in their manufacture, handling and installation as well as improved durability and
service lives. They will also be further adapted for non-​circular tunnels and rehabilitation
projects to extend (or repurpose) the service lives of existing underground facilities.

BIBLIOGRAPHY AND ADDITIONAL READING MATERIALS

Many published documents were accessed and utilized in the preparation of this chapter. These are
listed below as well as many other related references as additional reading materials.

PUBLISHED BOOKS, REPORTS AND PAPERS

Bartlett, J.V., Noskiewicz, T.M., Ramsay, J.A., 1971, “Precast Concrete Tunnel Lining for Toronto
Subway”, Journal of the Construction Division, ASCE, Volume 97, No.C02, pp. 241–​256.
Burr, S.D.V., 1885, Tunnelling Under the Hudson River, The American Society of Civil Engineers,
New York.
Copperthwaite, William Charles, 1905, Tunnel Shields and the Use of Compressed Air in Subaqueous
Works (1912), D. Van Nostrand, New York.
Gilbert, Clare, 1991, St. Clair Tunnel; Rails Beneath the River, Boston Mills Press, Erin, Ontario,
Canada.
Glover, John, 1999, London’s Underground, 9th Edition, Ian Allan.
Greathead, J.H., 1896, The City and South London Railway: With Some Remarks Upon Subaqueous
Tunnelling by Shield and Compressed Air, D. Van Nostrand, New York.
Green, Oliver Green, 1987, The London Underground: An Illustrated History, Ian Allan in associ-
ation with the London Transport Museum, London.
Groseclose, William R., and Schoeman, Kenneth D., 1976, “Precast Concrete Segment Lining for
Buckskin Mountain Water Conveyance Tunnel”, Rapid Excavation and Tunnelling Conference,
Las Vegas, NV, Society for Mining Engineering, Denver, CO.
Hewitt, B.H.M. and Johannesson, S., 1922, Shield and Compressed Air Tunneling, McGraw-​Hill,
New York.
Lauchli, Eugene, 1915, Tunnelling–​Short and Long Tunnels of Small and Large Sections Driven
Through Hard and Soft Materials, McGraw-​Hill, New York.
McBean. R.J., and Harries, D.A., 1970, “The Development of High-​Speed Soft Ground Tunnelling
Using Precast Concrete Segments and Tunnelling Machines”, The Technology and Potential of
Tunnelling, Proceedings, The South African Tunnelling Conference, Volume 1, pp. 129–​134.
O’Rourke, T.D. (Editor), 1984, Guidelines for Tunnel Lining Design, prepared by the Technical
Committee on Tunnel Lining Design of the Underground Technology Research Council.
American Society of Civil Engineers, New York, Library of Congress No.84-​70576, ISBN
0 87262 402 1.
Perlini, Charles, 1912, Tunnelling: A Practical Treatise, D. Van Nostrand, New York.
Ricahrdson, Harold W. and Mayo, Robert S., 1942, Practical Tunnel Driving, McGraw-​Hill,
New York.
B.F. Sturtevant Company, 1927, The Eighth Wonder, Cambridge University Press, Cambridge, MA.
Széchy, Karoly, 1973, The Art of Tunnelling, 2nd English edition. Akademiai Kiado, Budapest.
Wolf, Donald E., 2010, Crossing the Hudson: Historic Bridges and Tunnels of the River, Rivergate
Books, Rutgers University Press, New Brunswick, NJ.
Wolmar, Christian, 2005, The Subterranean Railway, Atlantic Books, London.

REFERENCE PUBLICATIONS PREPARED FOR TECHNICAL

ASSOCIATIONS
AFTES Recommendations–​Choosing Mechanized Tunnelling Techniques, 2000, Association Franҫais
de Tunnels and de l’espace Souterrain.
 Precast segmental tunnel lining history 51

AFTES Recommendations–​Design, Sizing and Construction of Precast Concrete Segments Installed

at the Rear of a Tunnel Boring Machine (TBM), 1999, Association Franҫais de Tunnels and de
l’espace Souterrain.
Tunnel Design–​ Design of Concrete Segmental Tunnel Linings–​ Code of Practice, 2016, British
Standards Institute; ISBN 978 0 580 88170 1.
Tunnel Lining Design Guide, 2004, British Tunnelling Society and the Institution of Civil Engineers;
ISBN 07277 2986 1 1.
Specification for Tunnneling, 2010, British Tunnelling Society and the Institution of Civil Engineers;
ISBN 978 0 7277 3477 8.

REFERENCE PUBLICATIONS PREPARED FOR GOVERNMENT

AGENCIES
Craig, R.N. and Muir-​ Wood, A.M., 1978, A Review of Tunnel Lining Practice in the United
Kingdom–​Supplementary Report 335, Tunnel Division, Structures Department, Transport and
Road Research Laboratory, Crowthorne, Berkshire, England.
Golder Associates, James F. MacLaren Limited, 1976, Tunnelling Technology–​An Appraisal of the
State of the Art for Application to Transit Systems, Ontario Ministry of Transportation and
Communications, Toronto, Ontario, Canada.
Halverson, G.T., Kesler, C.E. and Paul, S.L., 1975, Concrete for Tunnel Liners: Mix Design
Recommendations or Prototype Extruded Liner System–​Final Report, United States Department
of Transportation, Federal Railroad Administration (Contract DOT FR 30022). Report No.
FRA-​OR&D 75-​89.
Herring, K.S. and Kesler, C.E., 1974, Concrete for Tunnel Liners: Behavior of Steel Fiber
Reinforced Concrete Under Combined Loads–​Final Report, United States Department of
Transportation, Federal Railroad Administration (Contract DOT FR 30022). Report No.
FRA-​ORDD 75-​7.
Mayo, Robert S., Adair, Thomas and Jenny, Robert, 1968, Tunneling–​ The State of the Art–​A
Review and Evaluation of Current Tunnelling Techniques and Costs, with Emphasis on their
Application to Urban Rapid Transit Systems in USA, U.S Department of Housing and urban
Development (Contract H-​766).
Mayo, Robert S., Barrett, James E., Jenny and Robert, J., 1976, Tunneling: The State of the
Industry–​Final Report, U.S Department of Transportation, Office of the Secretary
for Systems Development and Technology (Contract DOT-​ TSC-​ 1091). Report No.
DOT-​TSC-​OST-​76-​29.
Paul, S.L. and Stinnamon, G.K., 1975, Concrete for Tunnel Liners: Structural Testing of Segmented
Liners–​Final Report, United States Department of Transportation, Federal Railroad
Administration (Contract DOT FR 30022). Report No. FRA-​OR&D 75-​93.
Selander, C. E., Nelson, C. A. and Jones, B.V., 1980, Segmented Concrete Tunnel Liner and Sealant
Systems–​Final Report, U.S Department of Transportation, Urban Mass Transportation
Administration, Office of Technology Development and Deployment, Office of Rail and
Construction Technology (Contract (RA) 76-​22). Report No. UMTA-​MA-​06-​0100-​80-​9.
Tunnel Construction–​State of the Art and Research Needs–​Special Report 171, 1977, Transportation
Research Board, Commission on Sociotechnical Systems, National Research Council. National
Academy of Sciences.
Technical Manual for Design and Construction of Road Tunnel–​Civil Elements–​Publication No.
FHWA-​NHI-​10-​034, 2009, United States Department of Transportation, Federal Highway
Administration.
Technical Record of Construction, Buckskin Mountains Tunnel, 1983, United States Department of
the Interior, Bureau of Reclamation. Granite Reef Division, Central Arizona Project. Phoenix,
AZ: Arizona Projects Office.
52 Handbook of Precast Segmental Tunnel Lining Systems

REFERENCES, PERIODICALS AND FURTHER READING

MATERIALS
Armstrong, J. and Gourvish, T., 2000, “London’s Railways–​Their Contribution to Solving the Problem
of Growth and Expansion”, Japan Railway & Transport Review 23, March 2000.
Arthur, H.G., 1976, “The Buckskin Mountain Tunnel”, Western Construction, Volume 51, No. 10,
pp. 48–​50.
Donovan, H.J., 1974, “Expanded Tunnel Linings”, Tunnels and Tunnelling, Volume 6, No. 2,
pp. 46–​51.
Harding, A. and Chappell, M., 2015, “Segmental Linings: A Vision for the Future”, Tunnelling
Journal, February/​March Volume, pp. 31–​37.
Joy, A., 1976, “One-​Step Method Speeds Sewer Tunnel”, Heavy Construction News, Volume 20,
No. 41, pp. 28–​29.
“James Greathead Statue: London Remembers” (www.lond​onre​memb​ers.com)
The Mont Royal Tunnel, 1913, built by Mackenzie, Mann and Company. For The Canadian Northern
Montreal Tunnel and Terminal Co. Limited December 1913.
The Mont Royal Tunnel–​A Description of Construction of the Tunnel and Terminal Built by the Mont
Royal Tunnel and Terminal Company for the Canadian Northern Railway at Montreal, The
Engineering Institute of Canada, April 1919.
Railway Age Gazette, 1915, Volume 59, No. 19.
Wright, L., 2017, “James Henry Greathead and the London Underground”, Literator, Volume 38,
No. 1 (www.litera​tor.org.za).

Commercial literature resources

Charcon Tunnels Limited, 1990, Shaft and Tunnel Linings. UK.
Charcon Tunnels Limited, 1990, Bolted Tunnel Linings. UK.
Commercial Shearing Inc., 1975, Tunnel Lining Sealants, Product report prepared by Commercial
Shearing, Youngstown, OH.
Commercial Shearing Company, 1956, Civil Engineering, Contractors and Manufacturers of Precast
Concrete Lining for Shafts and Tunnels.
Buchan, C.V., 1956, Civil Engineering, Contractors and Manufacturers of Precast Concrete Lining
for Shafts and Tunnels.
Kinnear Moodie & Co., 1956, Civil Engineering, Contractors and Manufacturers of Precast Concrete
lining for Shafts and Tunnels. UK.
Kinnear Moodie & Co., 1989, Bolted Linings. UK.
Parmley, 1927, The Parmley System of Arch Construction, Catalogue E, Segmental Concrete–​Plain
and Reinforced. USA.
Chapter 2

Geotechnical investigations for

segmental tunnel lining design
Lead authors: Francis J. Arland and James A. Morrison
Contributing authors: Michael Law, Andrew Klaetsch, Gary Kramer,
and Ritika Kundu

2.1 INTRODUCTION
A tunnel system is designed and built based on a series of assumptions and simplifications
developed by the engineer to model the natural geologic environment. The actual conditions
of the geologic realm are far too complex to capture completely and accurately, and the
ability to sample the environment is only possible at discrete points of space and time. For
example, if 50 mm (2 inches) diameter specimens from borings are collected at 165 m (540
feet) intervals along a tunnel alignment, the engineer has only physically examined about
1.0 ppm of the soil that will impact the tunnel. The tools available to the engineer to create
the best possible model of the environment as a basis of design include accurate and reli-
able sampling and testing methods, and a thorough understanding of the natural science of
geology to connect the dots. Technology such as remote sensing, LiDAR, and drone surveil-
lance is also available to help the engineer in understanding the bigger picture and regional
perspective necessary to interpret geologic conditions.

2.2 GEOLOGY LEADS THE DESIGN OF THE SEGMENTAL

TUNNEL LINING
The role of geotechnology in the design process is to prepare and define the parameters for
a tunnel lining system that will meet the loading, performance and durability characteristics
required of the tunnel. To meet this, the designer is focused on developing a geotechnical
model that assures the following:

• Minimum/​maximum soil/​rock strengths are identified and incorporated with appro-

priate factor of safety
• Potential changes in loading over the life of the project are accounted for, either nat-
ural or man-​made
• Maximum water pressure and/​or seismic loads that could be potentially encountered
during the project lifetime are accommodated
• Construction materials are designed to be compatible with the chemistry of the nat-
ural environment, or the environment to which they will be subject

The contractor, on the other hand, is generally responsible to determine the appropriate
equipment and installation methods necessary to build the tunnel and assure that the

DOI: 10.1201/9781032675541-2
54 Handbook of Precast Segmental Tunnel Lining Systems

specifications of the designer are met. To meet this objective, the contractor will be focused
on developing a geotechnical model that defines the following:

• Actual range of materials and material strengths to be encountered

• Material quantities to be encountered and variability within materials
• Parameters affecting tool selection and wear such as abrasiveness and hardness
• Actual range of groundwater pressure and hydraulic parameters (such as water
inflows) to be encountered during construction
• Environmental exposure that affects the safety of workers

While there is overlap, the use of the geotechnical model is quite different between designer
and contractor. The geotechnical investigation, data and reporting prepared during the
design phase of the project ultimately needs to address both objectives. The art of preparing
and reporting geotechnical data that covers each objective is discussed briefly in Section 2.7
but warrants further discussion and integration with the overall contracting approach in
other reference documents.
The purpose of this section is to present the reader with a general overview and guidance
to the principles of geotechnical engineering that are typical in segmental tunnel design and
construction. It is not intended to be a comprehensive document to define geotechnology,
or to facilitate other types of subsurface design. Many detailed references have been devel-
oped as outlined in Sections 2.8 through 2.10, and the reader is directed to these references
for the detailed development and evaluation of specific parameters to be incorporated into
a design.

2.2.1 Difference between factors critical for design and factors critical

to construction
The designer is typically focused on contractual and regulatory compliance, evaluating the
loads imposed on a structure, and sizing the structural components to meet those load
cases. Factors such as material composition, weight, internal stress and material strength
are critical. Where a designer may be focused on designing a structure to support theoret-
ical maximum loads with the minimum design soil strength, a contractor’s interpretation of
geotechnical information is most likely focused on the maximum strength of a soil to design
efficient cutting tools or to size efficient equipment instead of theoretical maximum loads
used in design. In general practice, a single geotechnical field investigation is used by both
parties to execute the project. This difference in use and interpretation of geotechnical data
needs a proper geotechnical investigation to cover and address both considerations.
In determining the information to include in a specific geotechnical report the geotech-
nical engineer must first understand how and where the information will be used, and by
whom. The data should be aligned with the decisions that influence cost and time of per-
formance. This includes both the type of data, and the interpretation of possible ranges in
the data. Factors that impact cost are defined as those values of soil or rock strength or
composition that cause a shift from one construction method, or grade of equipment, to
another. Understanding the critical factors and cost impact points typically requires input
from other parties in addition to the geotechnical engineer. It is advisable to consult with
industry construction experts, specialty technical experts, and the lead design professionals
to develop and interpret the appropriate geotechnical parameters and interpretation of data
that align with cost.
 Geotechnical investigations for segmental tunnel lining design 55

2.3 SPECTRUM OF EARTH MATERIALS

Most tunnel construction occurs in naturally deposited geological environments, but
increasingly tunnel designers need to consider man-​made features and re-​worked soils in
urban environments. Our understanding and terminology for describing and analyzing geo-
technical behavior is based on soils originating from nature. The ability to project and pre-
dict behavior between sample points is based on our understanding of natural geology but is
far more complicated in altered urban settings. The following sections set out the common
classification and nomenclature systems used to describe the natural earth environment.

2.3.1 Soft ground
The materials described as soft ground are the weathering products of decomposed rock and
consist of an accumulation of individual particles of varying size and chemical composition.
Classification and behavior of soft ground is influenced by its deposition, chemical compos-
ition, grain size and shape, and stress and loading history. From an engineering perspective
soft ground materials are broadly grouped into two categories, cohesive and non-​cohesive,
based on their behavior and their properties of interaction between individual soil particles.
Cohesive soils are defined as a material that retains its strength, electrostatic forces and/​or
capillary forces between soil particles. From a tunneling perspective, cohesive soils exhibit
some degree of stand-​up time during excavation but will deform in a plastic manner while
retaining individual particle connections. Cohesive soils also generally exhibit low perme-
ability and act as a barrier to water flow and drainage and may exhibit sticky behavior
during excavation.
Non-​ cohesive soils represent an aggregate of particles that exhibit strength through
gravity and frictional forces between particles. The relative strength of a non-​cohesive soil
is related to the particle size gradation, layering, and shape or roughness of individual par-
ticles. The relative strength generally increases with increased confining stress. From a tun-
neling perspective, non-​cohesive soils will lose strength and displace by flowing or raveling
with the loss of confinement and/​or loss of capillary forces due to drainage.
Naturally deposited soils are often a mixture or layering of different soil types and par-
ticle sizes and will exhibit a complex overall behavior profile during mining. Further dis-
tinction is given to materials that behave somewhere between soil and rock, and these are
referred to as intermediate geotechnical materials or IGM discussed further below. Soil
types by deposition include the following:

• Alluvium
Includes soils deposited by flowing water. Alluvial soils tend to be stratified in depos-
ition, with particle size bands being related to the water velocity at the time of depos-
ition. Ancient alluvial deposits may appear to be highly variable in vertical and lateral
extent when exposed during an excavation. From a tunneling perspective, complica-
tions arise due to the variability of porosity and the concentration of water flow in the
open-​graded coarse deposits, and the water trapping potential of fine grained and clay
strata.
• Lacustrine
Includes soils deposited in still water or lake environments. Like alluvium, lacustrine
deposits will be stratified in composition, but are generally fine grained. Depositional
56 Handbook of Precast Segmental Tunnel Lining Systems

strata generally reflect seasonal variations in water flow. Silt and fine sand strata tend to
accumulate where a water body is active with precipitation or stream flow, such as spring
melt and run-​off or periods of wet weather. During periods of still water, the finer grained
silt and clay soil minerals fall out of suspension, forming clay strata. From a tunneling
perspective, lacustrine soils can manifest with stand-​up time and stability issues, where
water transports through the coarser grained strata resulting in slip planes. Instability can
also result if excess pore pressure develops in the finer grained silts and clays. Effectively
dewatering or depressurizing lacustrine deposits can prove problematic, due to the typ-
ical overall fine-​grained nature and thin horizontal bedding of the soils.
• Glacial
Glacial soils have been formed by the action of flowing ice and from distinctive pat-
terns of deposition. Kames and eskers are concentrated regions of generally coarse
material deposited by water flowing through or beneath a glacier. After the glacier has
retreated, these deposits form stratified ridges and long linear features in the topog-
raphy. Glacial till is the ground-​up residue typically formed as the ice sheet advances
across native ground or exposed rock.
Till is generally comprised of a wide range of particle sizes, from clay fraction to
cobbles and boulders. From a tunneling perspective, till deposits are typically dense
and have substantial strength, offering excellent stand-​up time during excavation.
However, cobbles and boulders result in obstructions to mining, and impact the effi-
ciency and design of mining equipment. Predicting the presence and concentration of
boulders is problematic, as the occurrence and frequency is difficult to capture with
conventional soil sampling methods.
• Wind-​blown
Wind-​blown soils deposits are typically referred to as loess and exist in large expanses
of the great plains of North America, and the Eurasian steppes. Loess is characterized
by fine, angular particles that lock together into soil deposits that have both strength
and low density. Stand-​up time for mining can be very advantageous, as well as the
natural ability to drain rapidly due to the relatively high void ratio. Problems with
loess deposits include the tendency to collapse and lose strength when disturbed by
excavation or vibration.
• Residual
Residual soils are the product of parent rock decomposing in place into a soil con-
sistency. Residual soils are extensive in the southeast United States in an area known
as the Piedmont region. Within the Piedmont, residual soils are extremely variable
over short distances due to weathering along ancient rock fractures and can result in
extreme variation from soft ground to rock within the space of a single excavation.
The formation and strata of residual soils mirrors the minerology and fracturing of the
parent rock. Subsequently, slip planes and water flow channels will appear random
and variable in an excavation.

2.3.1.1 Soft ground classification methods

Geotechnical classification systems have been developed by different groups for different
end purposes. The purpose of a classification system is to create a codified language so that
 Geotechnical investigations for segmental tunnel lining design 57

multiple parties can understand the terminology used in describing a soil or rock. From
this common language, a common understanding of expected behavior is communicated.
For tunneling, the Designer needs to understand the material type, its weight, density, and
strength characteristics. The Contractor will be interpreting the ground for its resistance to
excavation, ability to transmit water, and other physical characteristics necessary to define
his means and methods of construction.

2.3.1.2 ASTM D-​2487 and ASTM D-​2488

American Society of Testing and Materials (ASTM) has codified the description of soft
ground soils based on laboratory standardized testing (ASTM D2487) and by visual-​
manual methods (ASTM D2488). Based on grain size distribution, ASTM D2487 divides
soils into two categories, fine-​grained and coarse-​grained, based on whether 50% of the
particles within a soil mass are greater or smaller than 0.074 mm (0.003 inches) or passing
a No.200 sieve.
Further subdivision for coarse-​grained soils relates to the distribution of soil particle
sizes within the soil mass, whether the particle sizes are broadly distributed (W =​well
graded) or more uniform in particle size (P =​poorly graded). Fine grained soils are sub-
divided based on physical composition and the properties of inter-​particle cohesion based
on the results of the Atterberg limit tests. Silt (M) and clay (C) designations are defined
based on comparing the measurement of liquid and plastic limits, as defined in the ASTM
procedure.
Further definition of the properties of plasticity (ability to retain a deformed shape) are
also provided within the testing procedure; where H is defined as high plasticity, and L is
defined as low plasticity. Table 2.1 below presents the ASTM D2487 classification system
in its entirety.

2.3.1.3 Tunnelman’s soil classification and behavior descriptions

The Tunnelman’s soil classification was first developed by Karl Terzaghi in the 1950’s and
updated and refined by Ron Heuer in 1974 as a systematic method for describing ground
behavior during tunneling, as opposed to simply describing the physical characteristics of
the ground. Summarized in Table 2.2, the system was published in 1977 by Commercial
Shearing, Inc. “Earth Tunneling with Steel Supports” (for soft ground tunneling) by Proctor
and White [1977] and is a very common reference of terminology used for describing the
behavior of soils at an excavated tunnel face.

2.3.3 Organic soils
Encountered extensively in northern latitudes as fibrous peat or muskeg, and range to black
silt and clay where decomposition can occur over time. Properties of organic soils need to
be determined reflecting as close as possible to their natural state. Peat is fibrous, and sponge
like, which allows for very high-​water content values in the range of 50% to 2,000%.
Organic soils often have comparatively high strength per unit weight due to fibrous inter-
locking. Other general properties include high acidity, which can prove detrimental to cor-
rosion in concrete and steel, and often interferes with slurry additives used in drilling and
slurry shield TBM construction.
 newgenrtpdf
Table 2.1 ASTM 2487 – Unified Soil Classification System.

58
Summary of soil classification nomenclature and criteria as defined by ASTM

Handbook of Precast Segmental Tunnel Lining Systems

Group
Item Major division symbol Typical name Classification criteria
1

C U = D 60 / D 10 greater than 4

50% of more of coarse fraction

retained on No.4 ASTM sieve
Well-graded gravels and gravel sand
GW C c = D 230 /(D 60 × D 10)

gravels
mixtures, little or no fines

Clean
between 1 and 3
(more than 50% retained on No.200 ASTM sieve)

Gravels

2 Poorly graded gravels and grave-sand Not meeting both criteria

Less than 5% passing the No.200 ASTM sieve; GW, GP,

GP

More than 12% passing the No.200 ASTM sieve; GM,

mixtures, little or no fines for GW

5% to 12% passing No.200 ASTM sieve; borderline

Coarse grained soils

Classification on basis of percentage of fines.

3 Atterberg Limits plot below

classification requiring use of dual Symbols

Gravels with

GM Silty gravels, gravel-sand-silt mixtures A-line or Plasticity Index

less than 4
fines

4 Atterberg Limits plot above

GC Clayey gravels, gravel-sand clay mixtures A-line and Plasticity Index
greater than 7

GC, SM and SC.

SW and SP.
5 Well graded sands and gravelly sands, C U greater than 6
More than 50% coarse fraction

SW
little or no fines C c between 1 and 3
Clean
sands

6 Poorly graded sands and gravely sands, Not meeting both criteria
SP
little or no fines for SW
passes No.4
ASTM sieve

7 Atterberg Limits plot below

Sands

Sands with fines

SM Silty sands, and silt mixtures A-line or Plasticity Index

less than 4.
8
Atterberg Limits plot above
SC Clayey sands, sand-silt mixtures A-line and Plasticity index
greater than 7
 (Liquid limit 50% (Liquid limit 50% or
9 Inorganic silts, very fine sands, rock
ML

Silts and clays

flour, silty or clayey fine sands

(50% or more passes No.200

10 Inorganic clays or low to medium

less)
CL plasticity, gravelly clays, sandy clays,

Fine grained soils

silty clays, lean clays

ASTM sieve)
11 Organic silts and organic silty clays of
OL
low plasticity Check plasticity chart
11 Inorganic silts micaceous or

Silts and clays

MH diatomaceous fine sands or silts, elastic

or less)
silts

Geotechnical investigations for segmental tunnel lining design

11 CH Inorganic silts of high plasticity, fat clays
11 Inorganic clays of medium to high
OH
plasticity
11 Fibrous organic matter, will char, burn or glow.
Peat, muck and other highly organic
Highly organic clays Pt Readily identified by color, odor, spongy feel
soils
and fibrous texture.
Notes:
1
Boundary classification: soils possessing characteristics of two groups are designated by combinations of group symbol, for example, GW-GC, well-graded,
gravel sand mixture with clay binder

59
 newgenrtpdf
60
Table 2.2 Tunnelman’s ground classification of soils.
Summary data from Terzaghi (1950) and Heuer (1974) for tunneling in soft soils and their anticipated behaviors under various

Handbook of Precast Segmental Tunnel Lining Systems

ground and groundwater conditions

Classification Performance Typical soil types

Firm Heading can be advanced without initial support, and Loess above the water table; hard clay, marl, cemented
final lining can be constructed before ground starts to sand, and gravel when not highly overstressed.
move.
Raveling Slow raveling Chunks or flakes of material begin to drop out of the Residual soils or sand with small amounts of binder
arch or walls sometime after the ground has been may be fast raveling below the water table, slow
Fast raveling exposed. This is caused by loosening or overstress and raveling above. Stiff fissured clays may be slow or fast
“brittle” fracture (ground separates or breaks along raveling depending upon the degree of overstress.
distinct surfaces, as opposed to squeezing ground). In
fast raveling ground, the process starts within a few
minutes; otherwise, the ground is slow raveling.
Squeezing Ground squeezes or extrudes plastically into tunnel, Ground with low frictional strength. Rate of squeeze
without visible fracturing or loss of continuity, depends on degree of overstress. Occurs at shallow
and without perceptible increase in water content. to medium depth in clay of very soft to medium
Ductile, plastic yield and flow are caused by consistency. Stiff to hard clay under high cover may
overstress. move in combination with raveling at excavation
surface and squeezing at depth behind surface.
Running Cohesive-​ Granular materials without cohesion are unstable at Clean, dry granular materials. Apparent cohesion in
running a slope greater than their angle of repose /​f rom ±30 moist sand or weak cementation in any granular soil
to 35 degrees). When exposed at steeper slopes they may allow the material to stand for a brief period of
Running run like granulated sugar or dune sand until the slope raveling before it breaks down and runs. Such behavior
flattens to the angle of repose. is cohesive-​r unning.
Flowing A mixture of soil and water flows into the tunnel like Below the water table in silt, sand, or gravel without
a viscous fluid. The material can enter the tunnel from enough clay content to give significant cohesion and
the invert as well as from the face, crown, and walls plasticity. May also occur in highly sensitive clay when
and can flow for great distances, completely filling the such material is disturbed.
tunnel in some cases.
Swelling Ground absorbs water, increases in volume, and Highly pre-​c onsolidated clay with a plasticity index
expands slowly into the tunnel. above 30, generally containing significant percentages
of montmorillonite clay.
 Geotechnical investigations for segmental tunnel lining design 61

The following issues should be considered during tunnel design if organic soils are
identified:

• Settlement due to changes in water content

• Corrosion potential for concrete and steel
• Creation of hazardous or combustible gasses

Several references are available to further define the properties of organic soils such as
(McFarlane, 1969).

2.3.4 Intermediate geomaterials (hard soil/​soft rock)

The most complex materials for tunnel design are the spectrum of material that falls between
soil and rock, defined in many design references as an intermediate geomaterial (IGM).
Materials in this classification behave neither as soils or rock, and proper analysis and
design requires a combination of methods suitable to both. Materials that generally fall into
the category of IGM range from cemented sands and gravels to residual and decomposed
rock strata such as those found in the Piedmont regions of the eastern US, or decomposed
granite. IGM is often layered with both harder and softer material and is seldom found in
large mass formations. Unique features of an IGM include the following:

• Common definition of IGM: N=​>50 bpf, 480 kPa to 1,380 kPa (69 psi to 200 psi)
• Range of geologic materials:
• Cemented sand and gravel (glacial soils, ancient sand dunes)
• Salt cement
• Calcium cement
• Residual decomposed rock
• Mudstone
• Decomposed granite
• Cemented loess

2.3.5 Rock
Rock is typically defined as a geologic material with a material matrix stronger than 1.3
MPa (200 psi) and encompasses a very broad range of both material composition, and
mass structural properties. From a tunnel design perspective, the equipment and meth-
ods of construction are distinctly different than those used for soft ground tunneling. In
addition, the analytical methods used to evaluate and design the tunnel are distinctly
different. For a tunnel in rock, the comprehensive mass properties and fracturing geom-
etry of the geological formation play a more significant role in defining the behavior of
the material, and in determining the loads that affect a tunnel. In contrast to the overall
loading and design properties, cutting tool selection and durability is affected by specific
material composition, and specific material strength. Both the micro-​behavior (discreet
samples) and the macro-​behavior (overall formation) of materials must be understood in
detail to design and build a tunnel. Further details on design properties of rock are pre-
sented in Section 2.6.1
62 Handbook of Precast Segmental Tunnel Lining Systems

Figure 2.1 Rock quality designation (RQD).

Note: After Deere (1967) and example of calculating RQD

Table 2.3 Descriptive classification of rock based on RQD

RQD designation Descriptive classification

<25% Very poor
25%–​5 0% Poor
50%–​7 5% Fair
75%–​9 0% Good
90%–​1 00% Excellent

2.3.5.1 Rock quality designation (RQD)

The rock quality designation index (RQD) was developed by Deere in 1964 (updated
1989) and is commonly used to provide a quantitative estimate of rock mass quality based
on recovered rock fragment size and fracture spacing as recovered from a borehole. RQD
is defined as the percentage of intact core pieces longer than 100 mm (4 inches) within the
total length of core drilled. The procedures for measuring core pieces and calculating RQD
are summarized in Figure 2.1
Descriptive classification of rock quality based on RQD from boreholes is generally sum-
marized as shown in Table 2.3.
The RQD classification system is used extensively in North America and is generally rec-
ognized as a first-​pass indicator of overall rock quality. Care is required in interpretation of
results to understand and account for disturbance and effects of:
 Geotechnical investigations for segmental tunnel lining design 63

• Thinly laminated or non-​isotropic rock properties

• Quality of drilling resulting from equipment (mechanical core breaks), personnel,
orientation, and depth of recovery
• Rock mineralogy and inclusions

2.3.5.2 Rock mass rating (RMR)

The rock mass rating (RMR) system was developed by Bieniawski in 1976 to estimate rock
mass strength and support requirements. The RMR system has been modified over the years
based on review of case studies, and changes made through 1989 to the ratings assigned to
the parameters. RMR is calculated as the sum of the ratings assigned to each of the follow-
ing six parameters:

• Strength of intact rock

• Determined from uniaxial compressive strength tests or point load index tests.
• Rock quality designation (RQD)
• Measured in cores or excavation face.
• Spacing of discontinuities
• Average distance between parallel joints, measured directly in the field or back cal-
culated from oriented core or borehole geophysical logs.
• Condition of discontinuities
• Qualitative description of joint surfaces, including roughness, persistence, aperture,
infilling materials, and weathering.
• Groundwater conditions
• Rating based on measured inflow, ratio of pressure to major principal stress, or
qualitative description.
• Discontinuity orientation
• Adjustment to RMR value based on favorable or unfavorable orientations for tun-
nels, foundations, or slopes.

These parameters are evaluated using geologic maps, rock cores, borehole logging data
and hydraulic conductivity testing or observations of water inflow. Ratings are assigned
based upon the criteria shown in Tables 2.4a to 2.4d. The RMR ranges from 0 to 100, rep­
resenting “very poor” to “very good” rock, respectively. For large projects it may be neces-
sary to divide the tunnel alignment into segments based on structural features or changes
in spacing and character of discontinuities, which permits separate ranges of RMR for
each segment to be defined. Published correlations relate RMR to Q (quality index from
Q-​system of rock classification) such that only one rock mass classification system may be
needed under certain circumstances.

2.3.5.3 Q-​system
The Q-​system of rock classification was developed at the Norwegian Geotechnical Institute
(NGI) (Barton, et al. 1974) in the early 1970’s and updated several times to incorporate
advances in tunneling and rock support technology. The quality index Q is computed as the
product of three quotients as shown below. The first quotient is related to block size, the
second is related to joint friction, and the last is related to active stress around an opening.
 newgenrtpdf
Table 2.4a Rock mass rating system –​ classification parameters and ratings.

64
Summary of rock mass classification parameters for the rock mass rating system (RMR)

Handbook of Precast Segmental Tunnel Lining Systems

Item Parameter Range of values
A Classification parameters and their ratings
1 Point load strength >10 MPa 4 to 10 MPa 2 to 4 MPa 1 to 2 MPa For this low range–​
index uniaxial compressive
Strength of strength test is
intact rock preferred
material
Uniaxial compressive >250 MPa 100 to 250 MPa 50 to 100 MPa 25 to 50 MPa 5-​2 5 1-​5 <1
strength MPa MPa MPa
Rating 15 12 7 4 2 1 0
2 Rock quality designation (RQD) 90% to 100% 75% to 90% 50% to 75% 25% to 50% <25%
Rating 20 17 13 8 3
3 Spacing of discontinuities >2.0 m 0.6 to 2.0 m 200 to 600 mm 60 to 200 mm <60 mm
Rating 20 15 10 8 5
4 Condition of discontinuities Very rough Slightly rough Slightly rough Slickensided Soft gouge > 5 mm
surfaces. Not surfaces. surfaces. surface or, gouge thick or, separation
continuous. No Separation Separation < 5 mm thick or, > 5 mm continuous
separation. <1 mm. Slightly <1 mm. Highly separation 1 to
Unweathered weathered walls weathered walls 5 mm continuous
wall rock
Rating 30 25 20 10 0
5 Inflow per 10 m tunnel None < 10 10 to 25 25 to 125 > 125
length liters/​m in liters/​m in liters/​m in liters/​m in
Joint water 0 0.0 to 0.1 0.1 to 0.2 0.2 to 0.5 > 0.5
Ratio

Ground pressure
water Major principal
stress
General Completely dry Damp Wet Dripping Flowing
conditions
Rating 15 10 7 4 0
 Geotechnical investigations for segmental tunnel lining design 65

Table 2.4b Rock mass rating system –​ rating adjustments for joint orientations.
Summary of joint orientation for the rock mass rating system (RMR)

Very
Item Parameter favorable Favorable Fair Unfavorable Very unfavorable
B Strike and dip orientations of joints
Tunnels 0 -​2 -​5 -​1 0 -​1 2
1 Ratings Foundations 0 -​2 -​7 -​1 5 -​2 5
Slopes 0 -​5 -​2 5 -​5 0 -​6 0

Table 2.4c Rock mass rating system –​ rock classes determined from total ratings.
Summary of rock mass determination for the rock mass rating system (RMR)

Item Parameter Range of values

C Rock mass classes determined from total ratings
1 Ratings 100 <<< 81 80 <<< 61 60 <<< 41 40 <<< 21 < 20
2 Class number I II III IV V
3 Description Very good Good rock Fair rock Poor rock Very poor
rock rock

Table 2.4d Rock mass rating system –​ meaning of rock mass classes.

Summary of rock mass classification for the rock mass rating system (RMR)

Item Parameter Range of values

D Meaning of rock mass classes
1 Class number I II III IV V
2 Average stand-​ 10 years for 6 months for 1 week for 10 hours for 30
up time 15 m span 8 m span 5 m span 2.5 m span minutes for
1 m span
3 Cohesion of > 400 kPa 300 to 400 200 to 300 100 to 200 < 100 kPa
the rock mass kPa kPa kPa
4 Friction angle >45° 35° to 45° 25° to 35° 15° to 25° < 15°
of rock mass

The value of Q can range from 0.001 (exceptionally poor) to 1,000 (exceptionally good).
The Q index is computed using the following equation:

RQD Jr J
Q= * * w (2.1)
Jn J a SRF

where
RQD Rock quality designation, ranging between 0 and 100
Jn Joint set number
Jr Joint roughness number
66 Handbook of Precast Segmental Tunnel Lining Systems

Ja Joint alteration number, qualitative description of joint condition

Jw Joint water reduction factor
SRF Stress reduction factor

While the RQD and joint set number can be determined from drill cores, the remainder
of the parameters are best determined by direct observation in the excavation. In deter-
mining Q, it is advisable to prepare a range of values for individual parameters from many
measurements, and from those compute the minimum, maximum, and mean Q value. More
information and tables of values to be used when selecting parameters to compute Q can be
found in Barton et al. (1974) and NGI (2015).
The Q-​system is commonly used to delineate tunnel alignments into separate support cat-
egories based on rock quality. Correlations have been developed to relate Q to other useful
parameters such as seismic P-​wave velocity and rock mass static modulus of deformation
(Barton, 2002).

2.3.6 Hydrogeology and water pressures for design

One of the most significant factors affecting both the design and construction of a tunnel
lining system is the presence and management of groundwater. Developing a design requires
both a comprehensive understanding of the hydrogeologic regime along the tunnel route,
and a recognition of potential changes that can occur over the design life of a project.
Hydrogeologic design for construction is typically focused on the variation and manifest-
ation of water pressures over a shorter period, specifically the period of construction.
For design, it is common practice to assume that the full hydrostatic pressure from the
surrounding ground will manifest against the tunnel lining system over the life of a tunnel.
Geologic and environmental features to be considered include the following:

• The presence of aquitards within the stratigraphic profile, and the water pressures
associated with each stratigraphic sequence
• Variability of water pressures over time due to seasonal fluctuations, or tidal
fluctuations
• The impact of constructing a tunnel through different geologic strata, and the poten-
tial for the tunnel itself to connect different hydrogeologic strata. (The design may
need to incorporate barriers to flow in this case.)
• The presence of artesian pressure due to regional topographic features
• Predictions for change over time due to influences such as global warming and sea
level rise
• The chemistry of groundwater, and the corrosion potential for various construction
materials

Groundwater within or moving through naturally acidic or alkaline soils, organic soils and
soils contaminated from industrial processes can be corrosive to the reinforcing in concrete
liners. Groundwater rich in sulfates, chlorides, and/​or with high pH can result in degrad-
ation of the concrete of the tunnel liner. As a general practice, chloride and/​or sulfate con-
centrations greater than 500 ppm are corrosive (FHWA, 2017).
For the tunnel construction process, the engineer is faced with all the above concerns, but
limited in time to variability within the period of construction. In addition, the short-​term
 Geotechnical investigations for segmental tunnel lining design 67

manifestation of water flow and ground stability during the excavation process should be
considered. Several texts have been developed (see Bibliography) that address the evaluation
of hydrogeologic properties, estimating groundwater inflow rates, and other specific ana-
lyses that are required for design, and are beyond the scope of this book.

2.3.7 Seismic considerations
Historically, the tunnel industry has considered tunnels to be naturally resistant to seismic
movements, as they did not experience the same high levels of shaking as surface structures.
This perception was supported by the relatively good historic performance of tunnels and
underground structures, especially of tunnels in rock during large earthquakes. Dowding
and Rozen (1978) presented one of the first compilations of damage to rock tunnels due to
earthquake shaking. They collected information on 71 tunnels and compared their response
to ground shaking to estimated peak ground accelerations (PGAs) and peak ground veloci-
ties (PGVs).
Hashash, Hook, Berger, and Yao (Hashash et al. 2001) summarized the state-​of-​the-​
art for seismic analysis and design for underground structures, as part of the activities of
the International Tunneling Association (ITA) Working Group 2. This research placed par-
ticular emphasis on practice within the United States. The report provides a summary of the
historical performance of underground structures under seismic loading and an introduc-
tion to the types and magnitude of movements and shaking that tunnels can be subjected
to under seismic events (see Figure 2.1). Additionally, the report describes approaches to
quantify those effects, aspects of seismic design criteria and traditional (primarily linear
elastic) closed form methods and approaches for analysis.
The following provides general observations regarding the performance of under-
ground structures when subjected to shaking during seismic events [from Hashash
(2001)]:

• Underground structures experienced a lower rate of damage than surface structures,

but the report indicates that some underground structures have experienced signifi-
cant damage (shallow cut-​and-​cover station boxes, sloped portal structures, unlined
rock tunnels and immersed tube structures)
• Cut-​and-​cover tunnels are more vulnerable to seismic events than circular bored
tunnels
• Reported damage decreases with tunnel depth
• Underground structures in soil/​soft ground can be expected to suffer more damage
than tunnels constructed in competent rock
• Lined and grouted tunnels are safer than unlined tunnels in rock.

2.3.7.1 Seismic site characterization

In areas of high seismicity, geotechnical investigations should contain a seismic site charac-
terization that includes a probabilistic seismic hazard assessment (PSHA) based on nearby
fault systems and their historic seismic behavior. The result of this effort is the establishment
of site-​specific time histories of appropriate ground motion parameters or base excitation
that will act as a source of ground movements (accelerations, velocities, displacements or
response spectra) at the site.
68 Handbook of Precast Segmental Tunnel Lining Systems

Figure 2.2 Seismic deformation modes of tunnels.

Sources: Adopted from Hashash et al. (2001) and Owen and Sholl (1981). 

2.3.7.2 Establishment of design earthquake motions

The site-​specific reference ground motions are typically established at a reference ground
horizon (at surface, bedrock depth or a soil depth with suitable stiffness). For tunnels, gen-
erally displacement time histories of base excitations at rock or reference soil depth are
appropriate.

2.3.7.3 Prediction of response to seismic events

Once the site has been characterized and design earthquake levels have been established,
the following seismic analyses are primarily conducted for tunnels lined with pre-​cast
segments:

• Axial compression/​tension and longitudinal bending of the tunnel

• Ovalization of the tunnel

Other considerations for areas with potentially liquefiable soils to include buoyancy ana-
lyses and the effects of shear strength loss (and increases in lateral earth pressures) on the
lining systems.
 Geotechnical investigations for segmental tunnel lining design 69

2.3.7.4 Fault rupture

The effect of the site base excitations needs to be translated to the tunnel horizon to con-
sider any attenuation or spectral response of the ground mass itself. This is typically done
using a method of convolution (rock or reference soil depth up to the tunnel) where the
time histories are applied to the base of a one-​dimensional soil column with the body waves
propagating up vertically through the soil column. This process predicts the dynamic iner-
tial effects and hence the spectral response of the ground mass to the base excitations based
on the variability of dynamic properties of the soil units within the column. Several soft-
ware packages exist for this including SHAKE2000. Typically, a set of displacement profiles
is obtained that reflect the variation with depth of the ground shearing strains. Generally,
the deformations resulting in maximum shearing strains with one to two tunnel diameters
above and below the tunnel are of most interest.
Once the displacement profiles are known, they are applied to a ground structure inter-
action (GSI) model of the tunnel within the soil mass.

2.4 DESKTOP DATA STUDY

The geologic history and site history sections in geotechnical reports are often overlooked
by design engineers, but in fact are very useful in offering a grid for interpretation of geo-
technical data and a basis for predicting the behavior of subsurface conditions. Before devel-
oping the geotechnical input parameters for design for a tunnel lining system, a thorough
review of available reference and historical material is recommended.
Desktop studies serve two general purposes in design:

• To gather available site history and subsurface information at the conceptual stage to
guide later investigations and gain a general understanding of the geology and general
subsurface conditions to be encountered.
• To provide guidance in interpreting between boreholes or other investigative meas-
ures. The general understanding gained from the desktop study guides the geotech-
nical engineer in developing the appropriate ground investigation program required
to establish an engineering design.

A select list of typical reference sources that can often be found online include the
following:

• Regional geology technical references

• State geologic and soil survey reports
• Water well records
• Mining records
• Design reports from other structures
• Lidar and GIS databases

In addition to geologic references, historical site maps should be reviewed (especially in

urban or developed areas) to identify potential man-​made obstructions and other features
that impact both design and construction, such as the following:

• Seawalls and historic shoreline structures

• Tiebacks and earth retention systems from previous construction
70 Handbook of Precast Segmental Tunnel Lining Systems

• Sewers and drainage features

• Buried stream channels
• Building foundations
• Mine shafts

For tunnels and construction projects in urban areas, the Sanborn map collection in the
United States is a useful tool to identify where old buildings and potential buried obstruc-
tion are located. The map series dates from approximately 1867 into the 1970’s and has
been updated at approximately 10-​year intervals for the purpose of assisting fire insur-
ance agents in assessing the fire hazard of structures for underwriting. Sanborn maps can
typically be found in public libraries or from on-​line sources. To the engineer, the Sanborn
map series presents a chronology of structures, streets, and utilities over time, and can
guide geotechnical investigation and planners on where old or buried structures may be
located.

2.5 GROUND INVESTIGATION PROCESS

Subsurface investigations are performed to collect specific information necessary to define
the ground conditions for segmental liner design and mining of the tunnel. Geotechnical
field investigations are often performed early in the design process. However, the needs and
use of that information evolves as the overall design process develops over time. It is recom-
mended that the investigation process be executed in stages, with each stage focusing on
answering questions and providing specific information relevant to the developing design
and construction. The engineer needs to understand the requirements of each stage of design
to establish the overall investigation program.

2.5.1 Planning stages
The questions to be answered during the early planning stages of a project typically involve
an overall understanding of the geotechnical environment. Regional features and general
geologic history are typically developed from the desk-​top study described above. However,
the planning phase of a project warrants the following investigations:

• Topographic information
• Such as LiDAR models and topographic survey to define the geometry of a project
area and establish a reference grid for design development
• Borings and test pits
• Performed in a grid pattern to establish the three-​dimensional conceptual model
of major earth strata and material types including rock, if within the tunnel
horizon.
• While many Codes provide guidance or prescribe minimum boring frequency, the
Engineer must adjust these values based on the complexity and unique nature of the
geology for each project site.
• Water wells
• To identify and monitor the hydrogeologic regime over time.
• Index testing
• To properly classify materials and establish the basic approach to design.
 Geotechnical investigations for segmental tunnel lining design 71

2.5.2 Detailed design stages

After the general geographic and geometric model of a project site is established, the design
process evolves into answering specific details about project components of the tunnel pro-
ject. For the context of this book, this stage of the geotechnical design provides the informa-
tion specifically required to input the analytical models described in Chapter 3. The focus of
the geotechnical investigation at this stage includes the following.

• Supplemental geotechnical borings

• To refine the fidelity of information and address specific issues identified during
the planning stage.
• Borings along a specific final alignment, and at the location of specific structures
such as shafts or foundation elements are developed in this phase of engineering.
• Specific laboratory testing of samples
• To define the range and level of geotechnical input parameters to be used in design,
such as strength, consolidation, elastic, and dynamic parameters, abrasivity, hard-
ness, permeability, fracture orientation, condition, and frequency of rock, etc.
• In situ physical testing
• Such as standard penetration test (SPT), pressuremeter, or over-​core stress testing
to evaluate and refine design parameters from design model input
• Environmental testing
• To identify contaminants and corrosive potential of earth materials and ground-
water, and to evaluate compatibility of construction materials in the environment.

2.5.3 Construction stages
The contractor’s precast segmental tunnel lining design often requires a refinement and
expansion of the geotechnical information and laboratory data that was collected for the
project design. At this stage of a project, the contractor is focused on answering the specific
details of quantifying materials and establishing construction performance predictions for
use in both project planning, execution, and final price development. While specific to each
project, the information required for construction planning often includes the following.

• Supplemental geotechnical borings

• To further refine site-​specific geologic profile gaps to a finer degree than was
required for design
• Specific water well testing
• To address specific hydrogeologic parameters required to size and plan for dewa-
tering and grouting
• Laboratory testing
• To provide specific data for the design and development of cutting tools and
equipment

2.5.4 Performing a subsurface investigation

2.5.4.1 Vertical boreholes
Borings are the most common method for investigating subsurface conditions, deter-
mining the subsurface profile, and direct sampling of subsurface materials. Depending
72 Handbook of Precast Segmental Tunnel Lining Systems

Table 2.5 Recommended Geotechnical Exploration Boring Spacing Guidelines.

Summary of geotechnical bore hole spacing guidelines published by AASHTO in 1988

Recommended bore hole spacing

Item Tunneling ground conditions Metric Imperial
1 Soft ground conditions
• Adverse 15 to 30 m 50 to 100 feet
• Favorable 90 to 150 m 300 to 500 feet
2 Mixed-​f ace conditions
• Adverse 8 to 15 m 25 to 50 feet
• Favorable 15 to 23 m 50 to 75 feet
3 Hard ground conditions
• Adverse 15 to 60 m 50 to 200 feet
• Favorable 150 to 300 m 500 to 1,000 feet

upon the degree of uncertainty, borings made during the preliminary phase are typically
widely spaced and extended to depths two to three diameters below the planned tunnel.
Where surface features preclude the drilling of vertical boreholes along the alignments,
inclined borings can be used to intersect the planned tunnel location at depth. After the
tunnel profile is established, the boring spacing for most projects is reduced, and borings
advanced 1.5 to 2 diameters below the tunnel invert. AASHTO provides guidelines for
boring spacing for bridges and highway structures based on ground conditions as shown
in Table 2.5.
Table 2.6 presents boring spacing recommendation by the U.S. Army Corps of Engineers
for rock tunnels.
Geotechnical borings are performed as part of most subsurface investigations, as they
allow for the direct sampling of subsurface materials. Borings or, more specifically, open
boreholes are needed to enable direct measurement of groundwater levels, sampling of
groundwater for environmental analyses, permeability testing at prescribed depth intervals,
and certain in situ testing methods.
Sampling methods must be selected based on the character of the subsurface mate-
rials and the intended laboratory testing regimen. For granular soil or cohesive soil
where undisturbed samples are not needed, a common sampling method is the Standard
Penetration Test (SPT), for driving a split-​barrel sampler into the soil with repeated
blows of a 63.5 kg (140 lb) hammer free-​falling 760 mm (30 inches) (ASTM D1586).
The SPT N-​value is determined from the number of hammer blows needed to drive the
sampler 300 mm (1 foot) into the soil. The N-​value can be used to estimate the friction
angle and relative density of granular soil, and shear strength of cohesive soil based on
published correlations as described in Section 2.6. Soil captured within the sampler is in
a disturbed state and is typically used for visual classification and certain index tests in
the laboratory.
Where undisturbed samples of cohesive soil are needed for consolidation or triaxial
strength testing, thin-​walled tube samples are obtained by direct push methods using spe-
cially designed tooling (ASTM D1587). Like the SPT method, the thin-​walled sampling
 Geotechnical investigations for segmental tunnel lining design 73

Table 2.6 
G uidelines for Assessing Exploration Needs for Tunnel in Rock (USACE, 1997).
Summary of guidelines for assessing geotechnical exploration needs in rock
conditions as published by U.S. Army Corps of Engineers in 1997

Approximate bore hole

spacing
Item Tunneling ground conditions Metric Imperial
1 Base case condition 150 to 300 m 500 to 1,000 feet
2 Extreme range condition 15 to 1,000 m 50 to 3,300 feet
3 For conditions noted, multiply base case
condition by the following factors
• Simple geology 2.0 to 2.5
• Complex geology 0.3 to 0.5
• Rural 2.0 to 2.5
• Dense urban 0.3 to 0.4
• Deep tunnel Increase bore hole spacing in
proportion to depth of tunnel
• Poor surface access 5.0 to 10+​
• Shafts and portals At least one each location
• Special problems 0.2 to 0.5 locally

device is typically 600 mm (2 feet) in length, thus requiring that the sampling interval be
prescribed in advance based on available subsurface information or determined in the field
based on observation of recovered soil.
In cases where continuous recovery of soil strata is needed, the SPT or thin-​walled sam-
pling methods may prove too cumbersome or time consuming. Sonic drilling may be a
cost-​effective alternative, as this method enables the retrieval of continuous soil or sedi-
ment cores up to 6.0 m (20 feet) or more with each trip out of the borehole. As the method
involves high-​frequency vibration during advance of the sampling barrel, the in situ soil
structure is likely to be disturbed.

2.5.4.2 Horizontal borings

Horizontal borings generally are considered under two circumstances.

• When vertical fracture features are present that cannot be intercepted by vertical bor-
ings, or
• Where the cost of drilling through deep overburden soils above rock can be avoided.

The horizontal boring is made by drilling a horizontal or inclined hole through over-
burden soils using a directional drill to reach the tunnel profile and then drilled horizon-
tally. Advantages of directional drilling is that several vertical boreholes can be replaced by
drilling horizontally, and a continuous record of ground conditions is provided along the
tunnel profile.
74 Handbook of Precast Segmental Tunnel Lining Systems

2.5.4.3 Cone penetration tests (CPT and SCPT)

The cone penetration test (CPT) method involves pushing a 35.7 mm or 43.7 mm (1.40
to 1.72 inch) diameter, 55° to 65° instrumented cone through the underlying subsurface
strata at a rate of 15 to 25 mm/​sec (0.60 to 1.0 inches/​sec) (ASTM D5778). A typical
truck-​mounted CPT rig is shown in Figure 2.3. The instrumented cone and data acquisition
system produce a digital log of tip and side sleeve resistance, induced pore pressure behind
the cone tip, and lithologic interpretation of the subsurface materials. The resulting output
is a continuous record of these parameters and their variation with depth, Figure 2.4. The
electric piezocone CPT data enables calculation of relative density, strength parameters and
equilibrium groundwater pressure. Measurements of the time rate of pore pressure dissipa-
tion can be made at prescribed depths to enable estimates of hydraulic conductivity.
The seismic cone penetration test (SCPT) method includes the capability of performing
downhole seismic velocity testing while also recording the data described above. Seismic
velocity measurements are typically performed at one-​meter intervals and allow for the
determination of soil shear wave velocity and dynamic modulus, useful for developing input
for continuum modeling of soil.
Investigations by the CPT or SCPT methods are typically augmented with traditional
borings to retrieve physical soil samples for visual classification and index testing, and to
“calibrate” the lithologic interpretation of stratigraphy by the CPT data. Where very dense
strata or cobbles and boulders are likely present, it must be anticipated that CPT soundings
may encounter refusal prior to reaching the target depth, and traditional borings may be
needed to characterize deeper strata.

2.5.4.4 Remote sensing and geophysical methods

Traditional site investigation methods such as geotechnical borings and CPT soundings typ-
ically involve the use of truck or track mounted machinery for work on land or barges and

Figure 2.3 Typical cone penetration testing truck rig.

Note: Cone penetration test (CPT) being performed by Craig Test Boring, Inc. in Long Island NY.
 Geotechnical investigations for segmental tunnel lining design 75

Figure 2.4 Typical CPT log.

Note: Summary of data presented for a typical CPT exploration.

service vessels for work over water. Some tunnel alignments, however, may pass beneath
areas that are inaccessible due to land use or local geology, or where access agreements or
temporary easements cannot be obtained. Although direct sampling of subsurface materials
may not be possible under such circumstances, valuable insight into local geomorphology
and surficial hydrology may be gained from high resolution aerial photography and LiDAR
survey from manned or unmanned aircraft, or satellite imagery.
Geophysical methods for identifying variations in subsurface strata or the soil/​bed-
rock interface include seismic refraction surveys, electrical resistivity, ground penetrating
radar (GPR), or the horizontal to vertical spectral ratio (H/​V) method. While these meth-
ods require site access for personnel, each can be performed without the need for large
machinery or vehicles. GPR and H/​V surveys can generally be performed by a single
operator with hand-​portable, battery powered equipment. Careful consideration must
be given to the site constraints and limitations of each method when planning the site
investigation.

2.5.4.5 Rock coring methods

Rock coring is performed to obtain samples of rock for classification and testing, identify
the thickness and sequence of rock strata traversed by the borehole, observe defects or
weaknesses in the rock, study the groundwater regime, and allow access for in situ test-
ing. Cores are typically made with NX-​size diamond or hard alloy steel drill bits, creating
76 Handbook of Precast Segmental Tunnel Lining Systems

rock cores roughly 50 mm (2 inch) diameter from a borehole up to about 100 mm (4 inch)
diameter. Larger diameter tools are available if needed for certain down-​hole testing that
may require larger core holes. Cores are typically retrieved in lengths of 1.5 m or 3 m (4.9
to 9.8 feet).
Core barrels are referred to as “double-​tube” if the core sample is captured in a steel
sleeve that does not rotate with the barrel and bit. Specialty “triple-​tube” core barrels,
containing an inner split barrel may be needed in weaker formations to protect the core
from damage when being extracted from the barrel. Water is pumped down through the
core barrel and circulated up the annular space along the drill string to cool the coring bit
and flush out pulverized rock. In deep boreholes, clay suspension or polymer drilling fluids
may be needed to efficiently evacuate drill cuttings. Proper selection of tooling and a skilled
driller are critical for obtaining high-​quality cores and allowing for accurate measurements
of core recovery and RQD.
Oriented core drilling was developed to enable the measurement of fracture angle and
orientation, which are essential data for evaluation of block or wedge stability in tunnels
and excavations. The process involves etching a line of known orientation along the axis
of the rock core during drilling. Advances in optical (OTV) and acoustic televiewer (ATV)
technology have all but replaced oriented coring in practice, as the automated process and
data acquisition available today eliminate the cumbersome process of manually measuring
the angle and orientation of every fracture in an oriented core. Typical output from OTV/​
ATV borehole logging are shown in Figure 2.5.

2.5.4.6 In situ testing in soil and rock

A suite of in situ testing methods is available for evaluating soil strength, elastic moduli,
density and compressibility, etc. Some in situ tests, such as the pre-​bored pressure meter test
(PMT) or vane shear test (VST) can be conducted in conjunction with traditional boring
and sampling. Other in situ tests, including CPT or SCPT soundings are direct push meth-
ods enabling testing to be conducted without the need for a pre-​drilled borehole. Selection
of appropriate in situ testing methods should be based on the understanding of subsurface
conditions and local experience.
As with CPT soundings, the presence of cobbles and boulders in the subsurface will pre-
clude the use of most direct push testing methods. A list of common in situ soil testing meth-
ods, description of the procedure, and the soil properties determined by each method, are
presented in Table 2.7. Not listed are common tests for assessing soil permeability, which
include pore pressure dissipation testing by electric piezocone and traditional rising, falling
or constant head tests or pumping tests conducted in monitoring wells.
In situ testing in rock is generally performed in boreholes or excavations, though testing
in boreholes is most common during the investigation phase of a tunnel project. Some com-
mon in situ test methods used to determine the rock mass modulus of deformation or shear
strength of discontinuities are listed in Table 2.8. Testing for permeability is typically con­
ducted by isolating segments of a borehole between inflatable packers and injecting water
at known pressure and recording flow rate.
To estimate the in situ state of stress within a rock mass, in situ testing by hydraulic flat
jack (ASTM D4729), over-​coring (ASTM D4623) or hydraulic fracturing stress measure-
ments (Haimson and Cornet, 2003) may be employed. The state of stress is critical informa-
tion for accurate modeling of tunnels that are deep or where it is anticipated that horizontal
principal stress levels may be elevated due to regional tectonics.
 Geotechnical investigations for segmental tunnel lining design 77

Figure 2.5 Typical OTV/​ATV borehole logging output.

Note: Borehole logging shown by Hager-​R ichter Geoscience, Inc.

• The flat jack method involves measuring displacement caused by stress relief as a slot
is cut into a rock surface, then inserting and pressurizing a flat jack until the displace-
ment is recovered.
• The over-​coring method involves measuring deformation at the base of a borehole as
stress is relieved by coring a larger diameter hole around a specialized deformation
gauge.
• The hydraulic fracturing method involves isolating a segment of borehole between
inflatable packers, then increasing fluid pressure in the test interval until a fracture
occurs in the borehole wall.
• The fluid pressure can stabilize, and several cycles of pressurization are conducted.
• Magnitude of the principal stresses are calculated from the recorded pressure
readings.
Table 2.7 In situ testing methods used in soil.

78
Summary of in situ testing methods used in soil as guidelines published by FHWA (2002)

Handbook of Precast Segmental Tunnel Lining Systems

Applicability
Item Testing method Procedure Soil types Soil properties Limitations and remarks
1 Electric cone A cylindrical probe is hydraulically Silts Sands Clays Estimation of soil No soil sample is obtained.
penetrometer pushed vertically through the soil Peat type and detailed
(CPT) measuring the resistance at the stratigraphy The probe may become damaged
conic tip of the probe and along if testing in gravelly soils is
the shaft; measurements typically Sand: φ’, D r, σ ho attempted.
recorded at 2 to 5 cm intervals. Clay: s u, σ p’
Test results not particularly
good for estimating deformation
characteristics.
2 Piezocone Same as CPT; additionally, Silts Sands Clays Same as CPT, with If the filter elements and ports
penetrometer penetration porewater pressures additionally: are not completely saturated, the
(CPTu) are measured using a transducer Peat pore pressure response may be
and porous filter element. Sand: u 0 /​ water misleading.
table elevation
Clay: σ p’, c h, k h, Compression and wear of a
OCR mid-​f ace (u 1) element will affect
readings.
Test results not particularly good
for estimating deformations
characteristics.
3 Seismic CPTu Same as CPTu; additionally, shear Silts Sands Clays Same as CPTu, First interval times should be
(SCPTu) waves generated at the surface are Peat with additionally: used for calculation of shear
recorded by a geophone at 1.0 m wave velocity.
intervals throughout the profile for Vs , G max, E max, ρ tot ,
calculation of shear wave velocity. and e 0 If first cross-​o ver times are used,
the error in shear wave velocity
will increase with depth.
 newgenrtpdf
4 Flat plate A flat plate is hydraulically pushed Silts Sands Clays Estimation of Membranes may become
dilatometer or driven through the soil to a Peat soil type and deformed if over-​inflated.
(DMT) desired depth at approximately 20 stratigraphy Total
to 30 cm intervals. unit weight. Deformed membranes will not
provide accurate readings.
The pressure required to expand a Sand: φ’, E, D r, m v
thin membrane is recorded. Two to Leaks in tubing or connections
three measurements are typically Clay: σ p’ K 0, s u, m v, will lead to high readings.
recorded at each depth. E, c h, k h
Good test results for estimating
deformation characteristics at
small strains.
5 Pre-​b ored A bore hole is drilled, and the Clays Silts E, G, m v, s u Preparation of the bore hole

Geotechnical investigations for segmental tunnel lining design

pressuremeter bottom is carefully prepared for and Peat most important step to obtain
(PMT) insertion of the equipment. good results.
Marginal response
The pressure required to expand in some sands and Good test for calculation
the cylindrical membrane to a gravels of lateral deformation
certain or radial strain is recorded. characteristics.
6 Full A cylindrical probe with a Clays Silts E, G, m v, s u Disturbance during advancement
displacement pressuremeter attached behind of the probe will lead to stiffer
pressuremeter a conical is hydraulically pushed Peat in sands initial modulus and mask lift-​o ff
(PMT) through soil and paused a select pressure (p 0)
intervals for testing.
Good test for calculation
The pressure required to expand the of lateral deformation
cylindrical membrane to a certain characteristics.
volume or radial strain is recorded.
7 Vane shear A four-​b lade vane is slowly rotated Clays Some silts s u, S t , σ p’ Disturbance may occur in soft
test while the torque required to rotate sensitive clays, reducing measured
the vane is recorded for calculation and peats if strength.
(VST) of peak undrained shear strength undrained Partial drainages may occur in fissured
The vane is rapidly rotated for conditions can be clays and silty materials, leading to
10 turns and the torque required assumed errors in calculation of strength.
to fail the soil is recorded for Not for use in
calculation of remolded undrained Vane diameter and torque wrench
granular soils capacity need to be properly sized
shear strength.
for adequate measurements in

79
various clay deposits.
 newgenrtpdf
Table 2.7 (Continued)

80
Table and Symbol Legend

Handbook of Precast Segmental Tunnel Lining Systems

Φ’ Effective stress friction angle G max Small-strain shear modulus

Dr Relative density G Shear modulus

σ ho’ In-​s ite horizontal effective stress E max Small strain Young’s modulus

su Undrained shear strength E Young’s modulus

σ p’ Preconsolidation stress ρ tot Total density

ch Horizontal coefficient of consolidation e0 In situ void ratio

kh Horizontal hydraulic conductivity mv Volumetric compressibility coefficient

OCR Over-consolidation ratio K0 Coefficient of at-rest earth pressure

Vs Shear wave velocity St Sensitivity

 Geotechnical investigations for segmental tunnel lining design 81

Table 2.8 In situ testing methods used in rock.

Summary of in situ testing methods and guidelines used in rock conditions as
published by FHWA in (2002)

Item Testing method Procedure Rock properties Limitations and remarks

1 Bore hole The dilatometer Fractured rock Poisson’s ratio of rock
dilatometer is lowered to the mass modulus must be assumed.
test test elevation and
the flexible rubber Test only affects a small
membrane is expanded area of the rock mass but
exerting a uniform several tests over the
pressure on the depth of influence can be
sidewall of the performed.
bore hole
2 Bore hole Jacks exert a uni-​ Modulus of Measured modulus value
jack test directional pressure fractured rock must be corrected to
on the walls of a bore mass modulus account for stiffness of
hole by means of two steel platens.
opposed curved steel
platens Test method can be used
to provide an estimate of
anisotropy.
3 Plate load Load is applied to a Modulus of Loaded area is limited so
test steel plate or concrete fractured rock may not be effectively
foundation using a mass modulus testing rock mass if joints
system of hydraulic are widely spaced.
jacks and a reaction
frame anchored to the Modulus values corrected
foundation rock. for steel geometry, effect
of rock breakage, rock
anisotropy and steel plate
modulus.
4 In situ direct Testing is typically Peak and Need to isolate a block of
shear test performed in an adit residual shear rock above a discontinuity
where reaction for the strength of surface without disturbing
shear load is provided by discontinuity the infilling.
an adit side wall. Normal or
load applied via jack discontinuity
system that uses the infilling.
adit roof for reaction.

Descriptions of the suggested use and limitations of each method are provided in the
references.

2.6 GEOTECHNICAL PARAMETERS USED FOR SEGMENTAL TUNNEL

LINING DESIGN
Development of design subsurface profiles and geotechnical design parameters for soil and
rock are fundamental components in the design of segmental tunnel lining. The required
geotechnical design parameters depend on the selected analysis approach and assumptions.
Refer to Chapter 3 for the development of design models for specific loading criteria. In
addition to a comprehensive subsurface investigation and laboratory testing program, the
interpretation and application design parameters should incorporate the lessons learned
82 Handbook of Precast Segmental Tunnel Lining Systems

from local experience. Consultation and review from local qualified practitioners are
recommended.

2.6.1 Shear strength parameters

Establishing appropriate strength parameters for design of segmental liners in soil and rock
is complex.
Soil shear strength as a material property is manifest by the cohesion between particles,
and the friction and/​or interlocking between the particles that provides resistance to soil
particle movement. For rock, the bonding of crystals or the cementing of particles provide
the material strength. The orientation of fractures and bedding planes, and the roughness
between discontinuities generally controls a design.
The Unified Soil Classification System (USCS) (ASTM D2487) is used to classify soils
into two broad soil groups, cohesive and non-​cohesive. However, the occurrence of cemen-
tation between particles, and water tensile forces between very fine particles can impart the
effects of cohesion into non-​cohesive soils. It is not recommended to assign strength values
to soil based on general material classification alone. Site-​specific investigations with field
and laboratory testing is recommended to develop parameters for design.
Commonly used field evaluation tools include standard penetration tests (SPTs), cone
penetration tests (CPTs), geophysical tests, field pressuremeter, dilatometer (DMT) and vane
shear tests. Field testing of boring samples using a hand-​held pocket penetrometer or tor-
vane can provide a general indication of the soils’ shear strength but should not be relied
on as definitive design values. Unconfined compression, triaxial and direct shear tests are
commonly used for shear strength and stiffness measurement using samples recovered from
borings. Consolidation tests are performed in fine-​grained soils to evaluate stress history
and time rates for settlement.

2.6.1.1 Cohesive soils

Parameters for cohesive soils or granular soils with plasticity can be approximated from
published correlations using results from the subsurface investigation program, such as the
standard penetration test. Table 2.9 provides a typical correlation between SPT and uncon­
fined compressive strength, and the nomenclature commonly used to describe the strength
properties of soils. Caution is advised in relying on such correlations as many local and site-​
specific factors will affect the results.
Also, as a caution, the sampling procedures and equipment used will influence the results
of SPT. As an example, many correlations available in the literature were developed from
data collected before the development and common use of automatic trip hammers, and the
energy efficiency was much lower than data collected with modern equipment. The results
can be that modern collected data may present lower blow count numbers than would have
been recorded from equipment at the time of the correlation. Therefore, it is important to
adjust the data to the energy input levels used to develop the original correlation.
The shear strength of cohesive soils used for design is typically measured from direct
testing or estimated from laboratory tests. The unconfined compressive strength, (qu), can
be determined from the unconfined compressive strength (UC) tests (ASTM D2166). These
tests are performed undrained, that is the excess pore pressure generated during shearing is
not allowed to dissipate, with no confining pressure and provide an approximate cohesive
strength in terms of total stresses. The undrained shear strength (cu) is computed as 0.50 qu.
 Geotechnical investigations for segmental tunnel lining design 83

Table 2.9 Guide for consistency of fine-​g rained soils (DM 7.01, 1986).
Nomenclature for describing consistency of fine-​g rained soils based on correlation to
the standard penetration test (SPT)_​_​_​_​_​_​_​_​_​_​_​_​_​_​_​_​

Estimated consistency
Estimated range of
Standard unconfined compressive
penetration strength
test (SPT) Estimated consistency based on standard
Item (blows/​f oot) penetration testing (SPT) Metric Imperial
1 <2 Very soft <0.024 <0.25
(Extruded between fingers when squeezed) MPa tons/​f t 2
2 2 to 4 Soft 0.024 to 0.25 to 0.50
(Molded by light finger pressure) 0.048 MPa tons/​f t 2
3 4 to 8 Medium 0.048 to 0.50 to 1.00
(Molded by strong finger pressure) 0.096 MPa tons/​f t 2
4 8 to 15 Stiff 0.096 to 1.0 to 2.00
(Readily indented by thumb but penetrated 0.192 MPa tons/​f t 2
with great effort)
5 15 to 30 Very stiff 0.192 to 2.00 to 4.00
(Readily indented by thumbnail) 0.383 MPa tons/​f t 2
6 >30 Hard >0.383 >4.00
(Indented with difficulty by thumbnail) MPa tons/​f t 2

Triaxial compression tests are performed to develop undrained and drained strength
parameters and stress–​strain relationships. The unconsolidated undrained (UU) triaxial
compression test (ASTM D2850) is performed with a confining pressure with no drainage
of the sample permitted. This test measures the undrained shear strength, cu, of the soil
sample. The friction angle, ø, is assumed as zero since there is no drainage during the test.
For the consolidated undrained (CU) triaxial compression test (ASTM D4767), the sam-
ples are first consolidated and stabilized at a defined confining pressure and then sheared
in compression without drainage. One of the reasons that the soil sample is consolidated is
to compensate for the sample disturbance that occurs after removal of in situ overburden
pressure during sampling. Measurements during testing of axial load, axial deformation,
and pore water pressure permit determination of both total and effective stress paths in the
sample, as well as total and effective shear strength parameters, cohesion, friction angle and
modulus of elasticity, E.
The consolidated drained (CD) triaxial compression test (ASTM D7181) is consolidated
and sheared at a specific strain rate to permit drainage without an increase in pore pressure
during shearing. Effective stresses and drained soil parameters, c’ and ø’, and modulus of
elasticity, E’ are determined from this test.

2.6.1.2 Non-​cohesive soils

In common practice, the friction angle of non-​cohesive or cohesionless soils is typically esti-
mated from correlations with the standard penetration test (SPT) “N” value or from cone
penetration tests (CPTs), since it is extremely difficult to obtain relatively undisturbed in
situ samples for laboratory triaxial testing. For most projects, the friction angles are often
84 Handbook of Precast Segmental Tunnel Lining Systems

estimated based on local experience or empirical correlation with SPT blow counts and/​or
cone penetration resistance.
Presence of larger particles such as cobbles and boulders in a non-​cohesive soil typically
increase the shearing capacity of a soil and therefore the design friction angle. Quantifying
the proportion of the subsurface strata comprising cobbles and boulders may not be pos-
sible, and sampling cobbles or boulders in the subsurface is difficult and time-​consuming.
Additionally, cobbles and boulders may result in refusal of CPT tooling or lead to erro-
neously high SPT N-​values and can lead to an over-​estimate of design values, as well as
resulting in complications during construction. Sonic drilling using 20 mm to 25 mm (0.75
to 1.0 inches) diameter casing has been used successfully to penetrate and sample soils with
appreciable cobbles and boulders.

2.6.1.3 Rock strength

Rock strength for design purposes requires an evaluation of both the intact material strength
of rock, and an assessment of the macro fracturing and bedding characteristics of the larger
rock mass. Table 2.10 presents a generalization representing the range of intact rock strength
based on geologic origin, and a generalized estimate of the ratio of unconfined compressive

Table 2.10 Representative strengths for different types of intact rock.

Summary of strengths (q u) or unconfined compressive strengths for different
types of intact rock per Goodman (1989) and the ratio of unconfined compressive
strength to tensile strength (q t)

Unconfined compressives
Strength (q u)
Item Intact rock material Metric Imperial Ratio q u/​q t
1 Berea sandstone 73.8 MPa 10,700 psi 63.0
2 Navajo sandstone 213.9 MPa 31,030 psi 26.3
3 Tensleep sandstone 72.4 MPa 10,500 psi
4 Hackensack siltstone 122.7 MPa 17,800 psi 41.5
5 Monticello Dam sandstone (greywacke) 79.3 MPa 11,500 psi
6 Solenhofen limestone 244.8 MPa 35,500 psi 61.3
8 Bedford limestone 51.0 MPa 7,400 psi 32.3
9 Tavernalle limestone 97.9 1MPa 14,200 psi 25.0
10 Oneota dolomite 90.3 MPa 13,100 psi 29.8
11 Flaming Gorge shale 35.2 MPa 5,100 psi 167.6
12 Micaceous shale 75.2 MPa 10,900 psi 36.3
13 Dworshak Dam gneiss (45° foliation) 162.0 MPa 23,500 psi 23.5
14 Quartz mica schist (ꓕ schistocity) 55.2 MPa 8,000 psi 100.4
15 Baraboo quartzite 319.9MPa 46,400 psi 29.1
16 Taconic marble 61.4 MPa 8,900 psi 53.0
17 Cherokee marble 66.9 MPa 9,700 psi 37.4
 Geotechnical investigations for segmental tunnel lining design 85

Table 2.10 (Continued)

Unconfined compressives
Strength (q u)
Item Intact rock material Metric Imperial Ratio q u/​q t
18 Nevada Test Site granite 141.3 MPa 20,500 psi 12.1
19 Pikes Peak granite 216.2 MPa 32,800 psi 19.0
20 Cedar City tonalite 101.4 MPa 14,700 psi 15.9
21 Palisades diabase 241.0 MPa 34,950 psi 21.1
22 Nevada Test Site basalt 148.2 MPa 21,500 psi 11.2
23 John Day basalt 355.1 MPa 51,500 psi 24.5
24 Nevada Test tuff 11.4 MPa 1,650 psi 10.0
Table Legend and Notes:
q
u Strength or unconfined compressive strength of the rock
q
t Tensile strength of the rock

strength to tensile strength for a variety of rock types. While these generalizations are useful
to gain an understanding of the range of strength one can expect based on geology, it is not
recommended to use such generalizations as a basis of design. Site-​specific laboratory test-
ing of rock is recommended, combined with an evaluation of local and regional fracturing
of the rock mass.
Intact rock strength is typically determined by laboratory uniaxial or triaxial testing of
core specimens recovered during subsurface investigations. Due to the variability of rock
strength and the presence or absence of core imperfections, a series of tests should be per-
formed to determine the typical range of strength for each rock type encountered.
Intact rock strength can be highly anisotropic in metamorphic rock with well-​developed
foliation or cleavage, and in sedimentary rock exhibiting prominent bedding. Point load
strength index (PLSI) testing (ASTM D5731) is a simple and inexpensive way to test sam-
ples in the field to estimate compressive strength and the anisotropy index, or ratio of
strength perpendicular to or parallel to natural planes of weakness. It should be noted,
however that the Point Load Strength index should be used with caution as results for rock
with compressive strength below about 25 MPa (3,625 psi) may be unreliable.

2.6.1.3.1 ROCK MASS PARAMETERS

Hoek and Brown developed a method to estimate the strength and deformation character-
istics of jointed rock masses based upon the geological strength index (GSI), a rock mass
rating system developed by Hoek (1994), which considers the lithology, structure and sur-
face conditions of the discontinuities within the rock mass as shown in Figure 2.6. The GSI
is used to reduce the laboratory determined values of uniaxial compressive strength σci, and
material constant mi, related to the frictional properties of the rock, to appropriate values
for the in situ rock mass. The Hoek–​Brown failure criterion is one of the most common
models for estimating the shear strength of rock with closely spaced discontinuities. A more
in-​depth discussion of the Hoek–​Brown failure criterion and GSI, and its utilization, is
found in (Hoek–​Brown, 2018)
86 Handbook of Precast Segmental Tunnel Lining Systems

Figure 2.6 Geotechnical strength index (GSI) chart.

Source: From Hoek and Brown, 2018. 

The use of GSI assumes an isotropic rock mass and should only be applied where discon-
tinuities are closely spaced compared to tunnel diameter. The rock can then be considered
as a homogeneous and isotropic mass of interlocking blocks and wedges.
Where the rock mass consists of strong, blocky rock with large block size relative to the
tunnel diameter, the rock mass behavior will be controlled by the displacement of blocks
or wedges along discontinuities that may be an order of magnitude weaker than the parent
 Geotechnical investigations for segmental tunnel lining design 87

material. Under such conditions, the blocks and wedges should be treated individually.
Computer programs such as “UnWedge” utilize block theory to enable 3D analyses of
blocky rock stability. Careful consideration must be given to the selection of discontinuity
strength parameters in such analyses. Barton (2013) has contributed a great deal of research
into shear strength of rock fractures, the non-​linear shear behavior of rock joints, and
displacement-​dependent joint roughness. The Barton–​Bandis joint model can be applied in
“UnWedge” and other jointed rock modeling software where appropriate.

2.6.1.4 Elastic modulus

Development of elastic parameters for design should consider the specific design applica-
tion. In-​tact material modulus is often significantly higher than the appropriate modulus
to be used for a fractured rock mass. Further guidance in development and interpretation
of modulus values are presented in Hoek and Brown (2018). Elastic modulus for design
input can be approximated using different methods in both the laboratory and in the field.
Laboratory tests, including compression and consolidation tests, can be used to characterize
the stress–​strain behavior of both soil and rock. Laboratory and field measurements of shear
wave velocity have been used extensively to measure the elastic modulus at the small strain
level, Emax.

2.6.2 Design using computer programs

Computer programs are commonly used to perform numerical analysis for design of seg-
mental tunnel linings. A number of these software programs, ABAQUS, FLAC, PLAXIS,
RS2 and UDEC to name a few, have implemented constitutive soil models with varying
degrees of complexity and limitations. Numerical methods allow engineers to evaluate
many different considerations, such as mixed subsurface conditions, construction staging,
unique excavation geometry, artesian groundwater and time-​dependent effects. In add-
ition, a numerical analysis that accounts for soil–​structure interaction provides results for
stresses in the structural elements that can be used to model the segmental tunnel lining.
Commercially available computer programs typically provide user friendly interface, mak-
ing numerical analysis an attractive design tool.
However, performing numerical analysis requires considerable expertise and experience
of the design engineer, particularly, the definition, and limitations, of the constitutive mod-
els that define the stress–​strain behavior of soils and rock, as well as the appropriate field
investigations and laboratory tests required to develop reliable design parameters.
When performing a numerical analysis for segmental tunnel lining design, it is advised
that a conventional analysis also be performed, allowing a “sanity check” on the numerical
results.

2.6.2.1 Soil models

Studies have recognized that the stress–​strain behavior of both sand and clay is highly
non-​linear with soil stiffness reducing significantly with increasing strain. Conventional
field and laboratory testing can be limited in its capability to measure an appropriate soil
modulus corresponding to the strain level typical for tunnels. The small-​strain modulus,
Emax, typically derived from the shear wave velocity measured in the field or in the labora-
tory, provides a useful reference modulus value, as it is the maximum stiffness that a
material can exhibit.
88 Handbook of Precast Segmental Tunnel Lining Systems

2.6.2.2 Rock models

The rock mass static modulus of deformation, Em for numerical modeling is typically esti-
mated by empirical correlations with measured properties such as intact rock modulus or
GSI (Figure 2.7, Table 2.11), or determined from in situ testing such as the borehole dilato­
meter and borehole jack (ASTM D4971).

2.6.2.3 Spring constants

Structural analysis of segmental tunnel lining often uses spring constants to represent inter-
action between lining and the ground. Springs can be placed in the radial and tangential
directions at each node. The magnitude of the spring constant at each node is calculated

Figure 2.7 Simplified relation between E m and GSI.

Source: After Hoek and Diederichs (2006). 

Table 2.11 Ratio of rock mass modulus to intact rock modulus.

Range of ratio of rock mass modulus for numerical modeling to intact
rock modulus per O’Neill et al. (1996)

Ratio of rock mass modulus to Intact

rock modulus (EM /​ ER)
Rock quality
Item designation (RQD) (percent) Closed joints Open joints
1 100 1.00 0.60
2 70 0.70 0.10
3 50 0.15 0.10
4 20 0.05 0.05
 Geotechnical investigations for segmental tunnel lining design 89

from the modulus of subgrade reaction of the surrounding ground multiplied by the tribu-
tary length of lining on each side of the node (FHWA, 2009).
The modulus of subgrade reaction is an empirical parameter that is dependent on the type
of soil or rock, the model selected for the analysis, as well as the tunnel diameter. While
there are many publications that document limitations of the use of the modulus of sub-
grade reaction in tunnel liner design, it remains common in many engineering offices and
agencies, and has been used in many successful projects. It is crucial that the modulus of
subgrade reaction (k) is selected by a qualified geotechnical engineer.

2.6.2.4 Non-​linear elastic modulus constitutive models

Selecting the appropriate constitutive model for numerical analysis is essential for evalu-
ating the ground behavior, as the constitutive model selected can significantly influence the
predicted stresses in the precast segmental tunnel lining, and therefore, affecting its struc-
tural design. Research, such as Meguid et al. (2002) and Law and Moore (2007), has shown
that the linear elastic and the linear elastic-​perfectly plastic constitutive models can provide
realistic results for buried structure behavior in soil, provided appropriate secant modulus
(selected based on the corresponding mobilized strain) is used in the analysis.
However, realistic soil behavior is generally non-​linear and stress dependent. The stress
history of soil also affects its behavior, resulting in different stiffness during virgin compres-
sion (initial loading) and recompression (unloading–​loading). These non-​linear behaviors
are not definitively represented in the linear elastic or linear elastic-​perfectly plastic models
and has led to the development of more advanced soil constitutive models, which permit
the analysis to consider effects such as strain dependency, strain hardening and softening,
stress history (preconsolidation effects), critical state, creep, swelling and shrinkage. The
more complex non-​linear constitutive models would require additional input parameters.

2.6.2.5 Soil hardening models

The soil hardening model typically features stress dependent stiffness, distinction between
virgin compression and recompression, and memory of stress history. Many soil hardening
models implemented into computer programs consider both shear and compression harden-
ing. The stress-​dependent stiffness is often expressed using the “power law” (for example,
Janbu 1963 and Duncan and Chang 1970):

m
 σ′ 
E = Eref  (2.2)
 pref 

where
E =​stress dependent modulus
Eref =​reference soil modulus corresponding to the reference confining pressure pref
m =​power for stress-​level dependency of stiffness
σ’ =​effective stress

For sands, development of soil hardening model parameters typically relies on correla-
tions from field data, particularly CPT and DMT, full scale testing, and local experience.
Laboratory results, including triaxial and consolidation tests, are typically used to derive
90 Handbook of Precast Segmental Tunnel Lining Systems

parameters for clays based on calibration. Typical reference modulus for sand typically
ranges between 10 MPa and 50 MPa (1,450 to 7,250 psi) at reference pressure of 100 kPa
(14.5 psi), depending on the relative density. For soft clays, the reference modulus can be in
the 1 MPa to 5 MPa (145 to 725 psi) range (at pref of 100 kPa) (14.5 psi). The value “m”
can be about 0.5 to 0.6 for sand and silt, and about 1.0 for soft clay.
Many available soil hardening models also consider the high soil stiffness at the small
strain level (typically defined as lower than 10-​5). Research has shown that constitutive
models considering the small strain stiffness can provide a more realistic ground settle-
ment trough (e.g. Brinkgreve et al. 2006). The input parameters for small strain stiffness
are typically developed based on measuring the shear wave velocity in the laboratory or
in the field.

2.6.2.6 Creep models

For soft clays that exhibit significant creep, some hardening soil models can also consider
the time-​dependent secondary compression. Typically, the creep constitutive models can
account for three-​dimensional creep. The development of input parameters for creep is
generally derived from consolidation test data. Refer to Chapter 3 for the development of
design models.

2.7 HOW TO REPORT THE DATA –​ GEOTECHNICAL REPORTS

The reporting of geotechnical data needs to align with its intended use, and the intended
use and limitations must be clearly defined. In common practice, both design engineers
and builders rely on data collected by, and interpreted by, a third-​party Geotechnical
Engineer. The Geotechnical Engineer in-​turn relies on information developed by a subcon-
tracted driller and in many cases an independent laboratory providing analytical results. It
is unlikely the Geotechnical Engineer fully comprehends the full scope of the end use for the
data. These results require completely different interpretations based on whether the end-​
user is a designer or a builder. Therefore, during the life of a project, the same data will be
analyzed from a completely different perspective. Consequently, geotechnical information
is often presented as a series of focused reports, each crafted specifically for one intended
purpose. For each report, it is recommended that the intended purpose is reflected in the
title using the below nomenclature, and that the limitations on the interpretations be clearly
presented.

2.7.1 Geotechnical reporting –​definitions and content

Common geotechnical report types typically prepared during the life of a tunnel design pro-
ject include the following.

2.7.1.1 Design memoranda (DM)

A document of specific focus written by a geotechnical engineer to a design professional for
a specific evaluation is called a design memorandum (DM). The geotechnical engineer will
present both data and interpretation specifically formulated to guide the design professional
performing a specific task. For example, a design memorandum describing the input param-
eters for stiffness evaluation will recommend spring constants or stiffness modulus values to
 Geotechnical investigations for segmental tunnel lining design 91

be used in design. The parameters are typically custom tailored for the analytical methods
employed by the designer.
The specific use and limitations of these reports should be clearly stated. The DMs should
be used for its intended purpose and the recommendations and data in it not extrapolated
for other uses or used by other parties without consultation with the geotechnical engineer.
Typical examples would include a memorandum describing the geotechnical design param-
eters for a structural analysis, or a pre-​bid interpretation of geotechnical factors affecting a
construction bid.

Written by Primary audience

Geotechnical Engineer Design Professionals, Construction Estimators

2.7.1.2 Geotechnical data report (GDR)

The geotechnical data report (GDR) is a comprehensive and general document incorpor-
ating the raw data from a project-​specific site investigation for the use and interpretation by
others. A GDR must clearly define the methods and standards used for data collection and
testing so that other parties can clearly interpret and understand the data.
GDRs are commonly included as a contract document for solicitation of bids. It is also
common in a GDR to summarize factual project history and performance from other nearby
or relevant projects to help guide the interpretation of data. The GDR must not provide
specific recommendations or limitations on the use of the data, but should instead focus on
factual accuracy, and a careful description of the sources used to collect data, so its accuracy
can be adequately understood.

Written by Primary audience

Geotechnical Engineer Contractor, Specialty Design subconsultants

2.7.1.3 Geotechnical interpretive report (GIR)

Often referred to as a design summary report, the geotechnical interpretive report (GIR)
summarizes and consolidates the information, geotechnical interpretation, and basis of
design used by the designer for the project. The contents and interpretations presented in
a GIR are specific to the design, and not necessarily comprehensive enough to be used for
construction. It is often made available to a contractor for bidding to understand the intent
of a design but is not intended to be relied upon as a contractual document in the develop-
ment of specific means and methods by a contractor.

Written by Primary audience

Geotechnical Engineer, Design Professionals Owner, Project Design Documentation

2.7.1.4 Geotechnical baseline report (GBR)

A geotechnical baseline report (GBR) is a specific contract document for a project that
defines a common interpretation of geotechnical information for the purpose of risk sharing
between the contracting parties for bidding by a contractor. The GBR is not intended to
justify the design, or be limited to the specific factual data only, but is intended to represent
92 Handbook of Precast Segmental Tunnel Lining Systems

a common interpretation of the materials and geotechnical behavior for the specific purpose
and use of a contractor developing his bid price, and in defining the criteria for additional
contractor compensation and/​or schedule extension should conditions be encountered that
are different than presented in the GBR at the time of the contractor’s bid.
The use of GBR’s on tunnel and underground construction was conceived and developed
during the implementation of the Washington DC Metro system in the 1970’s and has
since been refined and expanded into common use on underground projects. Guidelines
for the development and use of GBR’s has been summarized in an ASCE publication
titled: “Geotechnical Baseline Reports for Underground Construction–​ Guidelines and
Practices” [18].

Written by Primary audience

Geotechnical Engineer, Design Professionals, Contractor, Contract Documents for bid
Contract Manager development and change management

BIBLIOGRAPHY
For preparation of this chapter, the following references were used, which are categorized into two
categories of authority published documents and authored documents.

AMERICAN SOCIETY OF CIVIL ENGINEERS

American Society of Civil Engineers (ASCE) (1997). Geotechnical Baseline Reports for Underground
Construction –​Guidelines and Practices, New York.

FEDERAL HIGHWAY ADMINISTRATION (FHWA)

Federal Highway Administration (2002). Geotechnical Engineering Circular No. 5–​Evaluation of Soil
and Rock Properties, FHWA-​IF-​02-​034, Washington, DC.
Federal Highway Administration (2005). Road Tunnel Design Guidelines, FHWA-​IF-​05-​023,
Washington, DC.
Federal Highway Administration (2009). Technical Manual for Design and Construction of Road
Tunnels –​Civil Elements, FHWA-​NHI-​10-​034, Washington, DC.
Federal Highway Administration (2017). Geotechnical Engineering Circular No.5–​Geotechnical Site
Characterization, FHWA-​NHI-​16-​072, Washington, DC.

U.S. ARMY CORPS OF ENGINEERS (USACE)

U.S. Army Corps of Engineers (1997). Engineering and Design, Tunnels and Shafts in Rock, EM 1 1
10-​2-​2901, May.

U.S. NATIONAL COMMITTEE ON TUNNELING TECHNOLOGY

(USNTT)
U.S. National Committee on Tunneling Technology (1984). Geotechnical Investigations for
Underground Projects, National Academy Press, Washington, DC (2 volumes).

PUBLISHED CODES AND STANDARDS

For preparation of this chapter, the following references were used, which are categorized into two
categories of authority published documents and authored documents.
 Geotechnical investigations for segmental tunnel lining design 93

AMERICAN SOCIETY FOR TESTING MATERIALS

ASTM D1586/​D1586M-​18e1 (2022). Standard Test Method for Standard Penetration Test (SPT) and
Split-​Barrel Sampling of Soils, ASTM International, West Conshohocken, PA.
ASTM D2216-​19 (2019). Standard Test Methods for Laboratory Determination of Water (Moisture)
Content of Soil and Rock by Mass, ASTM International, West Conshohocken, PA.
ASTM D2434-​22 (2019). Standard Test Method for Hydraulic Conductivity of Coarse Grained Soils
ASTM International, West Conshohocken, PA.
ASTM D2487-​17e1 (2020). Standard Classification of Soils for Engineering Purposes (Unified Soil
Classification System), ASTM International, West Conshohocken, PA.
ASTM D420-​ 18 (2018). Standard Guide for Site Characterization for Engineering Design and
Construction Purposes, ASTM International, West Conshohocken, PA.
ASTM D4318-​17e1 (2018). Standard Test Methods for Liquid Limit, Plastic Limit, and Plasticity
Index of Soils, ASTM International, West Conshohocken, PA.
ASTM D4394-​ 17 (2017). Standard Test Method for Determining In-​ Situ Modulus of
Deformation of Rock Mass Using Rigid Plate Loading Method, ASTM International, West
Conshohocken, PA.
ASTM D4395-​17 (2017). Standard Test Method for Determining In-​Situ Modulus of Deformation
of Rock Mass Using Flexible Plate Loading Method, ASTM International, West
Conshohocken, PA.
ASTM D4554-​12 (withdrawn 2022), Standard Test Method for In Situ Determination of Direct Shear
Strength of Rock Discontinuities, ASTM International, West Conshohocken, PA, 2021
ASTM D4623-​16 (2019). Standard Test Method for Determination of In Situ Stress in Rock Mass by
Over-​coring Method–​Three Component Borehole Deformation Gauge, ASTM International,
West Conshohocken, PA.
ASTM D4630-​ 19 (2019). Standard Test Method for Determining Transmissivity and Storage
Coefficient of Low-​Permeability Rocks by In Situ Measurements Using the Constant Head
Injection Test, ASTM International, West Conshohocken, PA.
ASTM D4729-​19 (2019). Standard Test Method for In Situ Stress and Modulus of Deformation Using
the Flat Jack Method, ASTM International, West Conshohocken, PA.
ASTM D5084-​16a (2016). Standard Test Methods for Measurement of Hydraulic Conductivity of
Saturated Porous Materials Using a Flexible Wall Permeameter, ASTM International, West
Conshohocken, PA.
ASTM D5778-​20 (2016). Standard Test Methods Electronic Friction Cone and Piezocone Penetration
Testing of Soils, ASTM International, West Conshohocken, PA.
ASTM D5856-​15 (2016). Standard Test Method for Measurement of Hydraulic Conductivity of
Porous Material Using a Rigid-​Wall, Compaction-​Mold Permeameter, ASTM International,
West Conshohocken, PA.
ASTM D6913 /​D6913M-​17 (2021). Standard Test Methods for Particle-​Size Distribution (gradation)
of Soils Using Sieve Analysis, ASTM International, West Conshohocken, PA.
ASTM D6914 /​D6914M-​16 (2021). Standard Practice for Sonic Drilling for Site Characterization
and the Installation of Subsurface Monitoring Devices, ASTM International, West
Conshohocken, PA.

AMERICAN ASSOCIATION OF STATE HIGHWAY AND

TRANSPORTATION OFFICIALS (AAHSTO)
AASHTO (1988). Manual on Subsurface Investigations. MSI 88. American Association of State
Highway and Transportation Officials. Washington, DC.
AASHTO (2012). Classification of Soils and Soil-Aggregate Mixtures for Highway Construction
Purposes. M145-91. American Association of State Highway and Transportation Officials.
Washington, DC.
94 Handbook of Precast Segmental Tunnel Lining Systems

BRITISH STANDARDS INSTITUTE (BSI)

British Standards (1990). Methods of Test for Soils for Civil Engineering Purposes. BS 1377. British
Standards Institution. Lonon.

REFERENCE PAPERS
Atkinson, J.H., Sallfors, G. (1991). Experimental Determination of Soil Properties. Proceedings of the
10th ECSMFE, Florence, Volume 3, pp. 915–​956.
Barton, N. (1988). Rock Mass Classification and Tunnel Reinforcement Selection Using the Q-​System,
Rock Classification Systems for Engineering Purposes, ASTM STP 984, Louis Kirkaldie (Ed.),
American Society for Testing and Materials, Philadelphia, PA, pp. 59–​88.
Barton N. (1991), Geotechnical Design, World Tunneling, November 1991, pp. 410–​416.
Barton, N. (2002). Some New Q-​Value Correlations to Assist in Site Characterization and Tunnel
Design. International Journal of Rock Mechanics & Mining Sciences Volume 39, pp.
185–​216.
Barton, N. (2013). Shear Strength Criteria for Rock, Rock Joints, Rockfill, and Rock Masses: Problems
and Some Solutions. Journal of Rock Mechanics and Geotechnical Engineering Volume 5,
pp. 249–​261.
Barton, N., Lien, R. and Lunde, J. (1974). Engineering Classification of Rock Masses for the Design
of Tunnel Support. Rock Mechanics, Volume 9, No. 4, pp. 189–​236.
Bieniawski, Z. T. (1979). The Geomechanics Classification in Rock Engineering Applications,
Proceedings, 4th International Society of Rock Mechanics (ISRM) Congress, Montreux. A.
A. Balkema, Rotterdam, Volume 2, pp. 41–​48.
Bieniawski, Z. T. (1984). Rock Mechanics Design in Mining and Tunneling. A.A. Balkema, Rotterdam,
272 pages.
Bieniawski, Z. T. (1989). Engineering Rock Mass Classifications: A Complete Manual for Engineers
and Geologists in Mining, Civil and Petroleum Engineering. Wiley, New York, 251 pages.
Bieniawski, Z. T. (1992). Design Methodology in Rock Engineering. A.A. Balkema, Rotterdam,
198 pages.
Bieniawski, Z.T. (1973). Engineering Classification of Jointed Rock Masses. Transactions of the South
African Institution of Civil Engineers, Volume 15, No. 12, Dec. 1973, pp. 335–​344.
Bieniawski, Z.T. (1974). Geomechanics Classification of Rock Masses and Its Application in
Tunneling. In Advances in Rock Mechanics, Proceedings of 3rd International Congress,
Society of Rock Mechanics, Denver, 1974, Part A, pp. 27–​32. National Academy of Science,
Washington, DC.
Bieniawski, Z.T. (1978). Determining Rock Mass Deformability–​Experiences from Case Histories.
International Journal of Rock Mechanics and Mining Science & Geomechanics Abstracts,
Volume 15, pp. 237–​247.
Brinkgreve, R.B.J., Bakker, K.J. and Bonnier, P.G. (2006). The Relevance of Small-​Strain Stiffness in
Excavation and Tunnelling Projects. Proceedings of 6th European Conference on Numerical
Methods in Geotechnical Engineering, Graz-​Austria, September 6–​8, pp. 133–​139.
Canadian Geotechnical Society (1992). Canadian Foundation Engineering Manual. Canadian
Geotechnical Society, Richmond, BC.
Casagrande, A. (1948). Classification and Identification of Soils. Transactions ASCE, Volume 113,
pp. 901–​932.
Deere, D.U. and Deere, D.W. and United States Army Corps of Engineers (1989). Rock Quality
Designation (RQD) After Twenty Years; Issue 89, Part 1 of Contract Report, US Army Engineer
Waterways Experiment Station, Mississippi.
Duncan, J.M. and Chang, C.Y. (1970). Nonlinear Analysis of Stress and Strain in Soil. ASCE Journal
of the Soil Mechanics and Foundation Division, Volume 96, pp. 1629–​1653.
Edgerton, W. (2008). Recommended Contract Practices for Underground Construction. Society for
Mining, Metallurgy, and Exploration, Littleton, CO.
 Geotechnical investigations for segmental tunnel lining design 95

Essex, R. (2007). Geotechnical Baseline Reports for Construction, Suggested Guidelines. American
Society of Civil Engineers, Reston, VA.
Federal Highway Administration (2009). Technical Manual for Design and Construction of Road
Tunnels –​Civil Elements, FHWA-​NHI-​10-​034. U S Department Of Transportation; Federal
Highway Administration; National Highway Institute, Washington, DC.
Goodman, R.E. (1993), Engineering Geology–​ Rock in Engineering Construction, John Wiley,
New York.
Goodman, R.E. and Shi, G.H. (1985). Block Theory and Its Application to Rock Engineering, Prentice
Hall, New Jersey.
Hashash et al. (2001). Seismic Design and Analysis of Underground Structures, Tunnelling and
Underground Space Technology, Volume 16, Issue 4, 2001, pp. 247–​293, ISSN 0886-​7798,
https://​doi.org/​10.1016/​S0886-​7798(01)00051-​
Hatem, D. (1997). Geotechnical Baselines–​Professional Liability Implications. The Central Artery/​
Tunnel (CA/​T), Professional Liability Reporter, Volume 3, No. 1.
Heuer, R.E. (1974). Important Ground Parameters in Soft Ground Tunneling. Proceedings of Specialty
Conference on Subsurface Exploration for Underground Excavation and Heavy Construction,
ASCE, New York.
Hoek, E. (1982) Geotechnical Considerations in Tunnel Design and Contract Preparation. Journal of
Mines, Metals and Fuels, Volume 30, No. 1, pp. 573–​584.
Hoek, E. and Brown, E.T. (1980). Underground Excavations in Rock. Institution of Mining and
Metallurgy, London, 527 pages.
Hoek, E. and Brown, E.T. (2018). The Hoek–​Brown Failure Criterion and GSI–​2018 Edition. Journal
of Rock Mechanics and Geotechnical Engineering, Volume 11, pp. 445–​463.
Hoek, E. and Diederichs, M.S. (2006) Empirical Estimation of Rock Mass Modulus. International
Journal of Rock Mechanics and Mining Sciences, Volume 43, pp. 203–​215. http://​dx.doi.org/​
10.1016/​j.ijr​mms.2005.06.005
Hoek, E. and Marinos, P. (2000) GSI: A Geologically Friendly Tool for Rock Mass Strength Estimation.
ISRM International Symposium, Melbourne, Australia. ISRM-​IS-​2000-​35.
International Tunneling Association (ITA) (1988). Guidelines for the Design of Tunnels. Tunneling
and Underground Space Technology, Volume 3, No. 3, pp. 237–​249 (April).
International Tunneling Association (ITA) (1997). Immersed and Floating Tunnels State of the Art
Report, Working Group No.11, Tunneling and Underground Space Technology, Volume 12,
No. 2 (April).
Janbu, N. (1963). Soil Compressibility as Determined by Oedometer and Triaxial Tests. Proceedings
of ECSMFE, Volume 1, pp. 19–​25.
Law, T.C.M. and Moore, I.D. (2007). Numerical Modeling of Tight-​Fitting Flexible Liner in Damaged
Sewer Under Earth Loads. Tunnelling and Underground Space Technology: Special Issue on
Trenchless Technology, Volume 22, Issues 5 and 6, pp. 655–​665.
Liu, T.K. (1967). A Review of Engineering Soil Classification Systems. Highway Research Board
Record, Volume 156, pp. 1–​22.
Meguid, M.A., Rowe, R.K., and Lo, K.L. (2002). 3-​D Effects of Surface Construction Over Existing
Subway Tunnels. International Journal of Geomechanics, Volume 2, No. 4, pp. 447–​469.
NAVFAC DM 7 (2011). Soil Mechanics, Foundations and Earth Structures, vulcanhammer.net, ISBN-​
10: 1257064231.
NAVFAC DM 7 (2011). Soil Mechanics, Foundations and Earth Structures, vulcanhammer.net, ISBN-​
10: 1257064231.
Norwegian Geotechnical Institute (2015). Handbook: Rock mass classification and support design
using the Q-​System, NGI, Oslo.
Owen, G.N. and Scholl, R.E. (1981). Earthquake Engineering of Large Underground Structures.
Report no. FHWARD-​ 80195. Federal Highway Administration and National Science
Foundation, Washington, DC.
Peck, R. B. (1969). Deep Excavation and Tunneling in Soft Ground. Proceedings of the 7th International
Conference of Soil Mechanics, Mexico, State-​of-​Art Volume, pp. 225–​290.
96 Handbook of Precast Segmental Tunnel Lining Systems

Proctor, Robert V, and White, Thomas L. (1977) Earth Tunneling with Steel Supports, Commercial
Shearing.
Robinson, C.S. (1972). Prediction of Geology for Tunnel Design and Construction. Proceedings of
Rapid Excavation and Tunnelling Conference, Chicago, 1972 (editors K.S. Lane and L.A.
Garfield), pp. 105–​114. American Institute of Mining Engineers (AIME), New York.
Schmidt, B. (1969). Settlements and Ground Movements Associated with Tunneling in Soil. Ph.D.
Thesis, University of Illinois, Urbana and Champaign, IL, 448 pages.
Chapter 3

Analysis and design of precast

segmental linings
Lead authors: Mehdi Bakhshi and Verya Nasri
Contributing authors: Mahdi Soudkhah, Montazar Rabiei,
Pegah Jarast, and Toufic Maalouf

3.1 INTRODUCTION
The tunnel bore behind the tunnel-​boring machine (TBM) during the excavation in the soft
ground and the fractured and weak rock formations are supported almost immediately by
the installation of precast concrete segmental rings. The thrust needed for TBM driving is
provided by jacks that push against the circumferential joints of the newly installed fully
erected segmental ring. In the last approximately 40 years, precast concrete segmental rings
have showed progressively increased use, not only functioning as the initial ground support
but also acting as the final lining, avoiding the installation of a final cast-​in-​place lining,
therefore termed a “one-​pass lining” system. Precast segments must be designed to resist
the temporary loads from production, transportation and construction as well as permanent
loads from the ground and groundwater in the service stage.

3.1.1 Load and resistance factor design

Tunnel linings like other structural elements must be designed in a manner that the safety
of structure is ensured despite variabilities in the prediction of loads and variabilities in
the properties of materials and structural elements. The load and resistance factor design
(LRFD) method takes all these variabilities into account by specifying most critical load
combinations, applying appropriate factors on the action loads, and applying reduction
factors on the strength of materials and structural elements. LRFD employs specified limit
states to achieve its objectives of constructability, safety and serviceability. It is noteworthy
to mention that the LRFD method is also known as limit state design (LSD), a term that
is being used, for example, in Eurocode 1 (EN 1992-​1-​1:2004). Load combinations, load
factors, strength reduction factors and the design methodologies presented in this chapter
are based on American Concrete Institute (ACI) design codes such as ACI 318-​19, ACI
533.5R-​20 and ACI 544.7R-​16. Nonetheless, these factors and methodologies with some
adjustments can be extended to other structural codes such as EN 1992-​1-​1:2004.

3.1.2 Governing load cases and load factors

From the early stage of manufacturing in the precast plant, to the stage of transportation to
the job site, to the stage of erection and installation inside TBM and securing rings in-​place
with grouting, and finally to the final service condition, precast segments are subjected to
loads of various natures and magnitudes, which are designated by different load cases. The
most critical or governing load cases in the order of occurrence are presented in Table 3.1.

DOI: 10.1201/9781032675541-3
98 Handbook of Precast Segmental Tunnel Lining Systems

The first four load cases in this table, i.e. segment stripping (demolding), segment storage
(stacking), segment transportation and segment handling are categorized as “production
and transient stages” loads. These load cases are analyzed and designed using similar meth-
ods and occur in early stages of production compared to others. The next four load cases in
Table 3.1, i.e. accidental TBM thrust jack removal, TBM thrust jack forces, tail skin back
grouting pressure and localized back grouting (secondary grouting) pressure are catego-
rized as “construction stages” loads. Finally, the last three load cases in Table 3.1 are the
governing load cases are categorized as “final service stages” loads, which all occur after
completion of construction of segmental rings where the lining is completely embedded in
the surrounding ground.
All other loads that may apply to segmental linings but are dependent on the intended
use, the specific configuration/​arrangement or the specific location of the tunnel are referred
to as “other loads” in this chapter. Earthquake or seismic loads; loads of fire, blast and
explosion; loads specific to stacked arrangement of tunnel geometry; aerodynamic loads in
railway, subway or high-​speed rail tunnels; mechanical, electrical, plumbing and overhead
catenary support (OCS) loads; train wheel weights, longitudinal forces of braking and train
derailment loads; loads due to internal pressure in tunnels subjected to high internal water
pressure; are fit into this category and the loading considerations are superimposed on the
final service stage load case.
Table 3.1, in addition to the most critical load cases, presents the required strength (U)
and the required resistance in service (Rs) for each critical load case expressed in terms of
load combinations of all possible action loads and load factors. The required strength (U)
is used for the ultimate limit state (ULS) design and the required resistance in service (Rs)

Table 3.1 Ultimate strength and resistance in service.

Required ultimate strength (U) and resistance in service (Rs) for governing load cases

Required strength (U)

Load case Description required resistance in service (R s)
Load case 1 Stripping (demolding) U =​ 1.4w; R s =​ 1.0w

Load case 2 Storage (stacking) U =​ 1.4(w +​ F); R s =​ 1.0 (w +​ F)

Load case 3 Transportation U =​ 1.4(w +​ F); R s =​ 1.0 (w +​ F)

Load case 4 Handling U =​ 1.4w; R s =​ 1.0w

Load case 5 Accidental removal of TBM U =​ 1.05(w+​Pgr (gasket))

thrust
Load case 6 Thrust jack forces See Table 3.6
Load case 7 Tail skin grouting U =​ 1.25(w ± Pgr); R s =​ 1.0(w +​ Pgr)

Load case 8 Secondary grouting U =​ 1.25(w ± Pgr); R s =​ 1.0(w ± Pgr)

Load case 9 Earth pressure and groundwater U =​ 1.25(w ± WAp) ± 1.35(EH +​ EV) ±1.5 P 0;
load Rs =​ 1.0(w ± WAp) ± 1.0(EH +​ EV) ±1.0 P 0;

Load case 10 Longitudinal joint bursting U =​ 1.25(w ± WAp) ± 1.35(EH +​ EV) ±1.5 P 0;
Rs =​ 1.0(w ± WAp) ± 1.0(EH +​ EV) ±1.0 P 0

Load case 11 Additional distortion U =​ 1.4M distortion; R s =​1 .4M distortion

 Analysis and design of precast segmental linings 99

is used for the service limit state (SLS) design, which are the two most commonly specified
LRFD limit states by codes and regulations.
The load combinations and load factors in Table 3.1 are based on American Concrete
Institute (ACI) design codes of ACI 533.5R-​20 and ACI 544.7R-​16. Elsewhere in the world,
whenever local codes provide a different load combination and different load factor than the
ones shown in this table, those should be followed in lieu of the information in this table.
These load cases and their required strengths (U), together with analysis methods specific to
each load case are further explained in Sections 3.3 to 3.5 of this chapter. Performing ana-
lysis for each critical load case, the resulting internal forces of axial force, bending moment
and shear force are used for design of concrete and reinforcement, after applying reduction
factors on the strength of materials.

3.1.3 General design procedure

Design of precast concrete tunnel segments is an iterative process. The first step toward
the design of tunnel segmental lining is assumption or selection of the geometry of the seg-
mental ring. This step is explained thoroughly in Section 3.2 and is often done based on
past project experiences. The next step toward the design includes assuming material prop-
erties for precast concrete segments with the main parameters namely compressive strength
(f′c) and type and amount of reinforcement. Considering the selected material strength, the
design strength of segments (in ULS and SLS) is calculated and compared with required
strength or required resistance in service against critical loads of the production and tran-
sient stages, which is explained in Section 3.3, critical loads of the construction stages which
is explained in Section 3.4, and critical loads of final service stages, which is explained in
Section 3.5. Upon such comparison, design is optimized by changing the geometry and
material and structural strengths in a few iterations. Detailed design considerations are pre-
sented in Section 3.6, and tests and performance evaluations are discussed in Section 3.7.
Although design for service or SLS design has been discussed in all previous chapters along
the ULS design, a stand-​alone section, Section 3.8, provides more in-​depth insight about
various verifications for this specific design limit state.

3.2 GEOMETRY OF TUNNEL SEGMENTAL RINGS AND THEIR SYSTEMS

Most common shape for precast concrete segmental linings installed inside the TBM shield
is circular rings. Internal dimeter of the ring together with its thickness and its length in
the direction of tunnel axis define the geometry of the circular ring. These three geomet-
rical parameters will be discussed in the following three sub-​sections. Segmental ring sys-
tems, various ring configurations in terms of number of segments comprising a ring, and
geometries of each individual ordinary segment and key segment will be presented in the
subsequent four sub-​sections (Bakhshi and Nasri, 2018). Finally ring optimization using
advanced BIM modeling are discussed in the last sub-​section.

3.2.1 Internal diameter of segmental ring

Internal diameter of the segmental rings is determined mainly based on the space that is
required during the service life of tunnel. From this perspective, it is helpful to categorize
tunnels according to their intended use; i.e. railroad and subway tunnels, road tunnels,
water and wastewater tunnels and utility tunnels. The space proofing is exercised for these
100 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.1 Tunnel space proofing for TBM bored subway tunnels.

Note: Typical space proofing cross-​s ection for single-​t rack TBM-​b ored railway or subway tunnel.

tunnels from the early stages of design through the final stages to ensure minimum required
size of tunnels for their specific intended use and service has been provided.
In single-​track railroad and subway tunnels, as shown in Figure 3.1, the space proofing
at the early stages of design is done initially based on train’s static and dynamic clearance
envelopes considering the maximum super elevation in the tightest curve on the alignment,
track design and structure, size and elevation of safety walkways and emergency evacuation
corridor (egress space), and structure and envelop of the overhead catenary system (OCS)
contact line stays. The internal tunnel diameter is then determined by the smallest circle that
includes all these items and conditions and needs to be verified for providing sufficient ven-
tilation space. In addition, construction tolerances usually in the range of radial tolerance
of R =​±4 inches (100 mm) recommended by ACI 533.5R-​20 and DAUB:2013 should be
provided, which results in an additional 8 inches (200 mm) in diameter to be added to the
required internal structural boundary. This usually will determine the internal diameter of
the segmental ring during the tender design or early design stages. In later stages of design,
the sufficiency of the initially considered space is verified for inclusion of drainage trough,
and the electrical equipment and conduits, standpipe water pipes and other equipment such
as fire-​telephone-​cabinet (FTC), WIFI enclosures, blue light, blue light stations, handrails,
train antenna, train antenna power cable and lighting. Fire life safety requirements of NFPA
130 (2017) are recommended to be followed during the railroad or subway tunnel space
proofing exercise. In a double track or twin tunnel case, tunnel inner dimensions are add-
itionally governed by distance between the centers of tracks or the cross-​passageways.
 Analysis and design of precast segmental linings 101

Figure 3.2 Tunnel space proofing for TBM bored road tunnels.

Note: Typical space proofing for TBM-​b ored road tunnels: typical section.
Source: ITA WG 2, 2019. 

In the road tunnels, the space proofing at the early stages of design is done based on
horizontal and vertical traffic clearances, shoulders, barriers, fans and spaces for ventila-
tion, size and elevation of emergency egress corridor, and size and elevation of FTC cabi-
nets (cabinets for hose reels, fire extinguishers and emergency telephones). One important
parameter that may have a significant impact on the shape and size of traffic clearances is
the maximum super elevation on the tightest curve on the alignment. The smallest circle
encircling all these clearances and elements in addition to 8 inches (200 mm) diametrical
construction tolerance defines the internal tunnel diameter during the tender design or
early design stages. In later stages of design, as shown in Figure 3.2, space proofing is veri­
fied for the tunnel drainage, lighting, traffic control system, water supply pipes for fire-
fighting, passive fire protection, fiber optic cable and electric conduits, signals and signs
above roadway lanes, CCTV surveillance cameras, communication antenna and equip-
ment, and monitoring equipment of noxious emissions and visibility. Note that all new
road tunnels must conform to NFPA 502 (2017) standard requirements for fire life safety
which need to be strictly followed in the space proofing exercise. Also, when the area
under the roadway is designed for low-​point pump station (LPPS), as shown in Figure 3.3,
space proofing should be done with respect to providing sufficient space for access stairs,
dry well, wet well, discharge pipe, drainage pipe, LPPS pump and sump pump in the lower
level of tunnel section.
102 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.3 Tunnel space proofing for TBM bored road tunnels at pump station.
Note: Typical space proofing for TBM-​b ored road tunnels: section at low-​p oint pump station.
Source: ITA WG 2, 2019. 

For the wastewater or combined sewer overflow tunnels, the size of tunnel is defined by
the storage volume and occasionally based on volumetric flow rate of design storm, for
example a 100-​year storm, set forth by local authorities and updated collection system mod-
eling. This is similar to sizing the water conveyance tunnels, which are often defined by the
design flow rate.

3.2.2 Ring thickness and outside diameter

Thickness of the ring is a geometrical and structural parameter. The final thickness of the
ring is determined after verifying structural capacity of the precast segments for all expected
load cases during production, construction and final service stages. However, taking a rea-
sonable initial thickness makes the iterative structural design procedure of precast segments
computationally much more efficient. For this reason, we rely on historical data and pre-
vious experiences of bored tunnels with similar sizes and conditions.
Figure 3.4 compiles the data from more than 100 bored tunnel projects based on the
information in ACI 544.7R-​16, AFTES 2005, Groeneweg (2007) and Blom (2002), and
presents the lining thickness as a function of internal diameter (ID) of tunnel. A linear
regression to this data without any specific restriction or classification results in an ID-​
to-​thickness ratio of about 20 with an R-​squared value of 0.75. In order to further analyze
this data, the ID-​ to-​
thickness ratio is drawn in Figure 3.5 versus the internal tunnel
 Analysis and design of precast segmental linings 103

Figure 3.4 Tunnel diameter and lining thickness.

Note: Relationship between tunnel diameter and lining thickness.

Figure 3.5 Tunnel diameter and lining thickness –​ TBM projects.

Note: Ratio of internal tunnel diameter to lining thickness for different TBM tunnel projects.

diameter. General observation from Figure 3.5 is that the ID-​to-​thickness ratio increases
with the tunnel diameter; however, this trend stops around tunnel diameters in the range
of 23 feet (7 m). Breaking the figure into different size categories, Figure 3.5 shows that
for the tunnel diameter under 13 feet (4 m), disregarding the higher and lower outliers, the
internal diameter-​to-​thickness ratio is in the range of 11 to 20. When the tunnel dimeter was
between 13 feet (4 m) and 18 feet (5.5 m), the internal diameter-​to-​thickness ratio is in the
range of 14 to 26 disregarding one high outlier datum.
104 Handbook of Precast Segmental Tunnel Lining Systems

Finally, when the tunnel dimeter was more than 18 feet (5.5 m), the internal diameter-​to-​
thickness ratio is in the range of 17 to 31. In addition, JSCE (2007) recommends that the
ring thickness should be less than 4% of the tunnel outer diameter, a statement that leads
to an internal diameter-​to-​thickness ratio of 23 or less. Considering all the information and
analysis, as the best practice, it is recommended to take initial ring thickness as 1/​23rd to
1/​20th of the internal dimeter (ID) for tunnels larger than 13 feet (4 m) in diameter, and 1/​
15th of the internal dimeter (ID) for tunnels smaller than 13 feet (4 m) in diameter.
Segmental ring thickness in addition to providing sufficient structural capacity must sat-
isfy the conditions imposed by the contact joints. This means providing sufficient bearing
surface area at the circumferential joints compatible with TBM thrust jacks, sufficient bear-
ing surface at the longitudinal joints for transferring hoop forces without causing exces-
sive spalling stresses, and sufficient clear space for segment recesses, i.e. gasket recess and
stress relief recess. Also, from the durability perspective, the lining thickness, for example
in wastewater or combined sewer overflow tunnels, must account for an additional sacrifi-
cial lining thickness that will be sacrificed during the service life of tunnel due to microbial
induced concrete corrosion (Chapter 12). Due to durability considerations, DAUB:2013
recommends a minimum thickness of 12 inches (300 mm) for one-​pass tunnel linings.
Tunnel lining outer diameter is determined simply by adding the thickness of segmental
ring to its internal diameter. The TBM shield outer diameter is obtained by adding tunnel
lining outer diameter to a so-​called parameter “overcut,” a gap space between the segment
outer side (extrados) and the ground profile, which is filled by backfill annulus grout. This
gap space, which ranges from 4.5 inches (115 mm) to 10.8 inches (275 mm) (depending
upon the TBM diameter, ground condition and alignment curvature) is also equal to sum-
mation of the tail clearance inside the shield, shield skin plate thickness and an additional
TBM clearance with the ground at the tail due to the conical shape of the shield.
There is a clear relationship between the shield outer diameter and the minimum radius
of the tightest curve of the alignment as one can limit the other or vice versa. Historical data
from JSCE 2007 on the minimum curve radius versus shield outer diameter of more than
100 tunnel projects is presented in Figure 3.6. The best way to analyze this data is to cat­
egorize them into three categories of shield outer diameter: less than 20 feet (6 m), between
20 and 32 feet (6 and 10 m) and more than 40 feet (12 m). As one can see, the upper bound
limit for minimum curve radius for all size categories is 1,650 feet (500 m), but the lower
bound limit for the minimum curve radius is a function of shield outer diameter category.
For shield outer diameter in 20 feet (6 m) or less, the minimum curve radius is limited to
about 260 feet (80 m), and for shield outer diameter in the range of 20 to 32 feet (6 to 10
m), the minimum curve radius is limited to about 520 feet (160 m). This is 990 fee (300 m)
for the largest shield outer diameter category of more than 40 feet (12 m). These numbers
can be used as a guide to provide a relationship between shield outer diameters and lower
bound limits for the radius of tightest curve in the alignment. Note that the minimum curve
radius is not only related to the shield outer diameter or vice versa, but also to the ring
geometry (taper and ring width), overcut, shield design (articulated or not), and radial gap
between segment and tail skin.

3.2.3 Ring length
Length of segmental rings or the width of segments historically ranges 2.5 to 8.0 feet (0.75
to 2.5 m) with tunnels with smaller diameters often times toward the lower side of this
range and the ones with larger diameters toward the upper side. The data collected from
 Analysis and design of precast segmental linings 105

Figure 3.6 Tunnel minimum curve radius and TBM shield outer diameter.
Note: Relationship between minimum curve radius and TBM shield outer diameter.
Source: JSCE 2007. 

more than 60 projects in Japan before 2006 (JSCE, 2007) presented in Figure 3.7 indicate
that while the ring length increases by the size of tunnel, the relationship is not very strong.
This can be justified as the ring length or segment depends also on the tightest curve radius
in the tunnel alignment. In addition, narrower rings are favorable for transportation and
erection simplicity, construction of curved alignments and reducing the length of shield tail.
Wider rings are favorable for reducing the production cost, numbers of joints, the total per-
imeter of gasket length in project, the number of bolt pockets where leakage may occur and
increasing the construction speed (JSCE, 2007).
Among other parameters, the ring length depends on the available space for segment sup-
ply and handling inside the shield, the segment production easement, the segment weight
and the weight limitations during transportation to the site on public roads as often the main
limiting condition. The length of the ring is taken as an optimized value considering all these
conditions and parameters and the efficiency of tunnel works. The best practice is to use a
ring length of 5 to 5.5 feet (1.5 to 1.7 m) for tunnel diameters of 19 to 23 feet (6 to 7 m), a
ring length of 6 to 6.5 feet (1.8 to 2 m) for tunnels of 23 to 30 feet (7 to 9 m) in diameter,
and a ring length of 6.5 to 7.5 feet (2 to 2.3 m) for tunnels larger than 30 feet (9 m). Most
advanced ring length optimization methods by BIM modeling is presented in section 3.2.8.

3.2.4 Segmental ring systems

Main segmental ring systems include parallel rings, parallel rings with corrective rings,
right/​left or top/​down rings, and universal rings.
106 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.7 Segmental ring length and outer diameter.

Note: Ring length or segment width as a function of outer diameter of segmental ring.
Source: JSCE 2007. 

Figure 3.8 Schematics of different segmental ring systems.

Note: Different ring systems in plan view when negotiating straight drives and curves together
with presentation of ring taper parameter (k): parallel rings.

Figure 3.8 shows parallel rings, rings with parallel circumferential faces, which are per­
pendicular to the tunnel axis. These rings are not inherently suitable for curved alignments
and since most tunnels have curves, negotiating alignment curves is only possible through
the use of packers placed partially in circumferential joints. This ring system cannot be
 Analysis and design of precast segmental linings 107

Figure 3.9 Schematics of different segmental ring systems.

Note: Different ring systems in plan view when negotiating straight drives and curves together
with presentation of ring taper parameter (k): Left and Right-​h and rings.

properly sealed because packers reduce the compression in the gasket. Even in straight
alignment, rings can never be built perfectly straight and adopting methods to restore line
and grade require use of packers, which leads to sealing issues.
In parallel rings with corrective ring systems, corrective rings (up, down, left or right ring)
replace packers for negotiating curves. Up rings are the rings that do not have two parallel
circumferential faces and at least one face of the ring is tapered in a way that a sequence of
these rings turn in upward direction when key segment is located at or in the proximity of
the crown. By keeping the key segment in that place, in similar fashion but with a different
taper, the down rings turn in the downward direction, the left rings turn in the left and the
right rings turn in the right direction. The main disadvantage of this system is the require-
ment for different types of formwork sets, which are cost prohibitive.
The right/​left ring is a type of ring that tapered in a way that when the key segment is
above the springline, the ring turns in the right or left direction. Figure 3.9 shows most
common right/​left ring system with one circumferential face (often the front face) perpen-
dicular to the tunnel axis and the other face (often the rear face) inclined to the tunnel axis
with a slightly longer length on the right or left side when looking at the elevation view of
the ring. The difference between maximum and minimum ring length is called ring taper.
The entire tunnel lining can be built with different sequences of these right and left rings.
The sequence of right-​tapered (right) and left-​tapered (left) rings produces a straight align-
ment or a tangent alignment whereas a sequence of right-​right ring, as shown in Figure 3.9,
results in a curve to the right, and left-​left ring produces a curve to the left with a minimum
system radius.
Upward and downward turns/​curves are achieved through rotation of the right or left
rings by 90 degrees. Instead of using right/​left rings and following same concept but with
a taper in a different axis, the entire tunnel can be built with up/​down rings. On up/​down
rings, when looking at the plan view of the ring, the most common taper is when the ring
has one face (often the front face) perpendicular to the tunnel axis and on the other face
(often the rear) is tapered in a way that it has a slightly longer length on the upward or
downward side. Right/​left rings and up/​down rings have been also made of rings with
double taper, i.e. tapers on both faces of the rings with half of the required taper on each
end face of the ring to turn in the rings in the desired directions. Using the right/​left or up/​
down ring system, it is possible to always have the key segment on the top and, therefore,
108 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.10 Schematics of different segmental ring systems –​ universal rings.

Note: Different ring systems in plan view when negotiating straight drives and curves together
with presentation of ring taper parameter (k): universal rings.

to be able to build the ring from the bottom upward, which is preferable to some contrac-
tors. The right/​left or up/​down ring system provides a proper sealing performance for an
impermeable tunnel, with the only disadvantage being requiring different types of form-
work set.
In the universal ring system, the key segment can be located anywhere in the tunnel
including below the springline and at or near the invert. With this system, the ring can
turn in any desired direction of up, down, left, right and their combinations. As shown
in Figure 3.10, universal rings are often built with both circumferential faces of the ring
inclined to the tunnel axis. Nonetheless, universal rings with only one tapered side –​prefer-
ably the rear face –​are occasionally used. Depending on having both circumferential faces
of the ring tapered or only one, the required ring taper is divided in one or both ends of the
universal rings. Equation (3.1) can be used to calculate the required ring taper (k) based on
the size of the ring and tightest curve in the alignment.

k = ∅ A .bm /R (3.1)

where
∅ A is outer diameter of the segmental ring
bm is the average ring length at the tunnel springline
R is the minimum curve radius

It is necessary to consider the effect of correction curve drives, which are required in the
potential events of deviations from the theoretical alignment in order to return the tunnel/​
TBM back into the designed alignment. The minimum curve radius should be taken at least
20% less than the smallest curve radius in the horizontal or the vertical alignment (ACI
533.5R-​20, DAUB:2013). Figure 3.10 shows how straight drives or tangent alignments and
different curves and directional turns can be negotiated through the rotation of the universal
segmental ring.
 Analysis and design of precast segmental linings 109

In order to develop a straight drive or tangent alignment with universal rings, every
other ring must be turned by 180° in reference to the previous one. This means that if we
start with a ring that have a taper on the upward and downward sides with a key segment
located on the top at or near the tunnel crown (similar to up ring), by rotating the next ring
by 180°, the key segment will be located in the bottom at or near the tunnel invert. For
minimum right or left curve radius, such a universal ring must be rotated by 90o resulting
in the key segment placement at or near the tunnel springline. Such a design is preferred by
some designers and contractors who prefer to place the key segment where the ring length
is the narrowest.
Universal rings can be tapered alternatively on the right and left sides of the rings with the
key segment being located at or near the tunnel crown. While placement of the key segment
above the springline for the entire alignment is not an inherent characteristic of this ring
system, in recent years and using advanced software for guiding TBM, universal rings can
negotiate the curves and straight drives with keys always above springline through adjusting
the drive error of less than fractions of an inch (a few millimeters) in two or three rings. The
main advantage of the universal ring system is the requirement for only one type of form-
work set (ÖVBB 2011).

3.2.5 Ring segmentation
The number of segments that comprise a ring is one of the main parameters for the design
of segmental tunnel linings. From transportation, handling and erection process perspec-
tive, shorter segments or higher number of segments that comprise a ring is preferred.
Nonetheless, longer segments or a smaller number of segments that make up a ring results
in a stiffer segmental ring, reduced production cost, less hardware for segment connection,
shorter gasket length for entire project, fewer number of bolt pockets where cracking and
leakage can occur, and increased construction speed.
In tunnels smaller than 20 feet (6 m) in diameter, maximum segment lengths are nor-
mally limited by the space available to turn the segments inside the TBM shield and in the
backup gantries behind the shield. In larger tunnels, the weight of segment and its structural
capacity are usually the main factors for deciding maximum length of segments. For this
purpose, a parameter called segment slenderness (λ) is presented, which is defined as the
ratio between the breadth or curved length of segment along its centroid and the segment
thickness. Review of previous tunneling projects indicate that segment slenderness is usually
between 8 to 13. Up until a few years ago, slenderness ratio for fiber-​reinforced concrete
(FRC) segments were limited to 10. Nonetheless, in recent years and with the introduction
of high-​strength steel and advanced and innovative anchorage technologies for fibers, FRC
segment with slenderness of more than 10 and up to 13 have been used (ACI 544.7R-​16;
Bakhshi and Nasri, 2017a; Beňo and Hilar, 2013; Harding and Francis, 2013; ITA Working
Group 2, 2016). General criteria for ring segmentation of tunnels larger than 20 feet (6 m)
in diameter can be expressed as dividing the ring into as many segments that result in a
single segment’s slenderness ratio of 10 or higher.
In order to provide a guideline on the ring segmentation based on the tunnel size,
numerous authors have done an extensive review of previous TBM tunnel projects
(Bakhshi and Nasri, 2018). Results of this review indicate that TBM tunnels can be
divided into five different size categories shown in Table 3.2 according to the number of
segments in a ring. While Table 3.2 presents most common cases in practice, there will
110 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.11 5+​1 segmental ring arrangement.

Note: 5+​1 ring segmentation and virtual build of tunnel when negotiating the tightest curve.

be segmental ring segmentations for different sizes that are not shown in this table. In
general, the most common configurations comprise of an even number of segments with
an odd number of ordinary segments and one key segment, usually one-​third the size of
ordinary segments, or with an even number of ordinary segments and two key segments.
Such a design is most compatible with an even number of TBM thrust pads as often three
pads push against each ordinary segment and one pad pushes against the small key seg-
ment. On occasions, key segments are smaller than one-​third or as large as half or full size
of ordinary segments.
For tunnels 20 feet (6 m) in diameter or less, a ring division into six segments is typical.
The most common configuration is five ordinary segments and one small key segment, also
known as 5+​1 configuration (Figure 3.11). Also, a ring segmentation into four ordinary
segments and two smaller key segments (one at or near invert and one at or near crown)
also known as 4+​2, is common. For this size of tunnel, sometimes a ring segmentation with
three ordinary segments (e.g. each covering 72o on tunnel perimeter), two counter key seg-
ments (e.g. each covering 56.5o on tunnel perimeter), and one key segment (e.g. covering
31o on tunnel perimeter) also known as 3+​2+​1, or sometimes a ring segmentation with six
segments of the same length including the key segment (each covering 60o on the tunnel per-
imeter) have also been used. The same configuration concept consisting of an even number
of segments with an odd number of ordinary segments and one or key segment, or with an
even number of ordinary segments and two key segments is utilized and is the most common
also for tunnels 26 to 36 feet (8 to 11 m) in diameter as well as for tunnels larger than 46
feet (14 m) in diameter.
As shown in Table 3.2, 7+​1 and 6+​2 are the most common ring segmentations for the
former category, and 9+​1 with either small or large key and 11+​11 are the most common
ring configurations for the latter group. However, when the tunnel is 20 to 26 feet (6 to 8
m) in diameter or when the tunnel dimeter ranges 36 to 46 feet (11 to 14 m), for example,
a 5+​1 ring configuration with a small key for the former category and a 7+​1 with a small
key for the latter may result in very long segments with potential weight issues and insufficient
structural capacity. Solutions adopted for tunnels in the range of 20 to 26 feet (6 to
8 m) in diameter, is either to use a 5+​1 ring configuration with a large key (6 segment ring
 Analysis and design of precast segmental linings 111

Table 3.2 Common ring configuration and segmentation.

Most common ring configuration and segmentation for tunnel segmental lining
systems

Most common ring configuration

Item Tunnel diameter category and segmentation
1 Tunnel diameter < 20 feet (6m) 5+​1 ; 4+​2 ; 3+​2 +​1 *; 6
2 20 feet (6m) < tunnel diameter < 26 feet (8m) 5+​1 large key; 6+​1
3 26 feet (8m) < tunnel diameter < 36 feet (11m) 7+​1 ; 6+​2
4 36 feet (11m) < tunnel diameter < 46 feet (14m) 7+​1 large key; 7+​1 +​1 †
5 46 feet (14m) < tunnel diameter 9+​1 ; 9+​1 large key; 11+​1
Notes:
*
This configuration consists of three ordinary segments, two slightly smaller counter key segments, and one
small key segment.
†
In this configuration key can be one-​t hird of ordinary segments and one of counter keys can be as small as
two-​t hird of ordinary segments.

Figure 3.12 Developed plan view of hexagonal segment system.

Note: Developed plan on the intrados of a segmental ring system consisting of hexagonal
segments.

configuration) to reduce the length of ordinary segments, or adopting a 6+​1 ring segmentation
with a small key, which is not ideal but can solve the issues related to the length of
segments. Similarly, for tunnels between 36 and 46 feet (11 and 14 m) a 7+​1 ring configur-
ation but with a large key (eight segment ring configuration with each segment covering
45o on tunnel perimeter) or a 7+​1+​1 ring configuration with basically same segmentation as
eight segment ring but this time with dividing one full segment in one-​third and two-​third to
create key (covering 15o on perimeter) and counter key segment (covering 30o on perimeter).

3.2.6 Segment shape
Individual segment geometry can be categorized into four main categories of hexagonal,
rectangular, trapezoidal and rhomboidal segment systems.
The oldest segment geometry system is the hexagonal system shown in Figure 3.12. In
this system, TBM does not need to stop for the ring build as TBM jacks can thrust against
counter key segments (half of the ring) while erecting and installing the other half of the ring
112 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.13 Developed plan view of rectangular segment system.

Note: Developed plan on the intrados of a segmental ring system which mainly consists of rect-
angular segments (with the exception of key and counter key segments).

in shape of key segments. Then for the next ring, TBM jacks push against newly installed
half of the ring while the other half of the ring can be erected in an open teeth condition.
Disadvantages of this system include discontinuous ring build with this system, suscepti-
bility of gap opening between segments over successive rings, and inefficiency of gaskets in
this system for sealing the tunnel. Because the hexagonal segment system compromises the
watertightness of the tunnel, it is now only being used in long straight (tangent alignment)
tunnels where water infiltration is not a concern, or in part of two-​pass lining system as
the first lining system. When this system is adopted and especially in conjunction with the
double-​shield TBM it can result in very rapid advance rates.
In a rectangular system, all segments comprising a ring are in rectangular shape except
only the key and counter key segments that have a slightly tapered joint angle. With this
system, the ring is built continuously starting with erecting first the reverse key segment
(segment located opposite to key segment) and then erecting other segments, for example,
with a sequence shown in Figure 3.13 toward the key segment, which is installed last. The
main advantages of this system are the continuous ring build, efficiency of gaskets in sealing
the segment joints, and the simplicity of longitudinal joint geometry. However, there are
three major issues with this segment system.

First issue is the likelihood of star or crucifix joints as staggered joints are not inherently
guaranteed with these segments and sealing performance can be compromised.
Second issue is the difficulty of installing rectangular segments without impacting the gasket
on the adjacent segment because of the right angle at the joints.
Third issue is the difficulty of using fast-​connecting dowels with this system because dowels
can withstand only limited construction tolerances compared to time-​consuming bolting
systems. While bolts are fastened after complete segment insertion into the ring, dow-
els are preinserted into segments, which limits the segment approach path into a very
narrow path and may cause more friction between gaskets of adjacent segments (ACI
533.5R-​20).

This system is still being used in large-​diameter tunnels where there is a high demand for
shear capacity of connection system at the circumferential joints and the dowels unlike bolts
may not provide sufficient capacity.
 Analysis and design of precast segmental linings 113

Figure 3.14 Developed plan view of trapezoidal segment system.

Note: Developed plan on the intrados of a segmental ring system which consists of trapezoid
segments.

In trapezoidal segment system, all segments comprising a ring (usually an even number
of segments) are in trapezoidal shape and have the same length along the segment center-
line. As shown in Figure 3.14, half of the ring’s segments are reverse key type or counter
key type segments (wider on rear side of the ring) and the other half are key type segments
(narrower on the rear side of the ring). From the ring build perspective, as shown
in Figure 3.14, this system is similar to a hexagonal system assembled by erecting first
the reverse key type segments (e.g. segments 1, 2 and 3) like an in open-​teeth configuration
and then erecting the key type segments in the gaps between them (e.g. segments
4, 5 and 6). The main advantage of this system is the creation of staggered longitudinal
joints from otherwise star or crucifix joints (in case of rectangular segments) for negoti-
ating most important drives in the alignment such as the tightest right/​left or upward/​
downward curves. See Section 3.2.8.2 for more details on this subject. Another advantage
of this system is that every other segment is a key type segment and therefore with this
configuration, there is a freedom to close the ring at almost any given location. The main
disadvantage is the discontinuous ring build process.
In the rhomboidal or parallelogrammical-​trapezoidal segment system ordinary segments
are in the shape of parallelogram and only key and reverse key segments are in the shape
of a trapezoid. From the ring build perspective, for example as shown in Figure 3.15,
process starts with erecting trapezoid reverse key segment, which is No. 1, followed by
continuous placement of ordinary parallelogrammical segments Nos. 2 and 3 on each side
of reverse key segment, and then placing next ordinary segments Nos. 4, 5, 6 alternating
left and right, and finally closing the ring with the installation of key segment or segment
No. 7. Therefore, the ring is built continuously, from bottom to top, or right to left, or
vice versa.
Rhomboidal segment system is the most common segment shape system nowadays because
of eliminating all inherent disadvantages of previously discussed systems. First, this segment
system is compatible with efficient use of gaskets. This, together with replacing important
crucifix joints (present in rectangular segments) by the staggered joints provide the best
sealing performance among all systems. Continuous ring build process as explained above
is another major advantage of this system, especially when compared with trapezoidal seg-
ment system. Compatibility with use of fast-​connecting dowels instead of time-​consuming
114 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.15 Developed plan view of rhomboidal segment system.

Note: Developed plan on the intrados of a segmental ring system which consists of parallelogram
or trapezoid segments (with the exception of key and counter key segments).

and labor-​intensive bolting procedure is another major advantage of this system. This is
due to angled/​tapered longitudinal joints, which prevent early friction of the gasket of the
segment being installed with the gaskets of adjacent installed segment during the ring build
phase, resulting in reduced construction tolerance for connection devices and facilitating use
of dowels in circumferential joints.

3.2.7 Geometry of key segment

The key segment is the last segment installed in the process of the ring build to close a ring.
For this purpose, the key segment is always made in a trapezoidal shape (in plan view),
i.e. narrower on the rear side of the ring than the front side, to provide a smooth insertion
and easy sliding path in a wide-​angled gap that is provided by tapered joints of counter
key segments. As briefly explained in Section 3.2.5, the key segment is usually smaller than
ordinary segments, and as large as 1/​3rd, half, or 2/​3rd the size of ordinary segments. On
occasions, the key segment is smaller than 1/​3rd the length of ordinary segments or as large
as full size ordinary segments or is of a different length ratio with respect to ordinary seg-
ments. In addition to geometry and length of the key segment, tapering of its longitudinal
joints is also an important design and construction parameter.
Depending on the ring build procedure, i.e. ring build by segment erection in the radial
direction vs. ring build by segment erection in the longitudinal direction, two different key
segment tapering geometries have been historically used (ITA WG 2, 2000; JSCE, 2007).
For the procedure that segments are inserted radially from the inside of the tunnel, the longi-
tudinal side faces of the key segment are tapered with respect to the tunnel radius direction.
This ring build procedure has been rarely used in recent years, and therefore, readers are
encouraged to refer to ITA WG 2 (2000) for geometry formulations of this specific key seg-
ment, which is inserted in the radial direction. Most common practice is, however, to erect
all the segments including the key segment from the cut face side in the longitudinal direc-
tion of the tunnel. For this common ring build procedure, the longitudinal joints of the key
segment are tapered with respect to the longitudinal tunnel axis. This tapering angle usually
ranges 8° to 12° and is often taken as the same tapering angle as the ones for longitudinal
joints of ordinary segments in a rhomboidal or trapezoidal system.
 Analysis and design of precast segmental linings 115

3.2.8 Ring optimization using advanced BIM modeling

3.2.8.1 Ring length optimization
Negotiating curved alignment by a segmental ring, through a curved drive naturally pro-
duces minor deviations in line and grade that are restored using error minimization (EM)
algorithms. As each ring is rotated relative to the previous one, its centerline pivots in a cone
shape about its starting point (Figure 3.16). The base of this cone, representing the endpoint
of the most recently erected ring, deviates from the theoretical alignment in the curves where
the total distance vector caused by each rotation can be measured. It is important that
the maximum deviation at each ring does not exceed a predefined allowable deviation for
design and coordination purposes.
Past tunnel construction records indicate that the ring length has tended to become longer
over time. In the presented approach, preliminary ring optimization is performed in terms of
selecting the longest ring length for the minimum curve radius (tightest curve) in the tunnel
alignment after setting a maximum allowable deviation from the theoretical location as the
criterion. Figure 3.17 shows universal rings tapered on both end faces designed for negoti­
ating a curve.

where
‘R’ =​radius
‘k’ represents the taper
‘D’ represents the tunnel outside diameter
‘bm’ is the ring length, which is aimed to be optimized

The local maximum deviation ‘Δmax’ is determined by the face angle ‘θ’ and length of the
ring ‘bm’. Since taper k =​(Lmax –​ Lmin) is linearly proportional to the diameter, ‘D’, the same

Figure 3.16 Cone shape envelope of centerline locations for two adjacent rings.
Note: Possible ring centerline set-​o ut resulting from all relative rotations of the ring.
116 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.17 Ring taper and local maximum deviation of the ring.

Note: Universal ring tapered on end faces to negotiate curve drive (radius ‘R’).

face angle, ‘θ’, can be achieved for a fixed minimum path radius, ‘R’. Therefore, for a given
alignment, the diameter of the ring, ‘D’, has no direct impact on the resulting deviation.
In addition to the minimum curve radius ‘R’, the number of circumferential connections
in a ring ‘n’ –​and its evenness/​oddness –​is the only other parameter that affects the max-
imum local deviation of segmental rings in the curved drives. Although this parameter has
not been formulated in Figure 3.17, its impact on the resulting deviation is due to limiting
the maximum possible relative rotations of the ring. Parameter ‘n’ itself is a function of
a number of segments per ring, and number of circumferential connections per each seg-
ment. With the dowel system as the most common connection device for the circumferential
joints, often three dowels are designed for each ordinary segment and one dowel for the key
segment (often sized as 1/​3rd of the ordinary segments).
The most common ring segmentations and ring lengths categorized based on the tunnel
diameter (Section 3.2.5) are used in these examples to determine most common values for
parameter ‘n’. As a result of the primary analysis, as shown in the upper half of Table 3.3,
maximum deviation of the segmental ring is presented as function of minimum curve radius,
and total number of circumferential connections per ring ‘n’. As for the maximum allowable
deviation of the ring centerline from its theoretical location, 0.24 inches (6 mm) is often set as
the criteria. This roughly corresponds to 1.2 (30 mm) deviation of the TBM shield from the
TBM theoretical location since in general, length of the shield is about 5 times the length of a
single ring that is being installed inside it. Respecting maximum allowable ring deviation [e.g.
0.25 inches (6 mm)] and using the upper half of Table 3.3 when there is no other restriction
(see Table 3.4), one can determine the maximum length of the ring as a function of minimum
curve radius of the project and the number of circumferential connections per ring.
 Analysis and design of precast segmental linings 117

3.2.8.2 Ring optimization by eliminating crucifix joints

A crucifix joint is formed between adjacent rings when the longitudinal joints between the
segments directly align (Figure 3.18). These joints are more difficult to seal, are prone to
leakage, and thus should be prevented (or minimized) during the construction. As a com-
mon solution, this can be achieved by first choosing a rhomboidal segments system over
the rectangular segments’ system, and then eliminating ring rotations, which result in a cru-
cifix joint. On one hand with rhomboidal segments (angled longitudinal joints), the overall
number of crucifix joints in the entire alignment is drastically reduced.
On the other hand, as shown in Table 3.4, when comparing the number of relative ring
rotations that cause any crucifix joint, rectangular segments’ system may have fewer num-
bers than those for rhomboidal segments (9 vs. 10). Nonetheless, eliminating one in relative
rotation in rectangular segment systems (Figure 3.19) can be very consequential in terms of
increasing resulting deviation by a factor of 2 (Table 3.4) making rhomboidal segments a
favorable system.
Figure 3.20 show an example of a 5+​1 rhomboidal segmental ring system with 16 cir­
cumferential connections. On this ring, as shown in Figure 3.21, 10 out of 16 possible rela­
tive rotations result in developing at least one crucifix joint. By eliminating these rotations,
only six relative rotations can be available for design and negotiating alignment curves.
This leads to a significant increase in the resulting deviation, which is shown in the lower
half of Table 3.3 for typical rings with key segment an exact fraction (one-​third) of an
ordinary segment. For example, on a 7+​1 ring configuration and for a given curve radius
of 820 feet (250 m), even a ring length as short as 5.0 feet (1.52 m) results in a deviation
of more than 0.25 mm (6 mm), i.e. allowable maximum tolerance, when eliminating rota-
tions that correspond to crucifix joints. This can be compared to the case of primary ring
optimization on the upper half of the table where for the very exact same ring type and
curve radius, a ring length of 7.5 feet (2.29 m) results in an acceptable deviation under
0.25 inches (6 mm).

3.2.8.3 Ring optimization by slight size adjustment of one segment

Crucifix joints are unavoidable as long as the ratio of the arc angle of any segment in the
ring is determined from the total number of connections (n). This is always the case when,
for example, a key segment is an exact fraction (one-​third) of the size of ordinary seg-
ments. If, however, the arc angle of any segment is extended or reduced beyond such ratio,
it is possible to prevent any crucifix joints forming between adjacent rings. Depending on
the number of connections, arc angle, length and angle of longitudinal joints, the amount
that a segment angle can be adjusted is quite limited. Nonetheless, those few degrees can
make all the difference. For example, in Figure 3.22, the arc angle at the key segment has
been reduced by only 3.4° (19° to 15.6°) with the excess being taken up by the reverse key
segment.
Assuming an inner radius of 10.63 feet (3.24 m) this slight adjustment offsets the key seg-
ment by ( 3.4 / 360 × 2πri ) or approximately 3.5 inches (90 mm) on each side resulting in a
minimum 1.80 inches (45 mm) distance between adjacent segment joints in the worst-​case
scenario (Figure 3.23). With this slight adjustment, no relative rotations are eliminated to
prevent the crucifix joints. All possible relative rotations of the ring are available for nego-
tiating curves, which results in a drastic reduction of determined deviation for a given curve
radius (upper half of Table 3.3).
118 Handbook of Precast Segmental Tunnel Lining Systems

Table 3.3 Alignment deviation as a function of curve radius. Resulting alignment deviation as a

function of curve radius, ring length and number of circumferential connections
per ring. The maximum allowable deviation of the ring centerline from its
theoretical location is often set as 0.24 inches (6 mm).

Ring configuration Minimum radius

Diameter Segm’ts Conn’ns Length 80 m 120 m 160 m
Category 1 <6 5+​1 16 1.219 0.007 0.004 0.003
1.372 0.008 0.006 0.004
1.524 0.010 0.007 0.005
1.676 0.012 0.008 0.006
1.829 0.015 0.010 0.007
1.981 0.017 0.012 0.009
2.134 0.020 0.013 0.010
2.286 0.023 0.015 0.012
Category 2 6–​8 6+​1 19 1.372 0.004 0.003
1.524 0.005 0.004
1.676 0.007 0.005
1.829 0.008 0.006
1.981 0.009 0.007
2.134 0.011 0.008
2.286 0.012 0.009
2.438 0.014 0.010
Category 3 8–​1 1 7+​1 22 1.524 0.004
1.676 0.005
1.829 0.006
1.981 0.007
2.134 0.009
2.286 0.010
2.438 0.011
2.591 0.013
Category 4 11–​1 4 8+​1 25 1.676
1.829
1.981
2.134
2.286
2.438
2.591
2.743
 Analysis and design of precast segmental linings 119

180 m 200 m 250 m 300 m 350 m 400 m 450 m 500 m

0.003 0.003 0.002 0.002 0.002 0.001 0.001 0.001
0.004 0.003 0.003 0.002 0.002 0.002 0.001 0.001
0.005 0.004 0.003 0.003 0.002 0.002 0.002 0.002
0.006 0.005 0.004 0.003 0.003 0.002 0.002 0.002
0.007 0.006 0.005 0.004 0.003 0.003 0.003 0.002
0.008 0.007 0.006 0.005 0.004 0.003 0.003 0.003
0.009 0.008 0.006 0.005 0.005 0.004 0.004 0.003
0.010 0.009 0.007 0.006 0.005 0.005 0.004 0.004
0.003 0.003 0.002 0.002 0.001 0.001 0.001 0.001
0.004 0.003 0.003 0.002 0.002 0.002 0.001 0.001
0.004 0.004 0.003 0.003 0.002 0.002 0.002 0.002
0.005 0.005 0.004 0.003 0.003 0.002 0.002 0.002
0.006 0.005 0.004 0.004 0.003 0.003 0.002 0.002
0.007 0.006 0.005 0.004 0.004 0.003 0.003 0.003
0.008 0.007 0.006 0.005 0.004 0.004 0.003 0.003
0.009 0.008 0.007 0.005 0.005 0.004 0.004 0.003
0.004 0.004 0.003 0.002 0.002 0.002 0.002 0.001
0.005 0.004 0.003 0.003 0.002 0.002 0.002 0.002
0.006 0.005 0.004 0.003 0.003 0.003 0.002 0.002
0.007 0.006 0.005 0.004 0.003 0.003 0.003 0.002
0.008 0.007 0.006 0.005 0.004 0.003 0.003 0.003
0.009 0.008 0.006 0.005 0.005 0.004 0.004 0.003
0.010 0.009 0.007 0.006 0.005 0.005 0.004 0.004
0.011 0.010 0.008 0.007 0.006 0.005 0.005 0.004
0.004 0.004 0.003 0.003 0.002 0.002 0.002 0.002
0.005 0.005 0.004 0.003 0.003 0.002 0.002 0.002
0.006 0.005 0.004 0.004 0.003 0.003 0.002 0.002
0.007 0.006 0.005 0.004 0.004 0.003 0.003 0.003
0.008 0.007 0.006 0.005 0.004 0.004 0.003 0.003
0.009 0.008 0.007 0.006 0.005 0.004 0.004 0.003
0.010 0.009 0.007 0.006 0.005 0.005 0.004 0.004
0.012 0.010 0.008 0.007 0.006 0.005 0.005 0.004

(Continued)
120 Handbook of Precast Segmental Tunnel Lining Systems

Table 3.3 (Continued)

Ring configuration Minimum radius

Diameter Segm’ts Conn’ns Length 80 m 120 m 160 m
Category 5 > 14 9+​1 28 1.829
1.981
2.134
2.286
2.438
2.591
2.743
2.896
Category 1 <6 5+​1 6 of 16 1.219 0.019 0.013 0.01
(Eliminating
crucifix joint) 1.372 0.024 0.016 0.012
1.524 0.03 0.02 0.015
1.676 0.036 0.024 0.018
1.829 0.043 0.029 0.022
1.981 0.051 0.034 0.025
2.134 0.059 0.039 0.029
2.286 0.068 0.045 0.034
Category 2 6–​8 6+​1 7 of 19 1.372 0.011 0.008
(Eliminating
crucifix joint) 1.524 0.014 0.01
1.676 0.017 0.013
1.829 0.02 0.015
1.981 0.024 0.017
2.134 0.027 0.02
2.286 0.031 0.024
2.438 0.035 0.027
Category 3 8–​1 1 7+​1 8 of 22 1.524 0.01
(Eliminating
crucifix joint) 1.676 0.012
1.829 0.014
1.981 0.017
2.134 0.019
2.286 0.022
 Analysis and design of precast segmental linings 121

180 m 200 m 250 m 300 m 350 m 400 m 450 m 500 m

0.004 0.003 0.003 0.002 0.002 0.002 0.002
0.005 0.004 0.003 0.003 0.002 0.002 0.002
0.006 0.005 0.004 0.003 0.003 0.003 0.002
0.007 0.005 0.004 0.004 0.003 0.003 0.003
0.007 0.006 0.005 0.004 0.004 0.003 0.003
0.008 0.007 0.006 0.005 0.004 0.004 0.003
0.009 0.008 0.006 0.005 0.005 0.004 0.004
0.010 0.008 0.007 0.006 0.005 0.005 0.004
0.009 0.008 0.006 0.005 0.004 0.004 0.004 0.003
0.011 0.01 0.008 0.007 0.006 0.005 0.004 0.004
0.013 0.012 0.01 0.008 0.007 0.006 0.005 0.005
0.016 0.015 0.012 0.01 0.008 0.007 0.007 0.006
0.019 0.017 0.014 0.012 0.01 0.009 0.008 0.007
0.023 0.02 0.016 0.014 0.012 0.01 0.009 0.008
0.026 0.024 0.019 0.016 0.013 0.012 0.011 0.009
0.03 0.027 0.022 0.018 0.015 0.013 0.012 0.011
0.008 0.007 0.006 0.005 0.004 0.004 0.003 0.003
0.009 0.008 0.007 0.006 0.005 0.004 0.004 0.004
0.011 0.01 0.008 0.007 0.006 0.005 0.005 0.005
0.013 0.012 0.01 0.008 0.007 0.006 0.006 0.006
0.016 0.014 0.011 0.009 0.008 0.007 0.007 0.006
0.018 0.016 0.013 0.011 0.01 0.008 0.008 0.007
0.021 0.019 0.015 0.013 0.011 0.01 0.009 0.008
0.024 0.021 0.017 0.014 0.012 0.011 0.01 0.009
0.009 0.008 0.007 0.006 0.005 0.004 0.004 0.004
0.011 0.01 0.008 0.006 0.006 0.005 0.005 0.004
0.013 0.011 0.009 0.008 0.007 0.006 0.005 0.005
0.015 0.013 0.011 0.009 0.008 0.007 0.006 0.005
0.017 0.015 0.012 0.01 0.009 0.008 0.007 0.006
0.02 0.018 0.014 0.012 0.01 0.009 0.008 0.008
(Continued)
122 Handbook of Precast Segmental Tunnel Lining Systems

Table 3.3 (Continued)

Ring configuration Minimum radius

Diameter Segm’ts Conn’ns Length 80 m 120 m 160 m
2.438 0.025
2.591 0.028
Category 4 11–​1 4 8+​1 9 of 25 1.676
(Eliminating
crucifix joint) 1.829
1.981
2.134
2.286
2.438
2.591
2.743
Category 5 > 14 9+​1 10 of 28 1.829
(Eliminating
crucifix joint) 1.981
2.134
2.286
2.438
2.591
2.743
2.896

3.3 DESIGN FOR PRODUCTION AND TRANSIENT STAGES

Production and transient loadings include the loading stages starting from the time of seg-
ment casting up to the time of the segment erection within the tunnel boring machine (TBM)
shield. During these phases, the internal forces and stresses from stripping (demolding),
storage, transportation and handling are used for the design of precast concrete segments.
Production and transient loads during these stages result in significant bending moment
with no or negligible axial forces.

3.3.1 Segment stripping
Figure 3.24 shows the stripping (demolding) phase that is modeled by two cantilever beams
loaded under their own self-​weights (w). The design is performed with regard to the specified
 Analysis and design of precast segmental linings 123

180 m 200 m 250 m 300 m 350 m 400 m 450 m 500 m

0.022 0.02 0.016 0.014 0.012 0.01 0.009 0.008
0.025 0.023 0.018 0.015 0.013 0.012 0.01 0.009
0.009 0.008 0.007 0.006 0.005 0.004 0.004 0.003
0.011 0.01 0.008 0.006 0.006 0.005 0.004 0.004
0.012 0.011 0.009 0.008 0.007 0.006 0.005 0.005
0.014 0.013 0.01 0.009 0.007 0.007 0.006 0.005
0.016 0.014 0.012 0.01 0.009 0.008 0.007 0.006
0.018 0.016 0.013 0.011 0.01 0.008 0.008 0.007
0.02 0.018 0.015 0.012 0.011 0.01 0.008 0.008
0.022 0.02 0.017 0.014 0.012 0.011 0.009 0.009
0.009 0.007 0.006 0.005 0.005 0.004 0.004
0.011 0.009 0.007 0.006 0.005 0.005 0.004
0.012 0.01 0.008 0.007 0.006 0.005 0.005
0.014 0.011 0.009 0.008 0.007 0.007 0.006
0.016 0.013 0.011 0.009 0.008 0.007 0.007
0.018 0.014 0.012 0.01 0.009 0.008 0.007
0.02 0.016 0.014 0.012 0.01 0.009 0.008
0.023 0.018 0.015 0.013 0.011 0.01 0.009

strengths when segments are stripped or demolded (that is, about 6 hours after casting). As
shown in Figure 3.25, the self-​weight (w) is the only force acting on the segment and the
applied load factor in ultimate limit state (ULS) is 1.4 as recommended by ACI 544.7R-​16
and ACI 533.5R-​20.

3.3.2 Segment storage
Segment stripping (demolding) is followed by the segment storage phase in the storage yard
at the precast plant where segments are stored to gain specified strength before transporta-
tion to the construction site. As shown in Figure 3.26, in the most common scheme, all seg­
ments comprising a full ring are piled up within one stack. Designers, in coordination with
the segment manufacturers, provide the distance between the stack supports considering an
124 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.18 Developed plan view with crucifix joints.

Note: Developed plan view with crucifi x joints between two adjacent segmental lining rings.

Figure 3.19 Developed plan view of consequential segment ring rotation.

Note: Developed plan of consequential ring rotation causing crucifix joints in the rectangular
system vs. staggered joints in a rhomboidal system.

Table 3.4 Example of alignment deviations using rectangular and rhomboidal segments.

Resulting alignment deviation as a function of curve radius, ring length and number of
circumferential connections per ring

Segmental ring Tunnel curve radius (m)

Segmental Considered
ring type Length (m) ring rotations 250 300 350 400 500
Control 1.524 16 0.003 0.003 0.002 0.002 0.002
Rhomboidal 1.524 6 0.010 0.008 0.007 0.006 0.005
Rectangular 1.524 7 0.021 0.017 0.015 0.013 0.010

eccentricity of ‘e’ =​4 inches (100 mm) between the locations of the stack support for the
bottom segment and the supports of above segments. This load case can be represented by
a simply supported beam loaded under its self-​weight as shown in Figure 3.27. The dead
weight of segments positioned above (F) acts on the designed segment as a concentrated
load in addition to its self-​weight (w). Therefore, corresponding load combination can be
considered as 1.4w +​1.4F (ACI 544.7R-​16, ACI 533.5R-​20).
 Analysis and design of precast segmental linings 125

Figure 3.20 Example of a 5+​1 rhomboidal segmental ring system.

Note: With three dowels designed to be installed on each segment, this ring system has a total
number of 16 dowels where relative ring rotations happen.

Figure 3.21 All possible relative ring rotations for a rhomboidal segmental ring system.
Note: Resulting in crucifix joints (more transparent layouts) between adjacent rings.
126 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.22 6+​1 segmental ring arrangement.

Note: 6+​1 ring configuration with (right figure) and without (left figure) adjusted key segment
arc angle to prevent crucifix joints.

Figure 3.23 Developed plan view between two adjacent rings after adjustments.
Note: Developed plan between two adjacent rings where a crucifix joint has been eliminated by
slight adjustment of segment geometry.

Figure 3.24 Stripping (demolding) precast segments from forms.

Note: A perspective view of a segment being stripped from formwork in the precast plant.
 Analysis and design of precast segmental linings 127

Figure 3.25 Forces acting on freshly cast segments.

Note: Schematics of loads and support conditions.

Figure 3.26 Segment stacking for storage.

Note: A side view of a full ring segments piled up in one stack in the precast yard after demolding.

3.3.3 Segment transportation
During the segment transportation phase, stored precast segments in the storage yard
are transported to the construction site and TBM trailing gear. Segments may encounter
dynamic shock loads during this phase and as shown in Figure 3.28, half or all of the seg­
ments of each ring are transported on one dolly. Wood blocks provide supports for the
segments. An eccentricity of 4 inches (100 mm) is recommended for design. Note that the
wood blocks should be installed nearly parallel to the segment axis. Similar to segment stor-
age phase (Figure 3.27), simply supported beams represent the load case of transportation
with dead weight of segments positioned above (F) and self-​weight (w) as the acting loads
on designed segment. In addition to load combination of 1.4w +​1.4F per (ACI 544.7R-​16,
ACI 533.5R-​20), a dynamic impact factor of 2.0 is recommended to be applied to the F
force for the transportation phase.
128 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.27 Forces acting on precast segments in storage.

Note: Schematics of loads and support conditions (free body diagram).

Figure 3.28 Segment stacking for transport.

Note: A perspective view of segments being transported in the tunnel toward TBM trailing gear.

3.3.4 Segment handling
Segment handling inside the precast plant and from storage yard to trucks or rail cars are
carried-​out by specially designed lifting devices such as vacuum lifters, forklifts or mechan-
ical clamping or a combination of them (Figures 3.29–​3.31). For handling by vacuum lift-
ers and mechanical clamping the same analysis and design procedure utilized for segment
stripping (demolding) should be followed. However, when segments are handled by
forklifts as schematically shown in Figure 3.32, a loading scheme similar to segment stack­
ing for storage and transportation (Figure 3.27) should be adopted. In this case, the total
 Analysis and design of precast segmental linings 129

Figure 3.29 Segment handling with vacuum lifting.

Note: Segment handling using vacuum lifting device.

Figure 3.30 Segment handling with forklifts.

Note: Segment handling using forklift equipment.

eccentricity is equal to sum of spacing of forklift contact point from the wood blocks axis
and the minimum design eccentricity of e =​4 inches (100 mm). A dead load factor of 1.4
in ultimate limit state (ULS) and a dynamic impact factor of 2.0 are recommended for this
load case (ACI 544.7R-​16, ACI 533.5R-​20). In addition, when lifting inserts are used for
segment handling, the pullout capacity of lifting inserts and concrete should be checked to
130 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.31 Segment handling with mechanical clamping.

Note: Segment handling using mechanical clamping equipment.

Figure 3.32 Schematics of design forces and eccentricities for segment handling.

Note: Schematics of design eccentricities for segment handling with the forklift.

be greater than the self-​weight of the segment under dynamic loading conditions. Pullout
capacity of concrete in this case can be calculated following the pullout failure diagrams and
equations of Figure 3.33.

3.3.5 Summary of design checks and factors for production and

transient stages
Table 3.5 presents a summary of load cases; applied dynamic impact factors, if any; and
maximum developed bending moments and shear forces for the various manufacturing,
 Analysis and design of precast segmental linings 131

Figure 3.33 Pull-​o ut capacity for segment handling.

Note: Diagram of pullout capacity of concrete during handling.
Source: adopted from ACI 318-​1 9. 

Table 3.5 Internal forces developed in segments for first four load cases in Table 3.1.
Required design checks and factors for production and transient stages

Load Dynamic Maximum unfactored Maximum unfactored

case Phase impact factor bending moment shear force
1 Stripping (demolding) ― wa 2/​2 wa
2 Storage ― w(L /​8 –​ S /​2 ) +​ F 1e
2 2
Ws
w(S 2/​2 ) +​ F 1e Wl/​2 +​F 1
3 Transportation 2.0 w(L 2/​8 –​ S 2/​2 ) +​ F 2e Ws
w(S 2/​2 ) +​ F 2e Wl/​2 +​F 2
Handling (forklift) w(L 2/​8 –​ S 2/​2 ) +​ F 2e Ws
4 2.0 w(S 2/​2 ) +​ F 2e Wl/​2 +​F 2
Handling (other methods) wa 2/​2 wa
Notes:
1
F 1 is the self-​w eight of all segments completing a ring, excluding the bottom segment.
2
F 2 is the self-​w eight of all segments carried by forklifts or placed in one carriage for transportation phase,
excluding bottom segment.

transportation and handling stages. Precast concrete segments should be sufficiently rein-
forced to withstand developed bending moments and shear forces due to these actions at
an early age. Designers should follow structural codes and guides such as ACI 318-​19, ACI
544.7R-​16 and ACI 533.5R-​20 for calculating the bending strength (Mn) and check the
design strength vs. developed bending moments and shear forces.

3.4 DESIGN FOR CONSTRUCTION STAGES

Construction loads include the load applied during segment erection, boring and grouting
stages. They include loads induced due to accidental TBM thrust jack removal, TBM jack-
ing thrust loads on the circumferential ring joints and the pressures during the grouting
operation exerted against the exterior of the completed rings. Dowels in the circumferential
joints are designed to resist loads due to accidental jack removal, while reinforcement (steel
bar or fiber) in precast concrete segments are designed to resist significant bursting and
132 Handbook of Precast Segmental Tunnel Lining Systems

spalling tensile stresses that develop along the circumferential joints due to the advancing of
the TBM. The segments should also be able to resist the axial forces and bending moments
that develop when the annular space between the segments and the ground is pressure-​filled
with grout. Included is the primary backfilling of the tail skin void and the secondary grout-
ing that is needed in case a complete contact of the lining with the ground has not been
achieved through the primary grouting.

3.4.1 Accidental removal of TBM thrust jack forces

Construction of segmental lining inside the TBM shield starts with ring erection. During
this phase each segment is erected often by the vacuum erector and is held in its place by
the TBM thrust jacks one after another until completion of full ring erection. The critical
load during this phase occurs due to accidental removal of thrust jack forces leaving the
precast segment unsupported. Consequently, the dowel system must carry the self-​weight of
the connecting segment as well as the gasket reaction force to prevent accidental ejection/​
bounce back and the segment falling into the ring build working area. The key segment is
usually more vulnerable to this scenario, which is often located at the crown of the tunnel.
Other segments though larger are supported proportionally by a higher number of dowels,
and therefore, there is practically no difference between presenting this load case for key
segment or other ordinary segments.
For the strength design or ultimate limit state (ULS), as presented in Figure 3.34 by the
red color, the acting loads are the self-​weight of the connecting segment in the vertical dir-
ection (γdW) and the load due to the gasket pressure in the horizontal direction (γgPg). The
resisting loads are the ultimate pullout force of the dowel in horizontal direction (Pu) and
the ultimate shear force of the dowel in vertical direction (Vu).
In this load case scenario, the segment tends to rotate at the circumferential face down-
ward about a turning pivot point, which is the lowest point of segment under the dowel at
the intrados. By taking equilibrium of moments around this pivot point, pullout force of
dowels can be determined using following equation.

Pu (h/​2) =​ γd Wkey b/​2 +​ γg Pg,ULS (h –​ dg) (3.2)

where
γd is the load factor segment self-​weight
γg is the load factor for gasket reaction force
Wkey is the segment self-​weight
Pg,ULS is unfactored reaction force of gasket on segment at ultimate limit state of dowel
b is the width of segment
h is the thickness of segment
dg is the distance of gasket centerline from the segment extrados.

Therefore, the ultimate pullout force of the dowel (Pu) is calculated as

Pu =​[γd Wkey b/​2 +​ γg Pg,ULS (h –​ dg)]/​(h/​2) (3.3)

The ultimate shear force of the dowel (Vu) is simply calculated as

Vu =​ γd W (3.4)
 Analysis and design of precast segmental linings 133

Figure 3.34 Acting and resisting forces on segment at the time of accidental TBM thrust removal.
Note: Schematics of acting forces (red) and resisting forces (green) for the load case of acci-
dental removal of TBM thrust jack forces.

Pu should be compared with the reduced nominal pullout strength of dowel or ϕPNt,ULS
and Vu should be compared with the reduced nominal shear strength of dowel or
ϕPNv,ULS.
For this load case the mechanical properties of dowels must be obtained from the dowel
supplier on a case-​by-​case basis. For example, as shown in Figures 3.35 and 3.36, nominal
ultimate limit state (ULS) pull-​out capacity of the dowel (PNt,ULS), displacement of dowel at
ULS under pull-​out (dc,ULS), and nominal ULS shear capacity of the dowel (PNv,ULS) are 22.5
kips (100 Kn), 0.375 inches inches (9 mm) and 40.5 (180 Kn), respectively. Estimated
gap between gasket profiles (dg,ULS) corresponding to ULS displacement of dowel (in this
example dc,ULS =​0.375 (9 mm), as shown in Figure 3.37, can be determined as 0.60 inches
(15.4 mm) considering a linear approximation of displacement at the circumferential
joint. Under such large gap between gasket profiles, effect of the gasket reaction force, as
shown for example in Figure 3.38 for a typical gasket load-​gap diagram, is completely
dissipated.
Results of this design check should prove that the ultimate pull-​out force of dowel (Pu) is
less than the reduced ultimate pull-​out resistance of dowel (ϕPNt,ULS), and the ultimate shear
force developed in dowels (Vu) is less than the reduced ultimate shear resistance of dowel
(ϕPNv,ULS). Otherwise, a higher dowel capacity should be adopted.
134 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.35 Diagram of dowel pull-​o ut force from segment vs. displacement.

Note: Example of pull-​o ut tests on a typical dowel system for precast segments.

Figure 3.36 Shear test on dowels installed on segments.

Note: Example of shear tests on a typical dowel system for precast segments.
 Analysis and design of precast segmental linings 135

Figure 3.37 Linear approximation of displacements at circumferential joints.

Note: Linear approximation of displacements at the circumferential joint for the load case of
accidental removal of thrust jack forces.

3.4.2 TBM thrust jack forces

After assembly of a complete ring, the TBM advances by thrusting against the most recent
assembled ring, as shown in Figure 3.39. As part of this process, the TBM jacks bear against
the jacking pads placed along the exposed circumferential joint. The thrust jacks push the
shield forward in the direction of drive against the friction forces in the ground and against
the necessary support pressure (with face support). The shoes of the thrust jacks are placed
against the last ring to be installed, which thus serves as abutment (Maidl et al., 2012).
Schematics of a simplified disturbance area of strut under TBM jack shoes are presented
in Figure 3.40. High compression stresses develop under the jacking pads and result in
the formation of significant bursting tensile stresses deep within the segment. Furthermore,
spalling tensile forces also act between adjacent jack pads along the circumferential joint.
In the early stages of design, different methods can used for estimating TBM thrust due
to the various geologic materials that can be encountered. For tunnel excavation through
rock, TBM thrust can be estimated by the summing forces required to advance the machine.
These forces include the forces necessary for boring through the rock, countering the fric-
tion between the surface of the shield and the ground, and the hauling of trailing gears.
Methods found in Fukui and Okubo (2003) and Rostami (2008) can be used to evaluate
the rock thrust based on rock strength, tunnel diameter, and cutter characteristics (Bakhshi
and Nasri, 2013a). For soft ground tunneling applications, the method presented by JSCE
(2007) can be used to calculate penetration resistance based on earth or slurry pressure that
acts at the cutting face.
Once the required machine thrust has been estimated for the ground conditions, the
average thrust force per jack pair is determined simply by dividing the required machine
thrust by the number of jack pairs. On sharp curves, the machine thrust is higher on the
convex side of the curve than on the concave side. The higher thrust has to be accounted
for in the design. A simple technique can be used to account for the increased loading on
the convex side by doubling the jacking loads. Some design methods require an additional
20% increase in estimated jacking loads in the absence of accurate loads of TBM in early
design stage.
136 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.38 Example of load-​g ap diagram for typical gasket profile.

Note: This load case is considered as an extreme even limit state due to accidental nature of the
load case, and therefore, the Load Factors for acting loads can be taken conservatively as 1.05.
The unit weight of segments should be obtained from numerous tests on the samples taken from
the segment concrete batch in the pre-​p roduction stage and taking strength reduction factor for
dowel pullout as ϕ =​ 0.80.

Whenever the TBM’s total thrust is available, maximum thrust jack forces are known
and therefore cannot be exceeded. This important input is given by the TBM manufacturer
often by two values of nominal machine thrust and maximum exceptional machines thrust.
Nominal machines thrust is the maximum operational thrust forces applied on all jacks
required for boring and advancing the machine for the worst-​case ground condition while
maximum exceptional machines thrust are only used in exceptional cases when TBM is
stuck and cannot advance with normal thrust conditions and all exceptional machine thrust
needs to be mobilized for further TBM advancement.
Forces required for ultimate failure of precast segments in this load case are much higher
than maximum TBM thrust jack forces that are normally exerted on segments. Accordingly,
precast segments in this load case should be designed for a serviceability limit state (SLS)
of cracking. In addition to high thrust jack forces, eccentricity of jack forces with respect
 Analysis and design of precast segmental linings 137

Figure 3.39 Rendered 3D view of TBM thrust jacks pushing against segments.

Note: Thrust jacks pushing on circumferential joints.
Source: Groeneweg, 2007. 

Figure 3.40 Stress distribution under TBM thrust jack forces.

Note: Schematics of a simplified disturbance area of strut under TBM jack shoes.
Source: Groeneweg, 2007. 

to segment thickness centerline and asymmetric positioning of jacking pads with respect to
segment ends in the circumference of the ring (thrust jacks are irregularly spaced or segment
sizes are unequal) are considered as other major causes of cracking during the thrust phase
(SIG WG N. 2, Report N. 1 –​Damages of Segmental Lining (2019). Limiting magnitude of
jack forces, reducing the eccentricity of jack forces along the ring thickness and uniform dis-
tribution of jack pads with respect to segment ends can be considered as mitigation methods.
138 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.41 Radial section cut through a circumferential joint.

Note: Example of a circumferential joint for a segment with 11.8 inches (300mm) thick segment
with details of joint profile.

Nonetheless, in practice almost none of these methods can be guaranteed. Therefore, such
extreme conditions should be considered during the design procedure of segments.
One very important detail for this load case is the detail of circumferential joints with
gasket recess on the exterior side (extrados), and stress relief recess on the interior side
(intrados) of tunnel. For example, as shown Figures 3.41 and 3.42, while the total thick­­
ness of segment in this example is 11.8 inches (300 mm), maximum thickness of segment in
contact with TBM thrust pads can be only 7.9 inches (200 mm). Consequently, high eccen-
tricities of jack pads in radial direction toward extrados may result in reduced contact area
with segments and increased pressure on the contact zone. Nominal and maximum excep-
tional eccentricities of jack pad in the radial direction are recommended as en =​±1.20 inches
(30mm) and emax =​± 1.6 inches (40 mm), respectively.
The design procedure must consider all possible combinations of nominal and max-
imum exceptional thrust jack forces with nominal and maximum exceptional thrust
pad eccentricities. This results in a total number of 10 jack force scenarios. Each jack
force scenario is defined by an appropriate load factor of 1.00, 1.05 or 1.20 in order to
 Analysis and design of precast segmental linings 139

Figure 3.42 Transverse view of a circumferential joint.

Note: Example of a circumferential joint for a segment with 11.8 inches (300 mm) thick segment
with thrust pads positioning along the circumferential joint.

Table 3.6 Eccentricities and load factors for TBM thrust force scenarios.
TBM thrust force scenarios, eccentricities, load factors, and factored pressures for
load case 6

Factored pressure
Load Unfactored jack force Eccentricity Load factor Contact area on contact area,
case on each pad, F p (kN) e (mm) γ A contact (mm 2) P contact (MPa)

1 F n/​n thrust pads e =​ 0 γ =​ 1.20 A e0 γ F p/​ A contact

2 F n/​n thrust pads e n =​–​3 0 γ =​ 1.20 A en-​ γ F p/​ A contact

3 F n/​n thrust pads e n =​ +​3 0 γ =​ 1.20 A en+​ γ F p/​ A contact

4 F n/​n thrust pads e max =​–​4 0 γ =​ 1.05 A emax-​ γ F p/​ A contact

5 F n/​n thrust pads e max =​ +​4 0 γ =​ 1.05 A emax+​ γ F p/​ A contact

6 F max/​n thrust pads e =​ 0 γ =​ 1.05 A e0 γ F p/​ A contact

7 F max/​n thrust pads e n =​–​3 0 γ =​ 1.05 A en-​ γ F p/​ A contact

8 F max/​n thrust pads e n =​ +​3 0 γ =​ 1.05 A en+​ γ F p/​ A contact

9 F max/​n thrust pads e max =​–​4 0 γ =​ 1.00 A emax-​ γ F p/​ A contact

10 F max/​n thrust pads e max =​ +​4 0 γ =​ 1.00 A emax+​ γ F p/​ A contact

consider the variability in the prediction of loads and eccentricities. For example, the
likelihood of exceeding maximum exceptional eccentricity and maximum exceptional
machine thrust at the same time is zero, and therefore, a load factor of 1.0 is applied
on this thrust force scenario. Because eccentricities have an impact on the contact area
of thrust pads and segments, for all scenarios, contact areas and resulting pressure on
contact zone should be compiled in a table such as Table 3.6 as the first step toward the
design for this load case.
140 Handbook of Precast Segmental Tunnel Lining Systems

The eccentricity toward the extrados (upward in Figure 3.42) is shown with a positive
sign in this table and the eccentricity toward intrados (downward in Figure 3.42) is shown
with a negative sign. One quick conclusion from this table is that the maximum contact
pressure applied on circumferential faces by TBM thrust jacks which provides a basis for
unfactored specified compressive strength of the concrete mixture.
Different analysis and design methods available for this load case, including ACI 318-​19
equations for bursting forces, DAUB:2013 formulas, Iyengar Diagram (1962), and two and
three-​dimensional finite element simulations are presented in the following.

3.4.2.1 Simplified equations

For post-​tensioned anchorage zones of prestressed concrete sections, structural concrete
codes such as ACI 318-​19 allow the use of simplified equations [Equation (3.5)] to deter-
mine the bursting force, Tburst, and the centroidal distance from the face of the section, dburst.
These simplified equations are used to obtain the forces and stresses developed in the cir-
cumferential joints due to TBM advancement. DAUB:2013 recommends similar equations
[Equation (3.6)] specifically for the design of tunnel segments.

 h 
ACI 318 − 18: Tburst = 0.25Ppu  1 − anc  ; dburst = 0.5(h − 2eanc (3.5)
 h 

 hanc 
DAUB: 2013 Tburst = 0.25Ppu  1 − ; dburst = 0.4 ( h − 2eanc ) (3.6)
 h − 2eanc 

This load case and corresponding parameter are schematically shown in Figure 3.43.
Equations (3.5) and (3.6) represent the radial bursting stresses in the circumferential joints.
These equations are also applicable to tangential bursting stresses developed in the circum-
ferential joints.
Reinforcing bar or fiber reinforcement is designed to resist the bursting stresses developed
by jacking forces. Equations (3.7) and (3.8) have been adopted to determine the required
area (As) of reinforcing bars with a yield stress of fy for a reinforced concrete segment.

Figure 3.43 Bursting tensile forces and corresponding parameters.

Note: Bursting tensile forces and corresponding parameters recommended by ACI 318-​1 9.
 Analysis and design of precast segmental linings 141

Tburst = φfy As _ radial for radical reaction (3.7)

Tburst = φfy As _ tangential for tangential direction (3.8)

High compressive stresses can be developed under the jacking pads due to the TBM thrust
jacking forces. These compressive stresses, σc,j, can be estimated using Equation (3.9).

Ppu Ppu
σc, j = = (3.9)
Aj al hanc

Because only part of the circumferential segment face is in contact with the pads, the allow-
able compressive stresses (f′c) can be factored to account for the strength of a partially pres-
surized surface. ACI 318-​19 specifies the formula used for designing the bearing strength of
concrete [Equation (3.10)] with a partially loaded segment face. DAUB:2013 recommends
a similar formula for designing tunnel segment faces.

Ad at ( h − 2eanc )
fco′ = 0.85fc′ = 0.85fc′ (3.10)
Aj al hanc

3.4.2.2 Iyengar diagram

1 The analytical method of the Iyengar diagram (1962) has also been used for calculating
the bursting tensile stresses for the design of tunnel segments (Groeneweg, 2007). Similar
to previous methods, the extent of load spread, and the resulting magnitude of tensile
stresses depend on the dimensions of the loaded surfaces, β, and the final spreading
surfaces, a, shown in Figure 3.44. Using this approach, the bursting tensile stresses (σcx),
which vary significantly from the face that TBM jacks bear against toward the centerline
of segment, are determined as a fraction of the fully spread compressive stress (σcm =​ F/​
ab). Reinforcing bars are designed for the bursting tensile stresses shown in the diagram.
Total bursting force can be obtained as the integration of stresses or the area under the
curve and required reinforcing bar area is determined by Equations 3.7 and 3.8.

3.4.2.3 Linear elastic finite element analysis

As shown in Figure 3.45, in addition to the bursting stresses under the jacking pads, spalling
stresses develop in the areas between the jacking pads and in the areas between the jack-
ing pads and the end faces of segments due to the concentration of the jacking forces.
Analytical design methods for spalling stresses have not been provided in any of the previ-
ously reviewed approaches; however, analyzing the problem with a three-​dimensional finite
element method (FEM) is appropriate. While both linear elastic and nonlinear FEM simula-
tions can be performed, the latter is considered suitable for the service design as nonlinear
analyses capture the non-​linear response of the materials after cracking with respect to
crack opening. In conventionally reinforced concrete segments, TBM thrust jack forces,
among other methods can be accurately analyzed by elastic finite element method (FEM)
for ultimate limit state (ULS) design.
142 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.44 Iyengar diagram (1962) for determining bursting tensile stresses.

Note: Bursting stresses developed in segments as a function of the fully spread compressive
stress (σ cm), the size of final spreading surfaces (a) and the distance from loading surface (x).
Source: Groeneweg, 2007. 

Figure 3.45 Spalling and bursting stresses in segments.

Note: Spalling and bursting stresses in segment joints due to jack thrust forces (Blom, 2002).
Source: Blom, 2002. 

As shown in Figure 3.46, this load case can be simulated by modeling typical segments
of one ring or two adjoining rings. The jacking forces are applied along the contact area
between the jacking pads and the segment face. Recesses due to the gasket and stress relief
grooves are modeled between two segments to simulate the transfer of force through a
reduced cross-​section. With this approach, the translational degrees of freedom are fixed in
all directions behind the previously installed segment. Figures 3.47 and 3.48 show typical
 Analysis and design of precast segmental linings 143

Figure 3.46 Three-​d imensional finite element model.

Note: Three-​d imensional FEM model for load case of TBM thrust jack forces..

Figure 3.47 Bursting and spalling tensile stresses developed in segments.

Note: Example of bursting and spalling tensile stresses developed in segments due to TBM thrust
jack forces (Bakhshi and Nasri 2013b, 2014d): transverse stresses. Note size of contact area of
jacking pad is 7.8” (0.2 m) by 30” (0.8 m) and segment thickness is 15.7” (0.4 m).
144 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.48 Bursting and spalling tensile stresses developed in segments.

Note: Example of bursting and spalling tensile stresses developed in segments due to TBM thrust
jack forces (Bakhshi and Nasri 2013b, 2014d): radial stresses. Note size of contact area of jacking
pad is 7.8” (0.2 m) by 30” (0.8 m) and segment thickness is 15.7” (0.4 m).

results consist of the transverse and radial bursting stresses under the jack pad and the
spalling stresses in the areas between the jacking pads as a result of an elastic finite element
analysis. Examples of typical compressive stresses distribution due to this load case are
shown in Figure 3.49 (Bakhshi and Nasri, 2013b, 2014d).
Results from the three-​dimensional FEM simulation indicate that spalling tensile stresses
between the jack pads, and the jack pads and end faces, can be more significant than the trans-
verse bursting tensile stresses under the jacking pads (Bakhshi and Nasri, 2013b). Precast
tunnel segments should be designed to withstand these high tensile stresses. Reinforcement
bars or fibers are designed for the tensile forces determined by the integration of stresses
through the tensile zone, similar to Iyengar diagram (1962) method.

3.4.2.4 3D non-​linear finite element method simulation

When segments are only reinforced with fibers the spalling stresses are likely to exceed
the tensile strength. Therefore, a 3D non-​linear FEA (NFEA) simulation, such as the one
shown in Figures 3.50–​3.52, must be performed to validate limitation of the crack width
dimensions to the allowable values set forth by standards, guidelines and project specifica-
tions or design criteria. A 3D NFEA program such as DIANA can be used, which is cap-
able of adopting FRC materials model in the format of multi-​linear stress–​strain curves
 Analysis and design of precast segmental linings 145

Figure 3.49 Typical compressive stresses developed in tunnel segments.

Note: Typical compressive stresses developed in tunnel segments due to TBM thrust jack forces
(Bakhshi and Nasri, 2013b, 2014d). Size of contact area of jacking pad is 7.8” (0.2 m) by 30” (0.8
m) and segment thickness is 15.7” (0.4 m).

Figure 3.50 Fine mesh used for 3D NFEA simulation of TBM thrust jack forces.
Note: Example of a 3D non-​linear FEM simulation for the load case of TBM thrust jack forces
applied on fiber-​reinforced concrete segment: fine mesh used for simulation.

shown in Figure 3.53. These parameters are used in conjunction with the total strain
crack constitutive model to introduce and assign a nonlinear concrete material model to
the segments.
As for geometry, the exact shape of segments in rectangular, rhomboidal or parallelogram
shape with potential longitudinal joint taper angle (8.5o in presented example) and
exact positioning of thrust pads in actual asymmetric pattern with respect to segment cen-
terline and segment ends are imported to 3D NFEA (non-​linear finite element analysis)
program. Recesses along the segment thickness due to the gasket and stress relief grooves
(similar to shown in Figure 3.50) should be accurately modeled. Boundary condition for
146 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.51 Transverse tensile stresses as results of 3D NFEA simulation of TBM thrust jack
forces.
Note: Example of a 3D non-​linear FEM simulation for the load case of TBM thrust jack forces
applied on fiber-​reinforced concrete segment: transverse tensile stresses.

Figure 3.52 Crack width dimensions as results of 3D NFEA simulation of TBM thrust jack forces.
Note: Example of a 3D non-​linear FEM simulation for the load case of TBM thrust jack forces
applied on fiber-​reinforced concrete segment: crack width dimension. Size of contact area of
jacking pad is 6.8 inches (0.17 m) by 24 inches (0.6 m) and segment thickness is 12 inches (0.3 m).

three non-​loaded end faces of segments are set by a compression-​only constraint (blocking
movement of joints) normal to the joint faces.
Segment is modeled by solid linear hexahedron (dominant) elements and meshed with
element grading size varying from 0.31 inches (8 mm) near the thrust pads to 8.9 inches
(200 mm) towards the opposite boundary parallel to the thrust pads. The fine meshed model
(1.8 million hexahedron dominant elements in presented example) with higher density near
the jacking forces ensures that the model covers the crack initiation and propagation in
 Analysis and design of precast segmental linings 147

Figure 3.53 Specified mechanical properties of fiber-​reinforced concrete.

Note: Example of specified mechanical properties of fiber-​
reinforced concrete for precast
segments.

various dimensions of the segment including the thickness. Uniform surface pressure with
the intensity shown in Table 3.6 should be used to impose the effect of jacking forces acting
on the pads. In order to perform a nonlinear analysis, an incremental loading method
was introduced to capture the evolution of the plasticity in the model. The accuracy of the
results must be ensured with simultaneous convergence criteria based on the quasi-​Newton
iterative method.
In the presented example (Bakhshi and Nasri, 2020), as shown in Figure 3.53, tensile
cracking strength of fiber-​reinforced segments is 403 psi (2.77 MPa) and its post-​cracking
tensile strength is 233 psi (1.6 MPa). Figure 3.50 shows the transverse tensile stresses, and
Figure 3.51 shows the crack width dimension due to jacking force. As expected, cracks are
observed at the regions with highest spalling stresses in the linear analysis, i.e. between jack-
ing pads, and jacking pad and edge of segment. In this example the size of contact area of
jacking pad is 6.8 inches (0.17 m) by 24 inches (0.6 m).

3.4.3 Tail skin back grouting pressure

Tunnel rings are assembled within the shield of the TBM. At this location, the excavated
diameter of the tunnel is larger than the external diameter of the tunnel ring. As shown in
Figure 3.54, a tail void is created between the ground and the tunnel lining. This load case
is generated by back grouting or filling of the annular space using semi-​liquid grouts under
high pressure to control and restrict settlement at the ground surface as well as to ensure
complete contact between the ring and the ground. Grouting material generally consists of
sand, water, cement and several additives such as bentonite or plasticizers. This material is
not actually a liquid per se, but it does have very low shear yield strength of 0.003 to 0.015
psi (20 to 100 Pa), prior to its hardening. When grout flows, it can be continuously injected
into the annular space behind the TBM by means of grout pipes routed through the tail skin.
To enter the annulus space created around the tunnel lining, the grout pressure needs to
be greater than the water pressure surrounding the liner. The grout pressure should also
typically be less than the overburden pressure to prevent push off of the soil, heave, hydro-
jacking, or all of these. Grouting models (Zhong et al., 2011) can be developed that predict
148 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.54 Backfilling of tail skin void.

Note: Backfilling of tail skin void.
Source: Guglielmetti et al., 2007. 

the optimal grout pressure by considering the combined effects of groundwater level, plas-
ticity of grout, rate of advancement of the TBM, and the filling rate of the tail void. Using
Equation (3.11), the equivalent specific weight of the grout is determined by taking the
equilibrium condition between the upward component of the total grout pressure, tunnel
self-​weight, and tangential component of the grout shear stresses (Groeneweg, 2007). These
forces are schematically shown in Figure 3.55.

π 2
De bρeq = πDe hbρconcrete + 2De bτ yield (3.11)
4

The vertical gradient of the radial grout pressure between the crown and the invert of the
tunnel is determined by Equation (3.12).

∆Pg ,invert = ρeq ⋅ De

(3.12)

Note that AASHTO DCRT-​1-​2010 specifies that 10 psi (69 kPa) above the groundwater
pressure is the maximum permissible grouting pressure applied to the segmental ring.
However, a maximum permissible grouting pressure of up to 22 psi (150 kPa) above the
maximum groundwater pressure has been also considered (Ninić and Meschke, 2017). For
this load case, the lining is evaluated in a cross-​sectional plane perpendicular to the longitu-
dinal direction of the tunnel and modeled as a solid ring with a reduced flexural rigidity to
account for the segment joints. Because the lining is surrounded completely by semi-​liquid
and fresh grout materials at this point, no interaction is taken into account between the ring
and ground.
As shown in Figure 3.53, the back-​grouting load condition is modeled by applying radial
pressure varying linearly from the minimum grout pressure at the crown to the maximum
 Analysis and design of precast segmental linings 149

Figure 3.55 Forces and definitions for tail void back-​g routing.

Note: Forces and definitions for load case of tail void back-​g routing.
Source: Blom, 2002. 

Figure 3.56 Analysis and design checks for tail skin grouting –​ model.
Note: Example of analysis and design checks for the load case of tail skin grouting: geometry of
segmental ring model and imposed grouting pressure.

grout pressure located at the invert of the tunnel. The self-​weight of the lining and the grout-
ing pressure are the only loads applied to the tunnel lining at this stage of the analysis. For
a load combination of self-​weight and grout pressure, a load factor of 1.25 is suggested
to both loads (ACI 544.7R-​16, ACI 533.5R-​20). Analysis can be performed using general
structural analysis packages or 2D/​3D FEM model of the segmental ring with the exact
geometry of segments and joints, and with joints modeled by the Janssen (1983) model
(Figure 3.74). Because the lining is surrounded completely by a semi-​liquid fresh grout with
negligible interaction with the ground, an interface element with a very low (practically
zero) spring stiffness should be used for simulation of this interaction.
150 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.57 Analysis and design checks for tail skin grouting –​ results.
Note: Example of analysis and design checks for the load case of tail skin grouting: bending
moments in the ring as a result of FEM analysis.

As a result of analysis for this load case, significant axial forces and limited bending
moments (Figure 3.57) are usually developed in the lining. Precast segments are designed
for combined maximum bending moments and axial forces. As shown in Figure 3.58, for
the design checks, these results are compared with the axial force-​bending moment (M-​N)
interaction diagram.

3.4.4 Localized back-​g routing (secondary grouting) pressure

Localized back-​grouting, also known as secondary or check grouting, is performed through
holes that are manufactured into the segments. As shown in Figure 3.59, segments can be
fitted with grout sockets that are screwed into position and remain closed with non-​return
valves and plastic covers during the ring installation process. Prior to the introduction of
modern pressurized face machines, this grouting method was used to fill the annulus; how-
ever, the collapse of unstable ground into the annular space can generate significant settle-
ments. As such, this method is now primarily used for secondary grouting in closed face
tunneling.
To model the effects of secondary grouting, this load case consists of the forces applied
to segments to verify whether the annular gap has been closed, which is similar to the tail
skin back-​grouting method when only one of the grouting pipes is used. Following the ITA
WG 2 (2000) and ITA WG 2 (2019) guidelines, this load case can be simulated using the
force distribution shown in Figure 3.60. Using this procedure, secondary grouting pressure
is applied on one-​tenth to one-​sixth of lining perimeter on the crown.
As for the most simplified method, the lining for this load case can be modeled in the
cross-​sectional plane perpendicular to the longitudinal direction of the tunnel using a solid
ring with a reduced flexural rigidity to represent the segment joints. A more accurate simu-
lation includes a 3D FEM model of the segmental ring created with the exact geometry of
 Analysis and design of precast segmental linings 151

Figure 3.58 Analysis and design checks for tail skin grouting –​ design check.
Note: Example of analysis and design checks for the load case of tail skin grouting: comparison
with M-​N diagram.

segments and joints, and with joints modeled by the Janssen model, similar to Figure 3.56.
Because secondary grouting occurs long after the primary grouting materials have cured, it
can be assumed that tunnel lining is in full contact with the surrounding ground except in
the local area where the secondary grouting is to be performed. To simulate the boundary
condition for this case, the interaction between lining and surrounding ground or primary
hardened grout can be modeled using radial springs with the segments supported radially.
Linear translational springs have been used to represent this type of interaction.
The method described by USACE EM 1110-​2-​2901 (1997) can be used, as an example,
to determine the spring stiffness per unit of exterior tunnel surface. For example, consider-
ing a primary grout with f’c of 450 psi (3 MPa) and a modulus of elasticity of 43.5 ksi (300
MPa), radial spring stiffness of 368 lb/​in3 (100,000 kN/​m3) is a reasonable assumption for
a 20 ft (6 m) diameter tunnel. Using the same grout pressure on the crown as for the pre-
vious load case and with the radial spring stiffness, the bending moments and axial forces
developed within the lining can be determined from the localized grouting operation. This
loading case often results in small incremental axial forces with large bending moments
(Figures 3.61 and 3.62). Precast segments are designed for this load case using axial force-​
bending moment interaction diagrams.

3.4.5 TBM back-​up load

TBM back-​up load is the self-​weight of the back-​up equipment train behind the TBM shield,
shown by schematics in Figure 3.63, applied as a concentrated variable load onto the seg­
mental precast concrete lining. In general, TBM back-​up load may not be a critical or
governing load case. However, in sub-​sea and river crossing projects, to control buoyancy,
an additional weight should be provided by the back-​up system inside the tunnel after
152 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.59 Secondary grouting of segmental tunnel linings.

Note: Secondary grouting through segment grout hole.
Source: Gugliemetti, 2007. 

ring installation and before installation of precast buoyancy unit. TBM drawings, such
as Figure 3.64, show that this back-​up load is applied on specific locations in cross section
on contact areas with tunnel lining intrados. In addition, longitudinal TBM drawings
(Figure 3.65) reveal that the back-​up load is applied longitudinally (with respect to the tun­
nel direction) on two wheels on each ring. For this load case, two-​dimensional finite element
method (FEM) analysis is often sufficient, with the assumption that backup load is applied
uniformly on this area.
Typical analysis results including deformations, axial forces, bending moments and shear
forces are shown in Figures 3.66–​3.69. Resulting factored bending moments and axial forces
for all critical cases (for example, shallow cover and deep tunnel) are compared with axial
force-​bending moment interaction diagram of segmental lining. Results should confirm that
precast concrete tunnel segments can withstand axial forces and bending moments induced
 Analysis and design of precast segmental linings 153

Figure 3.60 Modeling localized grouting pressure of segmental tunnel linings.

Note: Modeling localized grouting pressure applied over one-​t enth of the lining.

Figure 3.61 Analysis results and design checks for secondary grouting –​ results.
Note: Example of analysis results and design checks for the load case of secondary grout-
ing: bending moments as result of FEM simulation.
154 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.62 Analysis results and design checks for secondary grouting –​ design check.
Note: Example of analysis results and design checks for the load case of secondary grout-
ing: comparison with M-​N diagram.

Figure 3.63 Groene Hart Tunnel –​ schematics of TBM, back-​u p train and tunnel lining.
Note: Schematics of TBM, backup train (gantries), and tunnel lining at the Groene Hart Tunnel
as segmental lining is being erected inside the TBM.
Source: Adopted from Talmon and Bezuijen, 2013. 

by TBM back-​up load or otherwise design should be modified. Punching shear strength of
the segmental lining should be also assessed following concrete design codes such as ACI
318-​19 and EN 1992-​1-​1:2004.
For the TBM back-​up load consideration, the early stage strength (setting time) of the tail
void grout and the TBM advance rate are also critical. The advance rate can be limited to
the tail void grout strength needed for support of the first and second gantries.
 Analysis and design of precast segmental linings 155

Figure 3.64 Example TBM Tunnel Project –​ sections behind the shield.

Note: Sections behind the shield through the heaviest part of backup train (gantries) where back-​
up loads applied on ring before backfilling: cross-​s ection.

3.5 DESIGN FOR FINAL SERVICE STAGES

The final service stages are represented by the long-​term loads imposed on the lining from
the ground; groundwater; surcharges; and other factors, which generally are specific to
the particular tunnel. Hoop force transfer along the longitudinal joints between segments,
additional distortion and other load cases specific to projects are also considered during
this stage. Some common load cases specific to projects include earthquake, fire, explosion,
adjacent tunnels and longitudinal bending moments. For the most part, service loads gener-
ally result in axial forces and bending in the segmental lining. With the hoop force transfer
in the longitudinal joints, significant bursting tensile stresses can be developed.

3.5.1 Earth pressure, groundwater and surcharge loads

Precast concrete segments are employed to withstand various loads from vertical and hori-
zontal ground pressure, groundwater, self-​weight, surcharge and ground reaction loads.
In accordance with load and resistance factor design (LRFD) principles and as explained
in Sections 3.1.1, 3.1.2 and 3.1.3, load factors and load combinations shown in Table 3.1
must be used to compute the ultimate limit state (ULS) and serviceability limit state (SLS).
Methods for analysis of segmental tunnel linings are presented in accordance with stand-
ards and guidelines from Europe, Asia and America (Bakhshi and Nasri, 2014a). The effect
of ground, groundwater and surcharge loads on segments is analyzed using elastic equations,
156 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.65 Example TBM Tunnel Project –​ sections behind the shield.

Note: Sections behind the shield through the heaviest part of back-​u p train (gantries) where
back-​u p loads applied on ring before backfilling: longitudinal section.

beam-​spring models, finite element methods (FEM) and discrete element methods (DEM).
Other acceptable methods of analysis include Muir Wood’s (1975) continuum model with
discussion from Curtis et al. (1976), Duddeck and Erdmann’s (1982) model, and an empir-
ical method based on tunnel distortion ratios (Sinha, 1989; Deere et al. 1969) that was
originally developed by Peck (1969). The results of these analyses are used to specify the
concrete strength and reinforcement. Reduction in bored tunnel segmental lining moment
of inertia also known as reduced flexural rigidity, where relevant, will be in accordance with
Muir Wood’s (1975) proposed method as presented in Equation (3.13).

Ir =​ Ij +​ (4/​n)² × I (3.13)

where
Ij is moment of inertia at joint (taken as zero in the design)
I is the moment of inertia for nominal lining thickness
n is the number of segments per ring excluding key segment (n ≥ 4).

3.5.1.1 Elastic equation method

The elastic equations method (ITA WG 2, 2000; JSCE, 2007) is a simple method for cal-
culating member forces of circular tunnels. As shown in Figure 3.70, the load distribution
 Analysis and design of precast segmental linings 157

Figure 3.66 Finite element analysis for TBM back-​u p load case –​ deformations.
Note: Typical results of FEM analysis for TBM back-​u p load case: deformations. Units are inches,
kips and kips-​feet.
Source: ACI 533.5R-​2 0. 

Figure 3.67 Finite element analysis for TBM back-​u p load case –​ axial forces.
Note: Typical results of finite element analysis for TBM back-​u p load case: axial forces. Units are
inches, kips and kips-​feet.
Source: ACI 533.5R-​2 0. 
158 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.68 Finite element analysis for TBM back-​u p load case –​ bending moments.
Note: Typical results of finite element analysis for TBM back-​u p load case: bending moment.
Units are inches, kips and kips-​feet.
Source: ACI 533.5R-​2 0. 

Figure 3.69 Finite element analysis for TBM back-​u p load case –​ shear forces.
Note: Typical results of finite element analysis for TBM back-​u p load case: shear forces. Units
are inches, kips and kips-​feet.
Source: ACI 533.5R-​2 0. 
 Analysis and design of precast segmental linings 159

Table 3.7 Equations to calculate radial and tangential springs.

Equations for calculating radial and tangential springs representing the interaction
between ground and tunnel

Item Guideline Bending moment Axial force

1 USACE EM K r =​ E r/​( R (1 +​ ν)) k t =​ 0.5 k r/​( 1+​ν )
1110-​2 -​2 901:1997 [Equation (3.14)]
2 Austrian (ÖVBB 2011) K r =​ E s/​R k t =​ 0
3 Groeneweg (2007) K r =​ E s/​R k t =​ 0
4 JSCE 2007 K r =​ Not available k t =​ 1/​3 K r
5 RTRI 2008 K r =​ varies (see RTRI 2008) k t =​ Not available
6 AFTES-​WG7:1993 K r =​ E r/​( R (1+​ν )) k t =​ Not available
7 DAUB:2005 K r =​ E s/​R k t =​ 0

Figure 3.70 Distribution of loads used in elastic equations.

Note: Distribution of loads used in elastic equations method.
Source: Adopted from JSCE, 2007. 
160 Handbook of Precast Segmental Tunnel Lining Systems

model consists of applying uniform vertical ground and groundwater pressures, a linearly
varying lateral earth pressure, self-​weight of the lining and a triangularly distributed hori-
zontal ground reaction between 45° and 135° from the crown. Member forces are calcu-
lated using the elastic equations contained in Table 3.8 (JSCE, 2007; ITA WG2, 2000).

Table 3.8 Internal forces by elastic equation method.

Equations for calculating internal forces by elastic equation method

Item Load Bending moment Axial force Shear force

1 Vertical load (1 − 2S ) PRc
2 2
/4 S Rc P
2
−SCRc P
(P = pe1 + pw1)
2 Horizontal load (1 − 2C 2 ) QRc 2 / 4 C 2 Rc Q −SCRc Q
(Q = qe1 + qw1)
3 Horizontal (6 − 3C −12C 2 + 4C 3 ) (C + 8 C 2 − 4 C 3 ) (S + 8SC − 4SC 2 )
triangular load (Q − Q′ ) Rc 2 / 48 (Q − Q′ ) Rc /16 (Q − Q′ ) Rc /16
(Q ′ = qe 2 + qw 2 )
(Q − Q ′ )
4 Soil reaction 0 ≤ θ ≤ π/ 4 0 ≤ θ ≤ π/ 4 0 ≤ θ ≤ π/ 4
(Pk = kδ h )
(0.2346 − 0.3536C ) Rc 2 kδ h 0.3536CRc kδ h 0.3536C Rc kδ h

π / 4 ≤ θ ≤ π /2 π / 4 ≤ θ ≤ π /2 π / 4 ≤ θ ≤ π /2

( −0.3487 + 0.5S2 + ( −0.7071C + C 2 + (C − 0.7071C 2S )

0.2357C 3 ) Rc 2 kδ h 0.7071S 2C )
Rc k δ h
Rc k δ h

5 Dead load for 0 ≤ θ ≤ π / 2 for 0 ≤ θ ≤ π / 2 for 0 ≤ θ ≤ π / 2

(P = π ⋅ g)
(3/ 8π − θS − 5/6C ) Rc 2 g (θS −1/6C ) Rc g (θS −1/6S ) Rc g
g

for π /2 ≤ θ ≤ π for π /2 ≤ θ ≤ π for π /2 ≤ θ ≤ π

( − π / 8 + ( π − θ) ( − πS + θS + πS2 ( − (π − θ) C + θS +
S − 5/ 6C − 1/ 2 πS 2
) −1/ 6C ) Rc 2 g πSC − 1/ 6S ) Rc g

Rc 2 g
6 Horizontal
δh =
( 2 P − Q − Q ′ + π g ) Rc 4
deformation at 24 (EI + 0.045kRc 4 )
springline (δ h )

Source: JSCE 2007; ITA WG2 2000.

Notes:
1
θ =​ angle from crown; S =​ sin θ ; S 2 =​ sin 2 θ ; S 3 =​ sin 3 θ ; C =​ cos θ ; C 2 =​ cos 2 θ ; C 3 =​ cos 3 θ ;
EI =​ flexural rigidity in unit width.
 Analysis and design of precast segmental linings 161

For this method, the segmental tunnel lining is modeled using a uniform reduced bending
rigidity (Muir Wood, 1975) that takes into account the effect of longitudinal joints between
the segments. Subgrade reaction modulus (spring stiffness) formulations recommended by
different guidelines are given in Table 3.7.
Here Es or oedometer stiffness has the following relationship [Equation (3.14)] with sur-
rounding ground Young’s modulus (Duddeck and Erdmann, 1982).

Es =
(1 − ν) Er (3.14)
(1 − 2ν) (1 + ν)

For example, assuming ν =​0.25, Es =​1.2Er.

3.5.1.2 Beam-​spring method

Using the beam-​spring (also known as bedded beam) method endorsed by AASHTO DCRT-​
1-​2010, JSCE 2007, and ÖVBB 2011, the lining can be modeled in the cross-​sectional plane
as a series of beam elements that span between the longitudinal joints of the segments.
As shown in Figure 3.71, the interaction between the ground and the lining is modeled in
two-​dimensional domain using translational springs in the radial and tangential directions.
When modeling in a three-​dimensional domain, interaction of lining with the ground in
longitudinal direction of the tunnel is also modeled by longitudinal springs accordingly.
Because the lining and ground are represented by a series of beams and springs, this method
is referred to as the beam-​spring or bedded-​beam method.
In the United States tunnel industry, the stiffness of the springs is generally calculated
using formulas recommended by USACE EM 1110-​2-​2901 (1997) [Equation (3.14)].
Furthermore, various two-​dimensional approaches are used to evaluate effects of the
segment joints, including solid ring models with fully bending rigidity, solid ring models
with reduced bending rigidity (Muir Wood, 1975), ring models with multiple hinged

Figure 3.71 Demonstration of springs used for modeling.

Note: Demonstration of springs used for modeling of the interaction between ground and the
lining.
162 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.72 Demonstration of springs used for modeling (double ring).

Note: Double ring beam-​s pring model with radial springs simulating ground, and joint springs
simulating longitudinal and circumferential joints.
Source: Tiberti, 2014. 

Figure 3.73 Demonstration of springs used for modeling (ring joint).

Note: Scheme of ring joint.
Source: Tiberti, 2014. 

joints, and ring models with rotational springs. However, two-​dimensional models can-
not be used to represent circumferential joints or staggered arrangement of segments
between rings.
As shown in Figures 3.72 and 3.73, a two-​and-​a-​half-​dimensional multiple-​hinged seg­­
mented double-​ring beam spring has been used to model the reduction of bending rigidity
and the effects from a staggered geometry. This manipulation is achieved by modeling the
segments as curved beams, flat longitudinal joints as rotational springs or Janssen joints
(Janssen, 1983), and circumferential joints as shear springs. Under final service loads, the
longitudinal joints may be open or closed.

2N
Closed joint: θ≤
Eblt
 Analysis and design of precast segmental linings 163

2N
Open jo int : θ>
Eblt

The Janssen rotational spring stiffness (kjr) is derived accordingly by the following equations:

blt2 E
Closed joint: kjr = (3.15)
12

2
 2M 
9blt EM  − 1
 Nlt 
Open joint: kjr = (3.16)
8N

where
b is the width of segment (contact area in longitudinal joint)
E is Young’s modulus
lt is the length of contact area between segments in longitudinal joints
N is the axial hoop force in segments
θ is the rotation

Refer to Figure 3.74 for presentation of parameters used in the Janssen model.
Two rings are used to evaluate the coupling effects; however, this method uses symmetry
conditions to remove complex support conditions and only half of the segment width is con-
sidered from each adjacent ring for the longitudinal and circumferential joint zone of influ-
ence. Considering the self-​weight of the lining, and distributing the ground, groundwater

Figure 3.74 Demonstration of Janssen joint model.

Note: Presentation of parameters used in Janssen model for determining rotational spring
stiffness.
Source: Groeneweg, 2007. 
164 Handbook of Precast Segmental Tunnel Lining Systems

and surcharge loads along the beam, member forces can be calculated using a conventional
structural analysis package.

3.5.1.3 Finite element method and discrete element method simulations

In soft ground, loose rock, and partially homogeneous solid rock, ÖVBB 2011 and
AFTES-​WG 7:N1993 recommend using FEM and finite difference method (FDM) to cal-
culate the forces in the tunnel lining. The discrete element method (DEM), as shown in
Figures 3.75 and 3.76, is generally considered more appropriate for tunnels in fractured
rock. Recommended engineering properties for analysis of segmental lining in the rock
formations includes properties of intact rock such as unit weight; modulus of elasticity;
unconfined compressive strength (UCS); internal friction angle; tensile strength; and prop-
erties of discontinuities such as joint spacing, joint apparent dip direction, and joint appar-
ent dip.
Other required discontinuities parameters for a DEM modeling for a typical TBM-​bored
tunnel in rock, includes peak joint friction angle, peak joint cohesion, residual joint friction
angle, residual joint cohesion, joint normal stiffness, joint shear stiffness, Eint/​Emass,
geological strength index (GSI), and Mi, which is a material constant for the intact rock
based on Hoek–​Brown failure criterion (Hoek and Brown, 2018). In rock tunneling, a
two-​dimensional approach is generally sufficient for continuous linear structures that do
not contain sudden changes in cross-​sectional geometry or high concentrations of loadings.
Three-​dimensional techniques are generally used in soft ground tunneling due to three-​
dimensional arching effect and in cases with more complex geometry and loadings such as
with crosscuts that intersect the main tunnel (ÖVBB, 2011).

Figure 3.75 Discrete element method model and developed internal forces (axial forces).
Note: Discrete element method (DEM) model and developed internal forces along the lining per-
imeter as a result of DEM analysis on large diameter tunnel excavation in fractured rock: axial
forces.
 Analysis and design of precast segmental linings 165

Figure 3.76 Discrete element method model and developed internal forces (bending moments).
Note: Discrete element method (DEM) model and developed internal forces along the lining per-
imeter as a result of DEM analysis on large diameter tunnel excavation in fractured rock: bending
moments.

Figure 3.77 2D finite element method (FEM) simulation for tunnel excavation in soft ground.
Note: Example of a 2D FEM model for a TBM tunnel in soft ground developed in Plaxis.

As shown in Figure 3.77, FEM is used to model the ground surrounding the liner, where
a continuum medium is discretized into smaller elements, which are connected along adjoin-
ing nodal points (Bakhshi and Nasri, 2013c). The advantage of this method is to be able
to model the ground deformations and the post yielding behavior of the liner materials to
166 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.78 Shield-​d riven TBM processes simulated in a finite element model.

Note: As depicted at this stage of analysis, rings Nos.1 to 11 have been installed; rings Nos. 12 to
13 are under tail-​v oid grout pressure and the shield is occupying location of future rings Nos.14
to 17 with the TBM facing future excavation phase 18. Longitudinal and circumferential joints are
shown on assembled rings.

include any redistribution of stress that results from deformation of the lining and excava-
tion of the tunnel (ÖVBB 2011). FEM analysis techniques can also be used to represent
non-​uniform and anisotropic stresses such as when non-​symmetrical features are present
in the ground. This can be the case when several different geologic formations or external
loads are present within close proximity of an existing structure (AFTES-​WG7:1993). Using
FEM techniques, complex underground conditions, and tunnel characteristics can be ana-
lyzed with a degree of accuracy, and the large axial forces and bending moments developed
in the segments can be reliably determined. Precast segments are designed using an axial
force-​bending moment diagram.
The most accurate results for this load case when tunneling in soft ground are obtained
from 3D staged NFEA modeling of shield-​driven tunnel excavation (Figure 3.78). In this
type of simulation, segmental ring configuration with exact geometry of each segments
(rhomboidal, trapezoidal or rectangular) and staggered joint are directly imported to the
model. Segments are modeled as shell elements and a multi-​linear tension softening and
compression functions (for example, Figure 3.53) is used for material modeling.
The Janssen non-​linear line interfaces simulate the segment joints, and surface inter-
faces are used between the lining and the surrounding ground. Excavation advances are
modeled in intervals equal to the length of segmental rings. A variable balancing face
pressure is applied along the tunnel cross section equal to the horizontal in situ stress.
Compression-​only gap elements are used to model the conical shield and the variable
gap between the ground and the shield. The length of shield is assumed to be 5 to 10
times the ring length and the grout pressure is only applied on the most recently acti-
vated segmental ring and one ring immediately behind, considering anticipated advance
rate and hardening time of grouts (two-​component grout here). A step-​by-​step simu-
lation is adopted to simulate the tunneling procedure. Figure 3.79 presents a typical
result of this 3D NFEA simulations in the format of internal bending moments in the
segmental rings.
 Analysis and design of precast segmental linings 167

Figure 3.79 Segmental lining bending moments in a 3D staged non-​linear finite element model.
Note: Example of segmental lining bending moments as a result of 3D staged non-​linear finite
element modeling of shield-​d riven tunnel excavation.

3.5.2 Longitudinal joint bursting load

Normal (hoop) forces developed in the section, due to the permanent earth and ground-
water pressures, are transferred through a reduced cross-​sectional area across the longitu-
dinal joints where gaskets and stress relief grooves are present. It is worth mentioning that
spalling is not generally of concern on longitudinal joints because that mode of failure is
not applicable when longitudinal joints are fully in contact with adjacent elements except
to account for eccentricity of adjacent segments. Nonetheless, bursting tensile stresses can
develop along the longitudinal joints similar to the TBM thrust jacking loads on the circum-
ferential joints. The design is undertaken for the maximum factored compressive force con-
sidered in ULS design. Simplified equations from ACI 318-​19, DAUB:2013, Iyengar (1962),
and two-​dimensional FEM simulations are the most common methods for carrying out the
analysis and design of longitudinal joint bursting (Bakhshi and Nasri, 2014b).
The simplified equation of ACI 318-​19 for post-​tensioned anchorage zones of prestressed
concrete sections [Equation (3.17)] can be used for analyzing this load case.

Here
Ppu is the maximum normal force from the permanent ground, groundwater, and sur-
charge loads
eanc is the maximum total eccentricity, consisting of the normal force eccentricity (M/​N)
and the eccentricity of the load transfer area.

 h 
ACI 318 − 19: Tburst = 0.25Ppu  1 − anc  ; dburst = 0.5 ( h − 2eanc ) (3.17)
 h 
168 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.80 Force transfer recommended by DAUB:2013 in longitudinal joints.

Note: Force transfer recommended by DAUB:2013 in longitudinal joints using a stress block con-
cept for case of hinge joint eccentricity in the same side as normal force eccentricity (left) and
case of eccentricities on different sides of section centroid (right).

Similar to ACI 318-​19, simplified equations by DAUB:2013 [Equation (3.14)] are used
for evaluating bursting stresses in the longitudinal (planar or flat) joints. Nonetheless,
DAUB:2013 presents more details about this specific load case using an approach that
transfers force by means of a stress block, as shown in Figure 3.80. Additional reinforce­
ment for spalling and secondary tensile stresses are placed when there are high eccentric
normal forces (e > d/​6) (DAUB:2013). Bursting, spalling and secondary tensile stresses are
calculated using the following equations:

Fsd = 0.25 ⋅ N Ed ⋅ (1 − d1 /ds ) (3.18)

 e 1
Fsd , r = N Ed ⋅  −  ; Fsd , 2 = 0.3Fsd ,r (3.19)
 d 6

where
The total eccentricity, e, consists of eccentricity from normal force, el, and the eccentri-
city of hinge joint, ek.
Therefore, e =​ el +​ ek =​ M/​N +​ ek, d1 =​ dk –​2e, and ds =​2e′=​ d − 2el.

These parameters are shown in Figure 3.80.

Bursting tensile reinforcement is placed at a distance of 0.4ds from the face of the seg-
ments; reinforcement for spalling and secondary tensile stresses, if necessary, are placed at
 Analysis and design of precast segmental linings 169

Figure 3.81 Finite element method analysis for longitudinal joint bursting (model).
Note: Example of finite element method (FEM) analysis for the load case of longitudinal joint
bursting: FEM model representing governing cases.

0.1ds and 2/​3d from the face of the segment, respectively (DAUB:2013). Simplified equa-
tions, which include Equations (3.18) and (3.19), can be used for this load case to determine
the compressive stress and the strength of the partially loaded surface.
Iyengar diagram methods (Figure 3.44) and FEM simulations can also be used as an alter­
native approach to determine the stresses within the longitudinal joints. For this load case,
two-​dimensional FEM models are developed to represent the recess of the gasket and the
stress relief grooves (curvature of elements is neglected). Bursting stresses at the vicinity of
the longitudinal joints are analyzed for cases of maximum hoop force and maximum posi-
tive and negative load eccentricities (el =​ M/​N) transferred between longitudinal joints. For
example, Figures 3.81 and 3.82 show the most governing cases considering partially open,
and closed joints due to extreme eccentricities, and the maximum uniform load transfer for
a typical mid-​size tunnel lining project. Reinforcement is designed to take these bursting and
compressive stresses, similar to the case for TBM thrust jacking loads.
In addition to the eccentricity of forces (M/​N) and of the load transfer area (because of
recesses), design for an additional eccentricity due to ovalization (or squatting) and mis-
alignment during the ring erection are sometimes asked for in project’s technical require-
ments. Such a load case is referred to as ring ovalization due to out of round ring build or
birds-​mouthing. The design procedure includes an assumption that the ring is initially built
in the shape of an ellipse, and that the chord length of the displaced segment, as shown in
Figure 3.83 does not change. Note that in this figure the x-​axis and y-​axis represent the
dimension of the quarter of the segmental ring in the Cartesian coordinate system. Solid
arc and dashed arc represent the shape of quarter of ring before and after ovalization,
respectively.
Specific out-​of-​round build allowance over diameter is considered [for example, 0.6 inches
(15mm)]. Joint rotation and joint opening distance are calculated with total joint rotation
causing birds-​mouthing (τ as in Figure 3.83) and opening distance due to poor ring build (δ
as in Figure 3.83) as the main parameters. Joint closure under minimum/​maximum embed­
ment loads on segment intrados and extrados are assessed by determining the load required
to close the gap and comparison with the hoop force due to embedment loads.
170 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.82 Finite element method analysis for longitudinal joint bursting (bursting stresses).
Note: Example of finite element method (FEM) analysis for the load case of longitudinal joint
bursting: resulting bursting stresses.

Depending on whether the joint remains open or closed, one of the two diagrams shown
in Figure 3.84 is used for calculation of birds-​mouthing eccentricity. Although provided
details are for flat joints as the most conventional joint shape, convex joint details are
similar in terms of design approach with consideration of a line load instead of a distributed
load at the joint locations.

3.5.3 Loads induced due to additional distortion

Segmental tunnel linings are designed to take an additional diametrical distortion in add-
ition to the deflections caused by the effects of ground, groundwater and surcharge loads,
which were discussed in the previous load case. This additional distortion may occur during
segment assembly under the self-​weight of the segments due to construction-​related events
such as joint misalignment, yielding of joint connectors, or excessive grouting pressure.
Furthermore, this distortion can result from ground movement caused by the construction
 Analysis and design of precast segmental linings 171

Figure 3.83 Ring ovalization due to out-​o f-​r ound ring build.

Note: In ring ovalization diagram, the x-​a xis and the y-​a xis represent dimensions of the quarter
of the segmental ring in the Cartesian coordinate system while solid arc and dashed arc represent
the shape of quarter of ring before and after ovalization, respectively.
Source: Adopted from Francis and Mangione, 2012. 

Figure 3.84 Bursting tensile stresses around longitudinal joints under hoop (normal) forces.
Note: Developed bursting tensile stresses around longitudinal joints under hoop (normal) forces
due to service loading condition and gasket pressure: (b) eccentric contact stresses.
Source: Adopted from Francis and Mangione, 2012. 
172 Handbook of Precast Segmental Tunnel Lining Systems

of an adjacent tunnel. This additional distortion is the difference between the movement of
the tunnel at the left and right springline or the crown and invert of the tunnel. Some local
authorities such as LACMTA 2013 and LTA 2010 require the design to accommodate this
additional distortion. The former specifies a minimum additional diametrical distortion of
0.5% of diameter due to imperfect lining erection and the latter specifies an additional
distortion of ±5/​8 inches (15 mm) on the diameter to allow for future development in the
vicinity of the tunnel. The following formula introduced by Morgan (1961) is commonly
used to calculate the additional distortional bending moment. Please refer to Equation
(3.20).

3EI δ d
Mdistortional = (3.20)
2r02

Using other approaches, the maximum distortion can be calculated based on the theory of
elasticity or finite element methods (FEM). Note that during excavation in certain ground
conditions such as clay materials, consolidation of underlying clay layers can result in add-
itional distortion to be a function of time. In such cases, additional distortion should be
considered and analyzed as a time-​dependent phenomenon.

3.5.4 Seismic loads
Bored tunnels are usually designed for two earthquake levels: maximum design earth-
quake (MDE), and operating design earthquake (ODE). The MDE has a low probability of
exceedance, approximately 4% in 100-​year service life, and the tunnel shall be designed to
maintain public safety during and after this design event. The ODE has a higher probability
of occurrence, approximately 50% in 100-​year service life, and the tunnel performance
objective for this event is to remain operational with minimal damage.
During a seismic event, a bored tunnel primarily undergoes three types of deforma-
tions: ovaling, axial, and curvature (Owen and Scholl, 1981), as shown in Figure 3.85.
Ovaling is mainly caused by shear waves propagating vertically perpendicular to the tunnel
axis (Wang, 1993). Axial and curvature deformations are generated when the seismic waves
propagate either parallel or oblique to the tunnel axis. In the following sections, some of
the most widely accepted approaches for estimating each of these three deformation types
will be reviewed.

3.5.4.1 Closed-​form solutions –​ovaling

Closed-​form solutions have been used to estimate lining ovality demands. These simplified
solutions are mainly based on the following assumptions:

• The lining has a constant thickness and behaves linearly

• The surrounding ground is an infinite, elastic, homogenous medium
• The elements inside the tunnel, including the structural walls and invert infill, do not
impact the tunnel seismic response.

In general, the closed-​form solutions are more suitable for early stages of design since
majority of tunnels are excavated in layered strata and there are different ground formations
near the tunnel, which result in some level of non-​linearity during MDE ground motion.
 Analysis and design of precast segmental linings 173

Figure 3.85 Tunnel deformation types during an earthquake.

Note: Tunnel transverse ovaling and longitudinal axial and curvature responses to traveling waves.
Source: Adopted from Wang, 1993. 

Therefore, more rigorous analyses, such as using numerical modeling, which is explained in
the next section, is recommended for detailed design of tunnels.
Estimating the seismic-​induced ovalization of tunnels using closed-​form solution consists
of the following steps:

1 –​Determining the maximum free-​field ground deformation

Expressed in terms of maximum shear strain ( γ max ), this is the maximum anticipated ground
deformation at tunnel springline elevation during MDE or ODE event in the absence of
tunnel. To evaluate this value, one-​dimensional linear or non-​linear site response analysis
software, e.g. SHAKE or DEEPSOIL, can be used. These software usually take the stain-​
dependent shear modulus and damping ratio of all ground layers as an input, which need to
be established based on site-​specific geotechnical investigation finding. The ground motions
used in the analysis should be spectrally adjusted to match the seismic hazard at project site
(LACMTA, 2013). As an alternative, simplified closed-​form solutions are also available,
which are mainly suitable for initial assessments, as they involve assumptions that might
not reflect the tunnel site conditions. Newmark (1965) proposed the following simplified
equation:

AF*Vs
γ max = (3.21)
Cs
174 Handbook of Precast Segmental Tunnel Lining Systems

Table 3.9 Ground motion attenuation factor.

Ratios of ground motion values at tunnel
depths to those at the ground surface
(AASHTO LRFD-​1 -​2 017)

Item Tunnel depth Bending moment

1 < 20 ft 0.95–​1
(< 6 m)
2 20–​5 0 ft 0.75–​0 .95
(6–​1 5m)
3 50–​1 00 ft 0.5–​0 .75
(15-​3 0m)
4 > 100 ft 0.5
(> 30m)

where
Vs is peak ground velocity (PGA)
AF is attenuation factor obtained from Table 3.9
Cs is effective shear wave velocity obtained from in situ (e.g. geophysical investigations)
or lab testing

2 –​Evaluaating the ground-​tunnel interaction

To quantify the lining stiffness relative to the host ground and its impact on tunnel seismic
response, previous studies (Merritt et al., 1985) have used the compressibility and flexibility
ratios (C and F ) defined as below:

Em (1 − vl2 ) R
C= (3.22)
El t (1 + vm ) (1 − 2vm )

Em (1 − vl2 ) R3
F= (3.23)
6El I (1 + vm )

where
Em is modulus elasticity of the medium
I is moment of inertia of the tunnel lining (per unit width)
Cs is effective shear wave velocity obtained from in-​situ (e.g. geophysical investigations)
or lab testing
R and t are nominal radius and thickness of the tunnel lining, respectively
vl and vm are Poisson’s ratios of the tunnel lining and medium, respectively.

To account for presence of longitudinal joints between precast segments, the moment of
inertia can be adjusted using Equation (3.13).

3 –​Calculating the lining demand

Assuming no shear force to develop at the tunnel-​ground interface (full-​slip condition),
Wang (1993) proposed the following equations for the maximum thrust and bending
moment in the tunnel lining:
 Analysis and design of precast segmental linings 175

1 Em
Tmax = ± K1 Rγ max (3.24)
6 (1 + vm )

1 Em
Mmax = ± K1 R2 γ max (3.25)
6 (1 + vm )

where

12 (1 − vm )
K1 = .
2 F + 5 − 6v m

This solution assumes both the tunnel and medium behave linearly during an earthquake.
For tunnels excavated in rock or hard soil, however, the interface condition is between full-​
slip and no-​slip. Therefore, it is recommended to conservatively assume no-​slip condition
when estimating the maximum thrust (Höeg, 1968; Schwartz et al., 1980).

Em
Tmax = ± K2 Rγ max (3.26)
2 (1 + vm )

where

1
F (1 − 2vm ) − (1 − 2vm )C  −
(1 − 2vm )2 + 2
K2 = 1 + 2
5 
F (3 − 2vm ) + (1 − 2vm )C  + C  − 8vm + 6vm2  + 6 − 8vm
2 

3.5.4.2 Numerical analysis –​ovaling

This approach involves utilizing a commercial software to model the tunnel, the surround-
ing ground, and the interaction between them. Finite element or finite difference programs,
e.g. PLAXIS, RS2 or FLAC, can be used to create a two-​dimensional, plain-​strain model of
the tunnel cross section. The model should properly account for site stratigraphy, ground
layers dynamic properties, lining joints, ground-​tunnel interaction, and any element inside
the tunnel that affects its ovality. Figure 3.86 shows an example model built using PLAXIS
2D with appropriate boundary conditions and geometry. The vertical boundaries should be
placed far enough from the tunnel to avoid any boundary effects.
Three of most widely used numerical analysis methods are described below.

1 –​Maximum deformation method

In this approach, first, the maximum ground free-​field ground deformation at tunnel eleva-
tion is estimated using site response analysis. Then, the model boundaries are displaced
till the same free-​field deformation is generated in the ground. It should be noted that the
mechanical properties of all ground layers in the model shall be adjusted to account for the
degradation of stiffness with shear strain increase. Figure 3.87 shows an example tunnel
model constructed in PLAXIS 2D with the maximum free-​field ground deformation applied
176 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.86 Typical finite element model of TBM tunnel for seismic analysis.
Note: Example model built using PLAXIS 2D for seismic analysis with appropriate boundary
conditions and geometry.

Figure 3.87 Free-​f ield ground deformation applied to tunnel model.

Note: An example tunnel model constructed in PLAXIS 2D with the maximum free field ground
deformation applied to the boundaries.

to the boundaries. In this case, because the underlying rock layers are significantly stiffer
than the overburden, the free-​field deformation is only applied over the surficial soil column
with an overall height of H.

2 –​Pseudo-​dynamic time-​history analysis

Overall, this method is similar to the maximum deformation method, except that instead
of using only the maximum ground deformation, the model vertical boundaries are stepped
through the free-​field ground deformation profile. The advantage of this method is the
ability of examining the surrounding ground and tunnel lining response throughout the
design earthquake duration rather than only looking at an instant. Use of this analysis
becomes crucial when the tunnel lining and/​or the ground are expected to experience a
notable degree of non-​linearity or advanced constitutive material models are utilized for
prescribing the behavior.
 Analysis and design of precast segmental linings 177

3 –​ Dynamic time-​ h istory analysis This is the most rigorous method among the three
approaches and involves applying different earthquake time histories at the base of the tunnel
cross section model. In addition to stiffness, the strain-​dependent damping of all ground lay-
ers shall be input into the software. For cases when the tunnel behaves dynamically different
than the surrounding ground, e.g. tunnels in liquifiable soils or with significant inertial effect,
adoption of this approach is necessary. The amount of data generated in this analysis is usually
large and engineering judgment might be needed to interpret and use them for design purposes.
It should be noted that some authorities may limit the designers to using more refined
analysis approaches. For instance, the Los Angeles County Metropolitan Transportation
Authority only recognizes the time-​history methods (pseudo-​static or dynamic) for ana-
lyzing the tunnels within its jurisdiction (LACMTA 2013).

3.5.4.3 Axial and curvature deformations

Ignoring the tunnel interaction with the surrounding ground, a simple approach to estimate
these deformations would be estimating the ground free-​field axial and curvature response
and then assuming the tunnel conforming to it. This solution gives an upper-​bound assess-
ment of the tunnel response and is deemed to be sufficient for most cases. St. John and
Zahran (1987) developed a set of equations for the ground free-​field axial strain due to
axial and curvature deformations. The combined total axial strain under different seismic
waves can be calculated as follows:

For P waves:

VP A
ε= cos2 ∅ + R P2 sin∅ cos2 ∅ (3.27)
CP CP

For S waves:

VS A
ε= sin∅ cos∅ + R 2S cos3∅ (3.28)
CS CS

For Rayleigh waves:

VR A
ε= cos2 ∅ + R R2 sin∅ cos2 ∅ (3.29)
CR CR

where
VP , Vs , VR are the peak ground velocities of P, S and Rayleigh waves at tunnel elevation,
respectively;
CP , Cs , CR are the effective propagation velocities of P, S and Rayleigh waves, respectively;
AP , As , AR are the peak ground acceleration of P, S and Rayleigh waves, respectively;
∅ = angle of seismic wave propagation with respect to the tunnel axis.

Among these three seismic wave types, generally, the S waves generate the largest axial
strains, and the seismic wave propagation angle should be set to generate the maximum
effect when checking the lining seismic response.
178 Handbook of Precast Segmental Tunnel Lining Systems

The above equations are overly conservative when the tunnel is significantly stiffer than
the surrounding ground. For such cases, the equations need to be adjusted using reduction
factors to account for tunnel-​ground interaction:
For axial strains:

2
El Al  2π 
Ra = 1 +   cos ∅ (3.30)
2
Ka  L 

For bending strains:

4
El Il  2π 
Rb = 1 +   cos ∅ (3.31)
4
Kh  L 

where
El is Young’s modulus of tunnel lining,
Al is cross sectional area of the lining (per unit width),
Il is moment inertia of the tunnel lining (per unit width),
Kh is transverse ground spring constant,
Ka is longitudinal ground spring constant,
L is wavelength of the P, S or Rayleigh waves.

Ra and Rb are multiplied into the first and second terms of Equations (3.27) to (3.29),
respectively. The maximum calculated axial strain should not exceed the tunnel-​ground
interface shear strength, Qmax , as defined by the following equation:

fL
Qmax = (3.32)
4

where
fL is tunnel-​ground interface shear strength per unit lining width.

For the cases where the tunnel traverses highly variable strata or undergoes abrupt stiff-
ness change, e.g. at tunnel interface with a station or a shaft, adopting a 3D modeling
approach might be crucial to properly capture the tunnel axial and curvature behavior. The
joints at the tunnel interface with other structures should be designed carefully to be able to
accommodate the anticipated seismic deformations while maintaining the water tightness
of both the tunnel and the adjoining structure. Figure 3.89 shows an example bored tunnel-​
station joint detail implemented in Crenshaw/​LAX Transit Corridor Project in Los Angeles,
CA, using Omega seals (Piek et al., 2017). Three-​dimensional time-​history analyses in 3D
LS-​DYNA were conducted to estimate the differential movements at the joints.

3.5.4.4 Fault crossing

The safest stagey when comes to active fault crossing is to avoid them. If this is not possible,
then the corresponding displacements should be accommodated by either employing
an oversized excavation, and backfilling it with compressible materials later, or using a
 Analysis and design of precast segmental linings 179

Figure 3.88 Bored tunnel-​s tation joint detail.

Note: An example of bored tunnel-​s tation joint detail used in Crenshaw/​L AX Transit Corridor
project in Los Angeles, CA.
Source: Adopted from Piek et al., 2017. 

flexible lining to minimize the potential damage. Fault crossing assessment usually carried
out in three steps: 1 –​estimating the average free-​field fault displacement, 2 –​imposing
the anticipated displacement to the tunnel using a numerical model capable of capturing
ground-​tunnel interaction, and 3 –​checking the tunnel lining demand versus its capacity.
The average free-​field fault displacement can be evaluated by either observing the mag-
nitude of fault displacement during past earthquakes or using empirical method that relate
the displacement to earthquake magnitude (LACMTA, 2013). As an example, Wells and
Coppersmith (1994) developed a graph, as shown in Figure 3.89, that correlates the max­
imum fault displacement with earthquake magnitude. It should be noted that the maximum
displacement occurs only along a very limited section of the fault. Therefore, the average
displacement, which is in the range of 20 to 80% of the maximum displacement, is recom-
mended to be used for design purposes (Wells and Coppersmith, 1994).
The estimated free-​field displacement is applied to the tunnel lining through the use of
non-​linear ground springs or, in the case of implementation of a continuum finite element
of finite difference modeling approach, via definition of appropriate boundary conditions,
as shown in Figure 3.90.
180 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.89 Maximum surface fault displacement versus earthquake moment magnitude.

Note: Relationship between maximum surface fault displacement and earthquake moment mag-
nitude for strike-​s lip faulting.
Source: Data from Wells and Coppersmith, 1994. 

3.5.5 Fire incident and explosion

A load case specific to road and railway tunnels is tunnel fires. One of the decisive factors
for the design of tunnel structures in the load case of fire is the heat release rate. Rates from
10 MW up to 200 MW have been proposed at different projects with the majority ranging
around 100 MW.
Another important factor is the duration of the design fire with values ranging from 30
minutes to 3 hours, sometimes with an added cooling phase. Both heat release rate and dur-
ation will depend on intended use of tunnel and project-​specific conditions such as the type
of traffic (train, cars, heavy goods or dangerous goods transport) and the required safety
level (with stricter requirements in the case of a possible impact on structures above the tun-
nel, or the risk of inundation) (Neun, 2012).
The actual time-​temperature curve resulting from such a design fire, which depends on
the individual cross section and air flow speeds inside the tunnel can be used for simulation
of a temperature gradient between the intrados and extrados of the tunnel lining. However,
since adequate computational fluid dynamics (CFD) modeling is very time consuming and
the results are not easily transferred to a structural model, it is common to use standard
deterministic time-​temperature curves for the projects (Neun, 2012).
Most common standard design fire curves for the structural design of tunnel structures
as shown in Figure 3.91 include RABT-​ZTV (Railways) also known as EBA, RABT-​ZTV
(Highways) also known as ZTV, ISO 834 Cellulose, HC Hydrocarbon Eurocode 1, HCM
Modified Hydrocarbon also known as HCinc, and Rijkswaterstaat also known as RWS
 Analysis and design of precast segmental linings 181

Figure 3.90 Tunnel fault crossing model.

Note: Actual geometry and idealized structural model at fault crossing.
Source: Adopted from ASCE, 1984. 

Figure 3.91 Standard fire load curves.

Note: Common fire temperature-​t ime diagrams used in tunneling.
Source: Data from ITA WG6, 2004. 
182 Handbook of Precast Segmental Tunnel Lining Systems

(ITA WG6, 2004). The ISO 834 curve is recommended up to an expected fire heat release
rate of 50 MW, above which the hydrocarbon curve (up to 100 MW) and thereafter the
RWS curve (up to the stoichiometric limit) should be applied. When cooling phase is con-
sidered, RABT curves can be used.
Once one of the standard design fire curves is selected (temperature versus time), it is
applied on the tunnel intrados. The increase in lining temperature vs. lining thickness is cal-
culated and the resulting reduction in concrete and reinforcement properties (modulus and
strength) is determined based on available data. This will determine the lining thickness loss
during the fire. If using basic structural analysis programs, applying non-​linear temperature
gradients are not allowed for, an equivalent temperature load can be established that has
the same impact on the equivalent section as the original temperature gradient has on the
original section (Neun, 2012). This can be followed by adopting a layered section analysis
where the normal force (N) and bending moment (M) can be determined by integration/​
summation of the stresses in the individual layers.
The most accurate results for the fire action are often obtained as a results two-​dimensional
finite element analysis using a coupled thermo-​mechanical model. For example, for fire acci-
dents in railway/​subway tunnels, RWS fire loading, which is typical for these types of tun-
nels is applied on either the entire tunnel lining intrados (Figure 3.92) or along the tunnel
lining intrados except those areas, which will be filled by track slab or ballast (Figure 3.93).
Alternatively, a fire curve specific to project can be obtained through the computational
fluid dynamics (CFD) simulation of actual train fire.
Initial temperature is typically considered as 68oF (20oC). Heat flow is simulated fol-
lowing recommendations of Eurocode 2 standard (EN 1992-​1-​2:2004) for structural fire
design. As shown in Figures 3.94 and 3.95, the concrete conductivity and the volumetric
specific heat (heat capacity) as a function of temperature are used as inputs to the model
following recommendation of EN 1992-​1-​2:2004 Section 3.3.2 and 3.3.3.
For the RWS fire curve example, the numerical results of the thermal analyses as nodal
temperature contours along a 21 feet (6.5 m) diameter tunnel lining are presented in

Figure 3.92 Locations of fire loading in FEM model.

Note: Example of locations of fire loading along the entire tunnel lining intrados.
 Analysis and design of precast segmental linings 183

Figure 3.93 L ocations of fire loading in FEM model excluding areas that will be filled by track
slab or ballast.
Note: Example of locations of fire loading along the tunnel lining intrados excluding areas that
will be filled by track slab or ballast.

Figure 3.94 Concrete heat conductivity.

Note: Concrete heat conductivity as a function of temperature (EN 1992-​1 -​2 :2004).

Figures 3.96 to 3.100 at 30 minute intervals. These results are also presented by means of
the temperature history at different points over the segment thickness in Figures 3.101 and
3.102, and by means of the temperature distribution along the segment thickness at differ­
ent fire exposure times in Figure 3.103. Note that for the fire analysis, it is recommended to
reduce the tunnel lining thickness by 0.8 in (20 mm) from the intrados to consider the effect
of potential spalling zone.
184 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.95 Concrete heat capacity.

Note: Concrete heat capacity as a function of temperature (EN 1992-​1 -​2 :2004).

Figure 3.96 Nodal temperatures at 0 min. for RWS fire curve analysis.

Note: Nodal temperature as a result of thermal analysis along the lining perimeter.

When performing a coupled thermo-​mechanical FEM analysis, in order to consider the

effect of fire damage along the cross section, the mechanical properties of concrete as a
function of temperature must be given to the model as input. In particular, the mechanical
property decay laws proposed by EN 1992-​1-​2:2004 Section 3.2.2.1 are recommended
for concrete properties under compression. The strength and deformation properties of
 Analysis and design of precast segmental linings 185

Figure 3.97 Nodal temperatures at 30 min. for RWS fire curve analysis.

Note: Nodal temperature as a result of thermal analysis along the lining perimeter.

Figure 3.98 Nodal temperatures at 60 min. for RWS fire curve analysis.

Note: Nodal temperature as a result of thermal analysis along the lining perimeter.

uniaxially stressed concrete at elevated temperatures are introduced to the FEM model as
stress–​strain relationships as a function of temperature. The main parameters of the stress–​
strain relationships of normal weight concrete at elevated temperatures are the compressive
strength (f’c or fck), the strain corresponding to peak compressive strength (εc1), and ultimate
compressive strain capacity (εcu). These parameters are available in EN 1992-​1-​2:2004
Table 3.1 and can be directly introduced to the model as shown in Figures 3.104 to 3.106.
In precast concrete segments reinforced with rebar, the tensile strength of concrete under
tension can be neglected, and recommendation of EN 1992-​1-​2:2004 Table 3.2a for the
strength and deformation properties of reinforcing steel at elevated temperatures should
186 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.99 Nodal temperatures at 90 min. for RWS fire curve analysis.

Note: Nodal temperature as a result of thermal analysis along the lining perimeter.

Figure 3.100 Nodal temperatures at 120 min. for RWS fire curve analysis.
Note: Nodal temperature as a result of thermal analysis along the lining perimeter.

be followed. In case segments are only reinforced with steel fibers, the mechanical decay
of the post-​crack residual flexural and tensile stresses of FRC should be used for analysis.
Experimental results of Caverzan et al. (2015) are recommended for this purpose and are
shown in Figures 3.107 and 3.108. In their study, feq1 is the nominal strength in crack mouth
opening dimension (CMOD) ranging between 0.012 and 0.02 inches (0.3 to 0.5 mm), which
is considered as a reference for serviceability limit state (SLS) residual strength and feq2 is
the average strength in CMOD for ultimate crack width which is regarded as ultimate limit
state (ULS) residual strength.
 Analysis and design of precast segmental linings 187

Figure 3.101 Temperature history along the lining thickness for RWS fire curve analysis (node
locations).
Note: Example of temperature history at different nodal points along the lining cross-​s ection
(segment thickness of 12 in or 300 mm) as a result of 2D FEM fire analysis: location of nodes.

Figure 3.102 Temperature history along the lining thickness for RWS fire curve analysis (analysis
results).
Note: Example of temperature history at different nodal points along the lining cross-​s ection
(segment thickness of 12 in or 300 mm) as a result of 2D FEM fire analysis: analysis results.
188 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.103 Temperature profile at different fire exposure times for RWS fire curve analysis.
Note: Example of temperature profile at different fire exposed times along the lining thickness.

Figure 3.104 Compressive strength of concrete as a function of temperature.

Note: f ’c or f ck as one of main parameters of compressive stress-​s train relationships at elevated
temperatures (EN 1992-​1 -​2 :2004).

An average stress decay factor applicable to post-​crack flexural and tensile strength can
be introduced to the model as shown in Figure 3.109. Also, according to Caverzan et al.
(2015), the ultimate crack width is reduced from 0.155 inches (3.94 mm) at 68oF (20oC) to
0.119 inches (3.02 mm) at 392oF (200oC), 0.083 inches (2.12 mm) at 752oF (400oC), and
0.046 inches (1.17 mm) at 1112oF (600oC). The fib Model Code 2010 defines the structural
characteristic length (lcs) for sections without traditional reinforcement as the entire section
thickness. In the referenced study (Caverzan et al. 2015) the section thickness of the samples
 Analysis and design of precast segmental linings 189

Figure 3.105 Strain corresponding to peak compressive strength as a function of temperature.

Note: ε c1 as one of main parameters of compressive stress-​s train relationships at elevated tem-
peratures (EN 1992-​1 -​2 :2004).

Figure 3.106 Ultimate compressive strain capacity as a function of temperature.

Note: ε cu as one of main parameters of compressive stress-​s train relationships at elevated tem-
peratures (EN 1992-​1 -​2 :2004).

was 30 mm. Considering this characteristic length, the ultimate tensile strain (εtu) at 68oF
(20oC), 392oF (200oC), 752oF (400oC) and 1112oF (600oC) are 0.13, 0.1, 0.07 and 0.039,
respectively. As for the first peak tensile strength decay of concrete materials under elevated
temperature, section 3.2.2 of EN 1992-​1-​2:2004 provides a relationship, which is shown in
190 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.107 Decay of post-​c rack residual flexural strength of fiber reinforced concrete.
Note: Serviceability limit state (SLS) strength as a function of temperature.
Source: Data from Caverzan et al. 2015. 

Figure 3.108 Decay of post-​c rack residual flexural strength of fiber reinforced concrete.
Note: Ultimate limit state (ULS) strength as a function of temperature.
Source: Data from Caverzan et al. 2015. 

Figure 3.110. All aforementioned tensile material properties are introduced to FEM models
as input tensile material models for a coupled thermo-​mechanical analysis.
The tunnel lining is commonly modeled as a non-​jointed solid ring due to difficulties in
simulation of the longitudinal joints in two-​dimensional (2D) plane strain FEM models –​
which are often used for the load case of fire. The interaction with the ground is simulated
using an interface with radial spring stiffnesses presented in Table 3.7. As a result of coupled
 Analysis and design of precast segmental linings 191

Figure 3.109 Average residual tensile stress decay factor.

Note: Average decay stress factor applied to post-​c rack flexural and tensile strength of fiber
reinforced concrete (FRC).

Figure 3.110 First peak tensile stress decay factor.

Note: Tensile strength decay parameter of Eurocode 2 (Section 3.2.2.2) for concrete at elevated
temperatures.

thermo-​mechanical analysis, the hoop stresses developed along the lining thickness can be
obtained. Example of hoop stresses in the lining at exposure time of two (2) hours due to
a project-​specific fire curve are presented in Figures 3.111, 3.112 and 3.113. Note that fol­­­
lowing the methodology of EN 1992-​1-​2:2004, in the compressive zone and near intrados
192 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.111 Hoop stress variation along the lining thickness.

Note: Example of hoop stresses developed along the lining thickness due to fire as a result of
coupled thermo-​m echanical FEM analysis.

Figure 3.112 Diagram of hoop stress along the lining thickness.

Note: Example of hoop stresses developed along the lining thickness due to fire as a result of
coupled thermo-​m echanical FEM analysis.
 Analysis and design of precast segmental linings 193

Figure 3.113 Tabular and graphical presentation of variations of hoop stress versus distance from
the lining extrados.
Note: Example of hoop stresses developed along the lining thickness due to fire as a result of
coupled thermo-​m echanical FEM analysis.

where very high temperatures are anticipated, isotherm 500oC (932oF) zone method should
be adopted by ignoring both strength and developed stresses in a zone with temperatures
higher than 932oF (500oC). This methodology basically implies that this part of section
contributes neither to capacity nor demand for structural calculations. The hoop stresses
in the segment thickness such as those presented in Figure 3.111 to 3.113 –​excluding the
isotherm 500oC (932oF) zone –​are integrated to obtain resulting axial forces and bending
moments in the tunnel lining.
In addition, existence of the longitudinal joints in the segmental ring results in reduction
of the bending moment. Note that according to Muir Wood’s (1975) formula [Equation
(3.13)], the effective moment of inertia of the lining also known as reduced flexural rigidity
considering the number of longitudinal joints in the ring is much lower than its full flex-
ural rigidity or moment of inertia. For example, for a ring with six longitudinal joints, the
effective moment of inertia of the lining is only 0.44 times its full flexural rigidity (Ir =​ Ij +​
(4/​6)2 × I =​0.44 I). Such bending stiffness reduction due to the effect of joints usually results
in reduction of non-​jointed solid rings’ bending moments by 30–​40%. Therefore, when for
example a 2D plane strain FEM model is adopted for a six-​segment ring, considering a 30%
reduction in the bending moment for effect of joints seems to be reasonable. Axial forces
and reduced bending moments obtained from presented fire analysis must be superimposed
by the internal forces resulting from the embedment loads in final service stage analysis
presented in Sections 3.5.1 through 3.5.4. The resulting internal forces are considered as
structural demands for the load case of fire.
In a simplified approach, the compressive and tensile capacity of the lining at specific
exposure time to fire such as two hours can be determined using average decay factors for
compression and tension. A table that presents variation of temperature and corresponding
compressive strength decay factor as a function of distance from intrados can be helpful for
194 Handbook of Precast Segmental Tunnel Lining Systems

this purpose. After excluding the isotherm 500oC (932oF) zone, an average decay factor for
the compression zone and an average decay factor for tensile zone are determined accord-
ingly. Compressive and tensile material properties related to each average decay factor are
used for the calculation of decayed lining capacity due to fire. Decayed bending moment-​
axial force (M-​N) interaction diagram can be constructed using closed-​form solutions of
Section 3.6.8 (Yao et al., 2018) and compared with the internal forces due to the fire and
embedment loads for design checks.
Explosions, on the other hand, are simulated by increasing the internal radial pressure on
the tunnel lining at the service condition by a representative value such as one atmosphere
or 14.5 psi (1 bar) (Caan et al., 1998). This internal radial pressure in the direction opposite
to ground and groundwater pressure results in reduced axial forces in the lining without
significant change in the bending moments. Recently an advanced and detailed design pro-
cedure for tunnels subjected to internal explosion and possibly preceded by fire accidents
was developed for FRC tunnel lining (Colombo et al., 2015). Simplified finite element model
and dynamic analyses were carried out to study the tunnel’s response under internal blast
loads in the form of pressure-​impulse (p-​i) diagrams and an ultimate limit state criterion
based on eccentric flexural capacity (M-​N interaction diagram) was generated. Also, a limit
state criterion considering the fire-​blast interaction was introduced through the modifica-
tion of the M-​N diagram. This procedure is suggested for an advanced blast-​fire analysis.

3.5.6 Break-​o ut analysis and shear recovery systems

Creating opening for cross-​passages and adit connections at TBM tunnel would result in
permanent disruption to the structural response of segmental lining and change in effective
stress of the ground in close proximity to segments where the cross-​passage is to be built.
During excavation of the cross-​passage, there will be a temporary stage of reduced ground
support to the tunnel lining. Temporary steel bracing is often installed inside the tunnel to
provide support mitigating ring ovalization resulting from opening and cross-​passage exca-
vation. The internal support will maintain the structural integrity of segments ensuring a
safe design.
As shown in Figures 3.114, 3.115 and 3.116, steel lintel and sill beams, “full-​moon”
support frame, and “half-​moon” support frames are among the most common temporary
opening support systems (Lee and Choi, 2017). Latest development includes use of shear
recovery bicone systems in conjunction with temporary “half-​moon” support frame, which
eases the problems with the creation of openings inside the segmentally lined tunnel by
minimizing the amount of temporary work and minimizing interference with the TBM
operations.
Bicones, as shown in Figure 3.117, prevent any offset between the rings during ring
assembly in the construction stage and absorb energy when the tunnel lining is partially
suppressed for opening or when instantaneous and temporary stresses occur in exceptional
instances. Bicones are designed to be used in the proximity of tunnel penetration areas such
as cross-​passages, entrances, ventilation adits and elevator adits. Although other methods
such as anchors, steel connectors and shear keys have been used in some specific cases for
this purpose, bicones are the most technically advanced solution that are now being com-
monly used.
Analysis is performed for the case of removal of several segmental rings with a maximum
height equal to the opening size. A three-​dimensional, non-​linear modeling approach using
an FEM package is adopted to evaluate the impacts of excavation. Figure 3.118 shows typical
 Analysis and design of precast segmental linings 195

Figure 3.114 Temporary steel lintel and sill beams in segmentally lined tunnels.
Note: Temporary steel lintel and sill beams as steel bracing structures (support frames) in seg-
mentally lined tunnels to mitigate ring ovalization resulting from opening and cross-​p assage
excavation.
Source: Lee and Choi, 2017. 

Figure 3.115 Temporary “full-​m oon” support frame in segmentally lined tunnels.

Note: Temporary “full-​m oon” support frame as steel bracing structures (support frames) in
segmentally lined tunnels to mitigate ring ovalization resulting from opening and cross-​p assage
excavation.
Source: Lee and Choi, 2017. 
196 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.116 Temporary “half-​m oon” support frames in segmentally lined tunnels.

Note: Temporary “half-​m oon” support frames as steel bracing structures (support frames) in
segmentally lined tunnels to mitigate ring ovalization resulting from opening and cross-​p assage
excavation.
Source: Lee and Choi, 2017. 

Figure 3.117 Use of bicones in opening areas of TBM tunnels.

Note: Shape and installation pattern of bicone shear dowels and temporary “half-​m oon” support
frame around the opening.
Source: Adopted from Lee and Choi, 2017. 

geometry, developed meshes and shear stresses around a penetration zone. Total shear force
of a ring around the penetration area is calculated and compared to the shear strength of the
bicones to determine the minimum number of bicones required for this action.

3.6 DETAILED DESIGN CONSIDERATIONS

Segmental tunnel linings installed in the rear of the TBM shield are generally in the shape of
circular rings. The size of the ring is defined by the internal diameter, thickness and length of
 Analysis and design of precast segmental linings 197

Figure 3.118 Shear stresses developed in segmental lining due to adit connection.

Note: Three-​d imensional model for openings inside the segmentally lining lined tunnel and devel-
oped shear stresses as results of the finite element analysis.

the ring. Other important design considerations include ring systems, ring configurations in
terms of number of segments that form a complete ring, geometries of individual segments,
and geometry and tapering of key segments (Bakhshi and Nasri, 2018).

3.6.1 Concrete strength and reinforcement

Several recommendations are available on the compressive strength of precast concrete tun-
nel segments.

• AASHTO DCRT-​1-​2010 recommends a compressive strength ranging from 5,000 to

7,000 psi (34 to 48 MPa) for one-​pass lining systems.
• Japan’s Railway Technical Research Institute (RTRI 2008) specifies segments com-
pressive strengths between 6,000 and 7,800 psi (42 and 54 MPa) whereas JSCE 2007
specifies a range of 6,000 to 8,700 psi (42 to 60 MPa).
• ÖVBB 2011 recommends 1,700 and 5,800 psi (12 and 40 MPa) as the minimum com-
pressive strength for stripping (demolding) from the forms and at 28 days, respectively.
• DAUB:2013, referring to local German guideline ZTV-​ING 2007, specifies a min-
imum stripping (demolding) strength of 2,200 psi (15 MPa), and a 28 day strength
between 5,000 and 7,200 psi (35 and 50 MPa).
• LTA 2010 specifies the minimum compressive strength for precast segments as 8,700
psi (60 MPa), which is the highest requirement of all guidelines and standards at the
moment.
• USACE EM 1110-​2-​2901 (1997) requires a concrete strength of 6,000 psi (42 MPa)
or higher for one-​pass segmental linings.

A summary of recommendations for compressive strength of precast concrete tunnel seg-

ments are presented in Table 3.10.
Precast concrete segments are reinforced with steel bar Grades 60, 75, or 80 (conforming
to ASTM A615/​ A615M and ASTM A706/​ A706M), welded steel wire reinforcement
198 Handbook of Precast Segmental Tunnel Lining Systems

Table 3.10 Recommendations for compressive strength of precast tunnel segments

Compressive strength Compressive strength

Item Authority (stripping) (28 days)
1 AASHTO DCRT-​1 -​2 010 Not provided 5,000 psi to 7,000 psi
(34 MPa to 48 MPa)
2 RTRI 2008 Not provided 6,000 psi to 8,700 psi
(42 MPa to 60 MPa)
3 ÖVBB 2011 1700 psi or 12 MPa 5,800 psi (minimum)
(minimum) (40 MPa) (minimum)
4 DAUB:2013 2200 psi 5,000 psi to 7,300 psi
(referring to ZTV-​ING 2007) (15 MPa) (35 MPa to 50 MPa)
5 LTA 2010 Not provided 8,700 psi
(60 MPa)
6 USACE EM 1110-​2 -​2 901: 1997 Not provided 6,000 psi (minimum)
(42 MPa) (minimum)

Grades 70, 75 or 80 (conforming to ASTM A1064/​A1064M), or steel fibers (conforming to

ASTM A820/​A820M). When reinforced with steel bars or welded wire mesh, reinforcement
is often categorized to three different types: transverse reinforcement, which is the main
reinforcement placed perpendicular to tunnel axis to resist forces and bending moments
developed under final service load cases; longitudinal reinforcement in the direction of tun-
nel axis and perpendicular to transverse reinforcement, often designed as minimum tem-
perature and shrinkage reinforcement; and joint reinforcement near segment and ring joints
for resisting bursting and spalling stresses.
A typical plan view of transverse and longitudinal bars in precast concrete tunnel seg-
ments is shown in Figure 3.119. No specifications or requirement can be found in tunnel
guidelines for minimum bar size. However, a review of segmental tunnel projects reveals
that transverse bar size ranges between No. 3 and No. 5 (10 and 16 mm diameter). The
general trend for recent projects built with even larger tunnel diameters has been the reduc-
tion of bar size to No. 3 (10 mm diameter) and using more bars in transverse direction by
reducing bar spacing to below 3 inches (80 mm). In the longitudinal direction, reinforcing
bar size ranges between No. 3 and No. 5 (6 and 16 mm diameter) with more closely spaced
bars in recent projects.
Typical sectional views of joint reinforcement in precast concrete tunnel segments are
shown in Figures 3.120 and 3.121.

• Circumferential joint reinforcement includes transverse and radial reinforcement.

• Transverse reinforcement in the circumferential joint is usually the same size as the
main transverse reinforcing bars unless analysis shows that reinforcement is not suffi-
ciently provided to resist stresses due to TBM thrust jack forces.
• Radial bars in circumferential joints in forms of ties or U-​bars are either No. 3 or No.
4 (10 or 13 mm diameter) bars designed either equally spaced or non-​uniformly with
more closely spaced bars around the TBM jack shoe locations.
• Longitudinal joint reinforcement in the radial direction is sometimes designed as lad-
der bars in addition to other forms (that is, ties and U-​bars) due to the simplicity of
welding ladder bars in a flat surface comparing to a curved face in the circumferential
direction.
 Analysis and design of precast segmental linings 199

Figure 3.119 Plan view of transverse and longitudinal bars.

Note: Typical plan view of transverse and longitudinal bars in precast concrete tunnel segments.

Figure 3.120 Sectional views of joint reinforcement–​c ircumferential joint.

Note: Typical sectional views of joint reinforcement in precast tunnel segments.

Manufactured rolled and cut welded wire meshes can be custom made as an alternate avoid-
ing the labors of tying. Radial bars have been designed in longitudinal joints as No. 3 or No.
4 (10 or 13 mm diameter). Longitudinal reinforcing bars in longitudinal joints are often of
the same size as the general longitudinal reinforcement inside the segment.

3.6.2 Early age strength for stripping

Depending on the production method, high early-​age strengths may be required for the con-
crete (usually 6 to 8 hours) to resist early production and transient load cases. The design
engineer should specify the required compressive strength at the time of segment stripping
(demolding), which is usually between 1,900 and 2,500 psi (13 and 17 MPa). For fiber
200 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.121 Sectional views of joint reinforcement–​longitudinal joint.

Note: Typical sectional views of joint reinforcement in precast tunnel segments.

reinforced concrete (FRC) precast segments, first-​crack and residual flexural strength, as
the basis of design, are specified in accordance with recommendations of ACI 544.7R-​16,
Chapter 4.

3.6.3 Concrete cover
DAUB:2013 recommends a minimum concrete cover of 1.6 inches (40 mm) on the sur-
faces of the tunnel segments. On the end faces of segments and in areas close to bolt pock-
ets, the minimum concrete cover recommended by DAUB:2013 is 0.8 inches (20 mm).
However, ACI 318-​19 specifies a minimum concrete cover of 1.5 inches (38 mm) for pre-
cast concrete elements exposed to earth. ITA Working Group on General Approaches to
the Design of Tunnels (1988) specifies a minimum 2.0 inches (50 mm) cover only for the
cast-​in-​place concrete lining at its outer surface in contact with ground and groundwater.
This specification does not apply to segmental lining especially when a one-​pass lining
system is adopted.
Among other codes, guidelines, and recommendations, AASHTO DCRT-​1-​2010 does
not specify a minimum concrete cover; JSCE 2007 specifies a minimum 1 inches (25 mm)
concrete cover over reinforcement and a minimum 1.4 inches (35 mm) concrete cover in
a corrosive environment for a one-​pass segmental lining system; ÖVBB 2011 refers to
EN 1992-​1-​1:2004, which specifies a minimum concrete cover of 1.0 to 1.8 inches (25 to
45 mm) depending on exposure conditions; AFTES:2005 specifies 1.2 inches (30 mm) as the
minimum cover on the intrados and extrados faces and 0.8 inches (20 mm) concrete cover
on other zones; and NEN 6720:1995 specifies 1.4 inches (35 mm) as the minimum concrete
cover for precast elements. A summary of recommendations for minimum concrete cover
over reinforcement are presented in Table 3.11.

3.6.4 Curing
Steam curing should be used as the primary curing method for the first six hours after con-
crete is placed into the forms. Immediately after segments have been cast, forms are placed
in a sealed, vapor-​tight enclosure to prevent escape of moisture and heat, but large enough
to allow complete circulation of steam. Segments should not be removed from the forms
until specified stripping (demolding) compressive strength is attained, determined by com-
pressive cylinder test results.
 Analysis and design of precast segmental linings 201

Table 3.11 Minimum recommended concrete cover.

Minimum recommended concrete cover for reinforcing bars in reinforced
concrete tunnel segments by national or international codes, guidelines, and
recommendations

Codes, guidelines or recommendations

ACI NEN ÖVBB 2011
Reference 318-​1 9 AFTES DAUB JSCE 6720 EN 1992-​1 -​1
document 2019 2005 2013 2007 1995 2004
Minimum 1.5” Intrados Intrados Non-​ 1.4” 1.0” to 1.8”
concrete (38 mm) and and corrosive (35 mm) (25 to 45 mm)
cover over extrados: extrados: environment: depending
reinforcing 1.2” 1.6” 1.0” (25 mm) on exposure
steel bars (30 mm) (40 mm) conditions
Other Joint faces Corrosive
zones: and near environment:
0.8” bolt 1.4” (35 mm)
(20 mm) pockets:
0.8”
(20 mm)

To avoid damages to early-​age concrete micro-​structure, as a common practice, enclosure

ambient temperature is kept below 100°F (38°C) for the first two hours of curing. After
two hours of curing, enclosure temperature is maintained between 90 and 120°F (32 and
49°C) until specified curing strength is achieved. As shown in Figure 3.122, it is crucial to
keep maximum temperature of the curing enclosure below 130°F (55°C) because this tem-
perature is approximately equivalent to 150°F (65°C) internal concrete temperature, above
which in ordinary portland cement concrete, delayed ettringite formation (DEF) as a major
durability factor can be generated in the presence of reactive aggregates. It is noteworthy to
mention that according to ACI 207.1-​21, the maximum concrete temperatures up to 185°F
(85°C) will not result in DEF when the concrete cementitious materials consists of: (a) 25%
by mass of ASTM C618 Class F fly ash; (b) 35% by mass of ASTM C618 Class C fly ash;
(c) 35% by mass of ASTM C989/​C989M slag cement.
In addition, care should be taken to maintain rate of temperature change under 30°F
(17°C) per hour. When specified strength is attained, segments are only allowed to cool
slowly so as not to exceed the aforementioned rate of temperature change. Following
primary steam curing, steam curing or moist curing methods can be adopted until 70 per-
cent of the design strength is reached. Note that when using the moist curing method,
segments should be placed in a special area where water is continuously sprayed or
atomized.

3.6.5 Reinforcement spacing
There is no specific recommendation available in ACI 318-​19 for reinforcement spacing in
the precast tunnel segments. However, general spacing limits for reinforcement include min-
imum and maximum clear bar spacing of 1 inch (25 mm) and 18 inches (457 mm), respect-
ively. DAUB:2013, provides a typical reinforcing spacing range of 4 to 6 inches (100 to
150 mm) for segmental tunnel linings, and specifies a minimum clear spacing of 3.5 inches
(90 mm).
202 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.122 Kiln temperature and thermocouples embedded in segment.

Note: Example of record of the kiln temperature and thermocouples embedded in segment core.

In the absence of any reinforcement spacing requirement by ITA (1988), AASHTO

DCRT-​1-​2010, and ÖVBB 2011, JSCE 2007 specifies a minimum bar spacing of 1.25 times
the width of the maximum size of aggregates plus the diameter of the reinforcement. AFTES
2005, referring to Section 4.5 of BAEL 91 Révisé 99: 2007, specifies the maximum spacing
for reinforcing bars as the smaller of 8.0 inches (200 mm) and 1.5 times the segment thick-
ness. NEN 6720:1995, on the other hand, specifies minimum bar spacing as the greater of
4/​3 of the maximum size of aggregates, the largest bar diameter, or 1.0 inch (25 mm).

3.6.6 Fiber reinforcement
Fiber reinforcement has emerged as an alternative to traditional reinforcing bars and
welded wire mesh reinforcement for precast concrete tunnel segments. Due to significantly
improved post-​cracking behavior and crack control characteristics, fiber reinforced concrete
(FRC) segments offer advantages over traditionally reinforced concrete segments such as
saving (especially labor) cost and reducing production time while developing a more robust
product with improved handling and long-​term durability (ACI 544.7R-​16; fib Working
Party 1.4.1 2017; ITA Working Group 2 2016).
FRC technology has developed in recent years with the introduction of high-​strength con-
crete, allowing the use of fibers as the sole reinforcement system for more challenging condi-
tions on larger diameter tunnel project. In other cases, fiber and reinforcing bars have been
used in conjunction to reinforce the tunnel segments (Yao et al., 2018a, 2018b). Ladder bars
at the longitudinal joint as well as bursting links at the circumferential joint may be signifi-
cantly reduced by adding steel fibers. Tunnels with different sizes with internal diameters
ranging between 7.2 and 39 feet (2.2 m and 11.9 m) have been built using fiber reinforce-
ment. Minimum and maximum thickness of the FRC precast segments in existing tunnels
 Analysis and design of precast segmental linings 203

are 5.9 inches (0.15 m) and 18 inches (0.46 m), respectively. These projects include water
supply, wastewater, gas pipeline, power cable, subway, railway, and road tunnels.

3.6.7 Codes, guidelines and design parameters

The FRC post-​cracking nominal residual strengths are the most important parameters for
design of FRC segments (ACI 544.7R-​16; fib Working Party 1.4.1 2017; ITA Working
Group 2 2016). ACI 544.7R-​16 and fib Working Party 1.4.1 2017 provide design proce-
dures for FRC tunnel segments to withstand all the appropriate temporary and permanent
load cases occurring during the production, construction and design life of segmental
tunnels, using specified post-​crack residual tensile strength, σp. Applying raw parameters
from ASTM C1609/​C1609M or EN 14651:2005+​A1:2007 beam tests, such as fD150 or fR,3,
requires caution to prevent over-​estimating the residual tensile strength when using elastic
analysis (Bakhshi et al., 2014). A back-​calculation procedure can be adopted to obtain spe-
cified residual tensile strength parameter σp (Soranakom and Mobasher, 2007).
Alternatively, the post-​crack flexural strength parameters determined in accordance with
ASTM C1609/​C1609M and EN 14651:2005+​A1:2007 can be scaled by an adjustment
factor ranging from 0.33 to 0.37 (Bakhshi et al., 2014; Mobasher et al., 2014; Vandewalle,
2000; Barros et al., 2005; fib Model Code 2010). As shown in Figure 3.123, this strength
parameter along with the specified compressive strength (f′c per ACI 318-​19 or fcd per EN
1992-​1-​1:2004) are used to obtain strain and stress profiles through the section to construct
the axial force-​bending moment diagram that is the key design tool for segments.
Although several different FRC constitutive laws (fib Model Code 2010; DBV 2001;
RILEM TC 162-​TDF; CNR DT 204/​2006; EHE-​08) have been used for design of these
elements, results of studies by Bakhshi and Nasri (2014c) shown in Figure 3.124, reveal
that the choice of constitutive laws does not have a significant effect on axial force-​bending
moment interaction diagrams and, therefore, the design outcome. Detailed design informa-
tion on FRC segments can be found in ITA Working Group 2 (2016), fib Working Party
1.4.1 2017, and ACI 544.7R-​16.

Figure 3.123 Strain and stress distributions.

Note: Strain and stress distributions through the section as part of it undergoes tension.
Source: Bakhshi and Nasri, 2014c. 
204 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.124 Axial force-​b ending moment interaction diagrams.

Note: Effect of choice of constitutive law on the axial force-​b ending moment interaction diagram
as a key design tool.
Source: Bakhshi and Nasri, 2014c. 

For fiber reinforced concrete (FRC) elements, following ACI 544.7R-​16 and fib Working
Party 1.4.1 2017, appropriate strength reduction factors or material safety factors should
be considered for flexure, compression, shear and bearing actions of concrete segments.
Such factors account for the uncertainty of post-​crack tensile strength when calculating the
design strength of FRC elements.

3.6.8 Hybrid solution of fibers and reinforcing bars

Under design conditions where use of FRC is not adequate as the sole reinforcing mech-
anism, a hybrid solution of fibers and reinforcing bars has been shown to be an alternative
solution. Often times, in large diameter tunnels, a hybrid solution of combined rebar and
fiber reinforcement has been adopted (Tiberti et al., 2015; Plizzari, 2009). Generally, for
segments with slenderness ratio of higher than 13, a hybrid reinforcement is recommended
(ACI 544.7R-​16). In shallow large-​diameter tunnel segments, low internal axial forces and
high bending moments are present. Such higher flexural demands impose high localized
stresses that can be better taken by reinforcing bars than fibers (Tiberti, 2009). It is believed
that in these cases in addition to fiber reinforcement, a certain amount of conventional
 Analysis and design of precast segmental linings 205

reinforcement is necessary for the required flexural capacity against ground and ground-
water loads at final service stage (de la Fuente et al., 2012).
In addition, TBM thrust forces in large diameter tunnels can be very high, especially
in soft ground tunneling. In some extreme cases, segments resisting these extraordinarily
localized bursting and spalling stresses due to the TBM jacking process needs to be rein-
forced by rebar in addition to fibers. Also, in large-​diameter tunnels with an increased ratio
between the tunnel diameter and lining thickness, segments are more likely to withstand the
high flexural stresses due to imperfections and irregular construction (Tiberti, 2009). As an
example, gaps between rings or outward eccentricity of thrust jack forces necessitate the
use of rebar for in-​plane deep beam behavior (Burgers et al., 2007). Collectively speaking
from the design point of view, if FRC is not adequate for design because of any of reasons
mentioned above, a hybrid solution of fibers and reinforcing bars may provide the required
strength at ultimate limit state (ULS) and serviceability limit state (SLS).
During the application of load cases occurring from the demolding stage all the way to
the final service stage, segments are most of the time subjected to combined axial force and
bending moment. Axial force-​bending moment interaction (P–​M) diagram is, therefore, an
important tool that tunnel engineers use for the structural design. Despite the popularity
and abundance of numerical methods for deriving P–​M diagrams, closed-​form solutions
offer important advantages for engineering problems.

First, algorithms expressed with closed-​form solutions are more computational efficient
and easier to implement into computer programs or simple hand calculations in com-
parison to finite element methods (FEMs) or complex equations that need non-​linear
numerical solvers (Vasdravelliset et al., 2012; Rodriguez and Aristizabal-​Ochoa, 1999;
Dinsmore, 1982).

Second, closed-​form solutions that are presented as symbolic math expressions offer a clear
view into how different variables interact with one another in affecting the results under
various modes of failure. Optimization algorithms and parametric studies can be conducted
readily using symbolic differentiation and integration to investigate the role of critical design
parameters such as reinforcement ratio, sample dimensions, concrete strength and tensile
residual strength.

Third, parameterized equations can be used for selection of variables using a design auto-
mation procedure; hence gradient-​based optimization algorithms can be conducted much
faster.

Finally, analytical model with closed-​form solutions can be implanted into a simple and
user-​friendly spreadsheet program that allows user to define the problem with a limited
number of parameters. This is much more difficult with numerical or FEM analysis as com-
plicated models have to be created each single time and the computation time will be much
longer.
Due to these advantages, Yao et al. (2018) presented material models (Figure 3.125), sec­
tional stress and strain diagrams at all failure modes (Figures 3.126 to 3.130), derivations
and for the first time closed-​form solutions (Table 3.12 and Table 3.13) to construct a full
range P–​M interaction diagram of hybrid reinforced segments (Figures 3.131 to 3.134)
considering the contributions of fibers in the post-​cracking strength in addition to reinfor-
cing bars.
206 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.125 FRC material models.

Note: Material models including (a) FRC tensile model: (b) FRC compressive model; (c) steel
model; (d) cross section.

Figure 3.126 Material strain and stress diagram models at model 1.1.

Note: Material strain and stress diagrams at mode 1.1: all compression with bottom fiber yielded
in compression.

Parametric studies have been conducted by Yao et al. (2018) on the effects of fiber addi-
tions represented by residual strength, the confined compressive strength, the longitudinal
reinforcement ratio and the steel strength. Results show an improvement in moment cap-
acity after tensile cracking with addition of fibers, because of the fiber contribution to tensile
resistance in cracked zones. The increased compressive strength due to lateral confinement
mainly contributed to the load and moment capacities in compression-​controlled mode of
 Analysis and design of precast segmental linings 207

Table 3.12 Normalized height of compression/​t ension zones, stresses/​forces for each mode.
Normalized height of compression/​t ension zones, stresses/​forces at vertices for
all modes

Mode 1 Mode 2 Mode 3

Parameter (χ > 0) (-​κ < χ < 0) (-​0 .005/​ε cr < χ <-​ κ)

1.1 1.2 2.1 2.2 3.1 3.2

Normalized -​
hc2 λ cu − ω k ( λ cu − ω ) k ( λ cu − ω )
height
h λ cu − λ λ cu λ cu

hc1 -​ ω−λ ωk ωk
h λ cu − λ λ cu λ cu

ht1 -​ -​ 1− k k −1 k −1
h β β

ht2 -​ -​ -​ (1 − k )(1 + β ) (1 − k )(1 + β )

h β β

Normalized fc2 γω γω γω
stress
E ε cr

γω γλ γω γω
f c1
E ε cr
nκ nκ nκ nχ
f s′
E ε cr

-​ -​ β 1 1
f t1
E ε cr
-​ -​ -​ µ µ
f t2
E ε cr
nκ nχ nχ nκ
fs
E ε cr
Normalized -​ -​ -​
force Fc 2 λ cu − ω γ kω( λ cu − ω )
γω
bhE ε cr λ cu − λ λ cu
γω γω γω
Fc1 ω2 − λ 2 γ kω 2
γω
bhE ε cr λ cu − λ 2 λ cu

(Continued)
208 Handbook of Precast Segmental Tunnel Lining Systems

Table 3.12 (Continued)

Mode 1 Mode 2 Mode 3

Parameter (χ > 0) (-​κ < χ < 0) (-​0 .005/​ε cr < χ <-​ κ)

ρg nκ ρg nκ ρg nκ ρ g nχ
Fs′
bhE ε cr

-​ -​
Ft1 β(1 − k ) (1 − k ) (1 − k )
bhE ε cr 2 2β 2β
-​ -​ -​
Ft 2 ( k − 1)(1 + β ) ( k − 1)(1 + β )
µ µ
bhE ε cr β β
ρ g nχ ρ g nχ ρg nκ
Fs
bhE ε cr

Table 3.13 Normalized force and moment for each mode

Parameter Mode Normalized expression

Force 1.1 P11′ = 2κ nρg + ωγ
1.2
( λ 2 + ω 2 − 2ωλ cu )γ + 2nρg ( χ + κ )( λ − λ cu )
P12′ =
2( λ − λ cu )

2.1
( ω 2 γ − 2ωγλ cu + βλ cu )k β
P21′ = − + nρg ( χ + κ ) +
2 λ cu 2
2.2
 ω2 γ  2βµ + 2µ − 1
P22′ =  − + ωγ  k + ( k − 1) + nρg ( χ + κ )
 2 λ cu  2β

3.1
 ω2 γ  2βµ + 2µ − 1
P31′ =  − + ωγ  k + ( k − 1)
 2 λ cu  2β

3.2
 ω2 γ  2βµ + 2µ − 1 nρg ( α − 1)
P32′ =  − + ωγ  k + ( k − 1) − nρg ( λ cu − κ ) + λ cu
 2 λ cu  2β k

Moment 1.1 M11′ = 0

1.2
C1 λ cu 2 + C2 λ cu + C3 β2 + 2ω3 γ
M12′ =
2(β − λ cu )2
 Analysis and design of precast segmental linings 209

Table 3.13 (Continued)

Parameter Mode Normalized expression

2.1 ′ = C 4 k 2 + C5k + C6
M21
2.2 ′ = C7 k 2 + C8 k + C9 ( χ − κ ) + C10
M22
3.1 ′ = C7 k 2 + C8 k − 2 C9 + C10
M31

C12
3.2 ′ = C7 k 2 + C8 k + C11 +
M32
k

where

λ cu
k= , C1 = 6nρg (2α − 1)( χ − κ ) , C2 = 12β2 nρg (2α + 1)(κ − χ ) − 3βγω( ω − 2 λ cu ) ,
β + λ cu

ω3 γ 3ω 2 γ 1  3ω 2 γ 
C3 = β2 C1 − γω 2 (2ω − 3λ cu ) , C 4 = − + − 3ωγ + β , C 5 = −  − 6ωγ + β
λ cu 2
λ cu 2  λ 
cu
,

β, ω − 3λ cu 2 6µ − 3 3µ − 2
C 6 = 3nρg (2α − 1)( χ − κ ) − C7 = − γω − 3( γω + µ ) − + ,
2 λ cu 2 β β2

3γω 2 18µ − 9 6µ − 4 6µ − 3 3µ − 2
C8 = − + 3( γω + µ ) + + , C9 = −3nρg (2α − 1) , C10 = − −
2 λ cu 2β β2 2β β2 ,

C11 = C10 − C9 (κ + λ cu ) , C12 = C 9 (1 − α )λ cu

Figure 3.127 Material strain and stress diagram models at model 1.2.

Note: Material Strain and stress diagrams at mode 1.2: all compression while bottom fiber is not
yielded.

failure while the effects in tension-​controlled failure is diminished. An overall improvement

in the section capacity was observed with increasing reinforcement ratio and steel yielding
strength. These results are presented in Figure 3.135.
Presented model has been verified by simulating experimental results from published
data for cases of hybrid reinforced concrete (HRC) columns (Figures 3.136 to 3.140) and
several other model-​ predicted and numerical results for HRC precast tunnel segments
210 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.128 Material strain and stress diagram models at mode 2.1.

Note: Material strain and stress diagrams at mode 2.1: compression controlled, no tension crack.

Figure 3.129 Material strain and stress diagram models at mode 2.2.

Note: Material strain and stress diagrams at mode 2.2: compression controlled with crack in
tension.

Figure 3.130 Material strain and stress diagram models at mode 3.

Note: Material strain and stress diagrams at mode 3: tension controlled.

(Figures 3.139 and 3.140) and cast-​in place concrete tunnel linings (Figure 3.141). These
results further verified the accuracy of the model and its suitability for design engineers.
A spreadsheet-​based program has also been developed and is available for general users
to readily construct the interaction diagram with the present method at https://​doi.org/​
10.13140/​RG.2.2.14437.09440/​1
 Analysis and design of precast segmental linings 211

Figure 3.131 Normalized axial force-​b ending moment interaction diagram.

Note: Normalized P-​M diagram showing different failure modes.

Figure 3.132 Parametric study (residual tensile strength).

Note: Parametric study on the effects of (a) residual tensile strength μ.

3.7 TESTS AND PERFORMANCE EVALUATION

Testing of concrete materials should be incorporated into the contract specifications to
confirm that the design parameters are met and for the case of FRC segments to determine
the optimal fiber content during production (AASHTO DCRT-​ 1-​
2010; Angerer and
Chappell, 2008). All of the required design parameters mentioned in Sections 3.3 to 3.5
should be determined by testing. Performance testing or proof testing is also approved by
212 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.133 Parametric study (yield strain).

Note: Parametric study on the effects of improved compressive yield strain due to confine-
ment ω c ,.

Figure 3.134 Parametric study (reinforcement ratio).

Note: Parametric study on the effects of reinforcement ratio ρ g .

most structural codes. This is especially true for repetitive units, where large numbers are
required to meet a particular purpose such as with the manufacturing of segments. Full-​
scale tests are often conducted to evaluate the design and performance of fiber reinforced
concrete (FRC) segments with slenderness ratio of more than 12 to 13. Non-​structural tests
on precast concrete tunnel segments include full-​scale fire test, which helps to ensure 2-​to-​3-​
hour fire resistance using standard fire load curves shown in Figure 3.91.
 Analysis and design of precast segmental linings 213

Figure 3.135 Parametric study (steel yield strength).

Note: Parametric study on the effects of steel yield strength.

Figure 3.136 Comparison of proposed axial force-​b ending moment interaction diagram.

Note: Comparison of proposed P-​M diagram and experimentally results for RC columns tested
by Chaallal and Shahawy (2000).
214 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.137 Comparison of proposed axial force-​b ending moment interaction diagram (HPFRC
columns).
Note: Comparison of proposed P-​M diagram and experimental results for HPFRC columns tested
by Foster and Attard (2001).

Figure 3.138 Comparison of proposed axial force-​b ending moment interaction diagram (UHPFRC
columns).
Note: Comparison of proposed P-​M diagram and experimental results for UHPFRC columns
tested by Steven and Empelmann (2014).
 Analysis and design of precast segmental linings 215

Figure 3.139 Axial force-​b ending moment interaction diagram for the tunnel lining segment.
Note: Axial force-​b ending moment interaction diagram for the tunnel lining segment in Reference
(de la Fuente et al., 2012) with a cross-​s ection of 4.92 feet (1,500 mm) by 1.0 foot (300mm), 8
φ10 rebar (4 in top layer and 4 in lower layer) and varying amount of fibers.

Figure 3.140 Locally magnified axial force-​b ending moment interaction diagram for tunnel lining
segments.
Note: Closer look at the range indicated in Figure 3.133 representing axial force-​ b ending
moment interaction diagram for the tunnel lining segment in de la Fuente et al. (2012) with a
cross-​s ection of 4.92 feet (1,500 mm) by 1.0 foot (300 mm), 8 φ10 rebar (4 in top layer and 4 in
lower layer) and varying amount of fibers.
216 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.141 Comparison between applied actions for a cast-​in-​p lace concrete lining.
Note: Comparison between applied actions for a 1,000 mm by 500 mm cross section cast-​in-​
place concrete lining in Chiaia et al. (2007) and the model-​p redicted interaction diagrams by
different methods.

3.7.1 Materials tests
Tests used to determine the required design parameters include compressive, flexural
and splitting tensile strength tests. Standard ASTM C39 tests with 6 × 12 inch (150 ×
300 mm) cylinders are recommended for determining the specified compressive strength,
f’c. This test is conducted on cylindrical concrete specimens. Peak flexural strength and
residual flexural strengths are obtained from ASTM C1609 standard tests, using 6 ×
6 × 20 inch (150 × 150 × 500 mm) fiber-​reinforced concrete (FRC) beams in three-​
point loading. EN 14651 (2005) test is used as an alternative for obtaining the residual
strength of FRC.
The ASTM C1609 or EN 14651 (2005) standard beam test is the most common test used
for the design of FRC segments on tunnel projects; however, the residual standard flex-
ural parameters cannot be used directly for design purposes because of the overestimation
that develops from using elastic methods of analysis (Bakhshi et al., 2014; Mobasher
et al., 2014). A stress–​strain diagram is required for designing the load cases mentioned in
Chapters 4 through 6 using methods recommended in ACI 544.7R-​16.
Alternative methods to obtain FRC tensile stress–​strain diagram include the cracked hinge
formulations by de Oliveira e Sousa and Gettu (2006) and Olesen (2001), and the fictitious
crack models by Zhang and Stang (1998) and Kitsutaka (1997). A simplified approach for
obtaining the post-​crack residual tensile strength consists of using the post-​crack tensile
plastic constitutive law with a scale factor of 0.33 to 0.37 (Bakhshi et al., 2014; Mobasher
et al., 2014; Vandewalle, 2000; Barros et al., 2005; fib Model Code 2010) applied to the
standard residual parameters. This simplified approach has been successfully used on several
 Analysis and design of precast segmental linings 217

tunneling projects (Vodička et al., 2012; Caratelli et al., 2012; Barwart et al., 2013; Tiberti
et al., 2008).
Although the specified residual tensile strength is used to check bursting and spalling
tensile stresses, ASTM C496 splitting tensile strength test is also recommended for verifica-
tion to make sure measured splitting strength is greater than the designed residual strength.
Also, results of a modified closed-​loop splitting test method equipped with crack opening
displacement (COD) gauges, including splitting tensile strength, can be used directly for
design against bursting and spalling stresses replacing σp parameter when a hardening-​type
behavior is exhibited. Statistical scattering of test results, which is usual for FRC materials,
requires at least three test data for averaging all of the aforementioned design parameters.
The aforementioned parameters are used for the construction of a stress block model such
as the CS TR63 (2007) model, which is typically used for calculation of the nominal resist-
ance bending moment of FRC members subjected to combined axial and bending actions.

3.7.2 Full-​scale tests
Bending and point load tests at full scale are conducted up to load levels much higher than
the TBM nominal service load, and strength results are compared with the actions presented
in Sections 3.3 through 3.5. Bending tests, as shown in Figures 3.142 to 3.144, are per-

Figure 3.142 Full-​s cale bending tests (setup).

Note: Full-​s cale bending test: test setup.
Source: Meda and Rinaldi, 2017. 
218 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.143 Full-​s cale bending tests (instrumentation).

Note: Full-​s cale bending test: measurement instrumentation.
Source: Meda and Rinaldi, 2017. 

Figure 3.144 Full-​s cale bending tests (instrumentation plan).

Note: Full-​s cale bending test: measurement instrumentation.
Source: Caratelli et al., 2012. 
 Analysis and design of precast segmental linings 219

Figure 3.145 Point load tests.

Note: Point load test set-​u p simulating TBM thrust jack force.
Source: Caratelli et al., 2012. 

formed to verify the design and performance of segments during the production stages
of stripping (demolding), storage, transportation and handling, as well as for asymmetrical
earth pressure at the service stage. Full-​scale point load tests, as shown in Figures 3.145 and
3.146, simulate the TBM thrust jack forces on the segment cross section during the excava­
tion process (Caratelli et al., 2012), as well as the force transfer through a reduced cross
section in longitudinal joints.
The cantilever load test, as shown in Figures 3.147 and 3.148, is another full-​scale test
that is used to investigate the circumferential joint strength under misaligned jacking loads
(Poh et al., 2009). Concrete tunnel lining strength has also been evaluated by full-​scale tests,
such as those shown in Figure 3.149 to 3.151 to simulate dominant effects of axial forces,
bending moments, and combined action of axial loads and bending moment (Mashimo
et al., 2002). In this study, half of the circular tunnel lining was simulated for a tunnel with
an outer diameter of 31.8 feet (9.70 m), a thickness of 12 inches (300 mm), and a height of
3.2 feet (1.0 m). Loads were applied on 17 locations every 10 degrees from the bottom with
two jacks per each location, each 1 foot (300 mm) and 2.3 feet (700 mm) from the bottom
of specimen. As shown in Figures 3.149 to 3.151, three types of loading were adopted.

Type (a) loading represents a case that a load of loose rock/​soil acts on the lining in the tun-
nel crown while rest of the lining is supported by the surrounding ground. To simulate
this load case, loading was applied in all 17 locations to the lining with 34 jacks up to
4.5 kip/​jack (20 kN/​jack). Except for jacks applying loads on three locations in the crown
that continued loading increase up to the failure, the remaining jacks maintained 4.5
kip/​jack (20 kN/​jack) axial force, simulating forces of springs as ground-​structure inter-
action. As a result of load type (a), axial force was the dominant internal force compared
to the bending moment.
220 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.146 Point load tests –​ sketch.

Note: Point load test set-​u p simulating TBM thrust jack force.
Source: Caratelli et al., 2012. 

Figure 3.147 Cantilever load tests.

Note: Cantilever load test set-​u p and instrumentation: (a) before testing; (b) after testing (photo
credit to Professor Tan Kiang Kwee, National University of Singapore, 2008).
 Analysis and design of precast segmental linings 221

Figure 3.148 Cantilever load tests –​ sketch.

Note: Cantilever load test set-​u p and instrumentation.
Source: Adopted from Poh et al., 2009. 

Figure 3.149 Full-​s cale tests on half of the circular segmental lining –​ case 1.
Note: Full-​s cale loading cases on half of the circular lining simulating: (a) load of loose rock and
soil acts on the lining in the tunnel crown while rest of the lining is supported by the surround-
ing ground.
Source: Adopted from Mashimo et al., 2002. 
222 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.150 Full-​s cale tests on half of the circular segmental lining –​ case 2.
Note: Full-​s cale loading cases on half of the circular lining simulating: beyond/​b ehind a cavern,
load of loose rock/​s oil acts on the lining in the crown with ground providing no support on large
part of tunnel shoulder.
Source: Adopted from Mashimo et al., 2002. 

Figure 3.151 Full-​s cale tests on half of the circular segmental lining –​ case 3.
Note: Full-​s cale loading cases on half of the circular lining simulating: ground is unfavorable and
only acting against the lining in all direction without providing any support.
Source: Adopted from Mashimo et al., 2002. 

Type (b) loading on the other hand is for a case that beyond/​behind a cavern, the load
of loose rock/​soil acts on the tunnel lining in the crown but ground does not provide
support on a large part of the tunnel shoulder, and only a limited part of the lining is
supported by the surrounding ground. Type (b) simulation was done by applying loads
in nine locations, with four locations from the left as well as right bottom, and from
one location at the crown. Loading was increased in the crown location after the axial
force in all other 16 jacks (two per location) reached and maintained 2.25 kip/​jack (10
kN/​jack) simulating loads of ground-​structure interaction springs. As a result, bending
moment in the tunnel shoulder was the dominant internal force in type (b) loading com-
pared to the axial force dominant effect in load type (a).
Type (c) loading represents a case that ground is unfavorable and only acts as loading
against the lining in all directions without providing any type of support. For this simu-
lation, loads in all 17 locations and 34 jacks (two per location) increase to failure. This
load case combined the action of axial loads and bending moments. Experimental results
include load-​displacement curves at the crown measured by the load cell and linear vari-
able differential transformer (LVDT) connected to the jack located at the crown.
 Analysis and design of precast segmental linings 223

3.8 DESIGN FOR SERVICEABILITY LIMIT STATE (SLS)

Conforming to the requirements of the load and resistance factor design (LRFD) method,
the design engineer needs to design precast concrete tunnel segments for the serviceability
limit states (SLS) (JSCE 2007). SLS are states beyond which specified service requirements
for a tunnel are no longer met. The SLS in segmental tunnel lining systems correspond to
excessive stresses, deflections and cracking of concrete segments as well as excessive stresses
and deformations of segment joints. These SLSs may cause reduction of tunnel inner space
due to excessive deflections, and durability and watertightness issues due to reinforcing bar
corrosion and water leakage from segment cracks or enlarged gaps between segment joints
(JSCE 2007; Mendez Lorenzo, 1998; Çimentepe, 2010). Figure 3.152 and Table 3.14 show
a flowchart and design checks for verifying the SLS of tunnel segments. These states and
particularly SLS of cracking are further discussed in this section.

3.8.1 Verification for SLS in tunnel segments

The SLS design for the tunnel segments is performed considering different types and com-
bination of loads acting on tunnel linings from the time of production through the final ser-
vice stage. Different SLS conditions are explained and corresponding calculation methods
and limiting values are presented according to international standards and guidelines.
As previously discussed, critical load cases for segment design include load cases of seg-
ment demolding, storage, transportation and handling, while construction loads include
TBM thrust jack forces, tail skin and localized back grouting pressure. Final service loads
include earth pressure; groundwater and surcharge loads; longitudinal joint bursting load;
and special loads, such as earthquake, fire, explosion and loads induced due to additional
distortion (ACI 544.7R-​16). Design engineers can refer to Table 3.1 and use all possible SLS
load combinations considering proper load factors which are often 1.0 for SLS loads.

Figure 3.152 Design flowchart for serviceability limit state.

Note: Examining stresses, cracks and deformations as the main serviceability limit state design
parameters.
Source: JSCE, 2007. 
224 Handbook of Precast Segmental Tunnel Lining Systems

Table 3.14 Design checks and limiting values for SLS.

Design checks and limiting values for serviceability limit states (SLS) of tunnel
segments

Item SLS states Location Items to check Limiting values

1 Stress Segment Stress in Allowable compressive stress of
Section concrete concrete
Stress in Allowable tensile stress of steel bars
reinforcement
Segment Stress in Allowable compressive stress of
Joints concrete concrete
Stress in Allowable stress of connecting bolts
connectors
2 Deformation Segmental Ring Allowable deformation
Ring deformation
Segment Joint opening Allowable gap between segments joints
Joints
Joint offset Allowable offset between segments
joints
3 Cracking Segment Flexural crack Allowable concrete crack width
Section width
Shear force Shear crack capacity

3.8.2 Stress verification
Critical stresses in the segments at SLS are calculated for a combination of maximum bend-
ing moments and corresponding axial forces. Compressive stresses are limited in the struc-
tural codes to avoid microcracking that may lead to a reduction of durability. Maximum
compressive stresses in both reinforced and fiber reinforced concrete (FRC) at service limit
state are limited to a restricted value of 0.4f′c according to JSCE 2007, and 0.6f′c accord-
ing to EN 1992-​1-​1:2004, AFTES-​WG7:1993, and fib Model Code 2010. On the other
hand, tensile stresses in the reinforcing bar are limited to fy according to JSCE 2007 and to
0.8fy according to EN 1992-​1-​1:2004 and fib Model Code 2010. AFTES-​WG7:1993 limits
reinforcement tensile stresses to 34.8 ksi (240 MPa) for detrimental cracking and 29 ksi
(200 MPa) for highly detrimental cracking.
Flexural stresses in the joints are calculated using maximum bending moments and cor-
responding axial forces at joints as results of analysis such as beam-​spring modeling or FEM
simulations. Developed stresses in the concrete at segment joints are limited to allowable
compressive stress of concrete. Developed stresses and forces in the bolts are limited to
allowable stress of connecting bolts reported by the manufacturer.

3.8.3 Deformation verification
Segment deformations are obtained directly as results of different models presented in
Sections 3.5.2 through 3.5.4 for the load case of earth pressure; groundwater; and surcharge
loads including elastic equation, beam-​spring, finite element method and distinct element
method. However, joint gap and joint offset are only obtained from models that simulate
joints between segments and rings. For serviceability limit state (SLS) verification, these
 Analysis and design of precast segmental linings 225

Figure 3.153 Ring deformation criteria.

Note: Ring deformation criteria defined with respect to locations on a completely circular ring
before taking any loads.
Source: Adopted from ÖVBB 2011. 

deformations are limited to allowable values recommended by standards, guidelines and

often project specifications. ÖVBB 2011 recommends allowable deformations, shown in
Figure 3.153, for both segments and joints of tunnels with a diameter of up to 26 feet (8 m).

3.8.4 Cracking verification
Cracking in segments is a major cause for reduction in serviceability due to reduction
of watertightness and reinforcement corrosion. In particular, cracking has a significant
effect on the durability of the tunnel in an environment with frequent freeze–​thaw cycles.
Examination using appropriate methods should be carried out to ensure that cracking in
segments does not impair the serviceability, durability, or intended purposes of the tunnel
lining (Bakhshi and Nasri, 2015a, 2015b). Among possible cracks induced in segments
under service loads are mainly the cracks due to bending moment and axial force. Cracking
should be examined by ensuring that the flexural crack width is not greater than the allow-
able crack width. The flexural crack width calculation for reinforced concrete (RC) and
fiber reinforced concrete (FRC) segments is presented in the following.

3.8.5 Flexural crack width in segments

Crack width in tunnel segments due to bending moment and axial force is calculated using
ACI 224.1R-​07, JSCE 2007 and EN 1992-​1-​1:2004 formulas as shown in Equations (3.33
through 3.35), respectively.

Equation (3.33)

2
fs  s
w = 0.076βfs 3 dc A × 10−3 and w = 2 β dc 2 +   (in.-​lb)
Es  2
226 Handbook of Precast Segmental Tunnel Lining Systems

2
fs  s
w = 0.011βfs 3 dc A × 10−3 and w = 2 β dc 2 +   (SI)
Es  2

Equation (3.34)

f   15  5 (n + 2)
w = s  s + ε csd
′  ; s > 13.97  + 0 .7  ⋅ ⋅
 Es   0.006895fc + 20
’
 7n + 8

(101.6dc + 17.78( s − ϕ)) (in.-​lb)

f   15  5 (n + 2)
w = s  s + ε csd
′  ; s > 0.55  ’ + 0 .7  ⋅ ⋅ ( 4 ⋅ dc + 0.7 ⋅ ( s − ϕ )) (SI)
 Es   fc + 20  7n + 8

Equation (3.35)

 fct ,eff  E A 
 fs − kt  As 


1+ s ⋅ s  
E A
   
A  fs 
w = sr, max  ≥s
r , max  0.6
 Es   Es 
 
 
 

fib Model Code 2010, CNR DT 204/​ 2006, RILEM TC 162-​ TDF, and DAfStB
Stahlfaserbeton:2012-​11 can be used to calculate the crack width in concrete sections rein-
forced by fibers with and without conventional reinforcement. The flexural crack width of
FRC segments has been well presented in fib Working Party 1.4.1 2017 by considering ana-
lytical sectional approaches as well as finite element methods.

3.8.6 Maximum allowable crack width

Cracking in tunnel segments is controlled by limiting the crack width to specific levels to
prevent durability issues due to increased permeability, excessive water leaks and reinforce-
ment corrosion. The allowable crack widths are recommended by standards and guidelines
considering the function, importance, service, life span, purpose, surrounding environment
and surrounding soil conditions of the tunnel (JSCE 2007). ACI 224.1R-​07 limits allowable
cracks in structures exposed to the soil to 0.012 inches (0.30 mm). Similar to ACI 224.1R-​
07, EN 1992-​1-​1:2004 recommends an allowable crack width of 0.012 inches (0.30 mm)
for reinforced concrete members.
In a more restricted manner, fib Model Code 2010 limits the allowable crack width to
0.008 inches (0.2 mm) if leakage into the structure is to be limited to a small amount and
only some surface staining is acceptable. Among references specific to tunnel segments are
LTA 2010, DAUB:2013 and JSCE 2007 that specify the allowable crack width to 0.01 inches
(0.30 mm), 0.007 inches (0.2 mm) with 0.006 inches (0.15 mm) when below the ground-
water table, and 0.004dc, respectively, where dc is the concrete cover over the reinforcing
bar. As the most comprehensive guideline, ÖVBB 2011 specifies the allowable crack width
 Analysis and design of precast segmental linings 227

Table 3.15 Allowable crack widths.

Allowable crack widths for tunnel segments

Requirement Allowable
Item class Designation Application Requirement crack width
1 AT-​1 Largely dry • One-​p ass lining with Impermeable 0.008”
very tight waterproofing (0.20 mm)
requirements
• Portal areas
2 AT-​2 Slightly • One-​p ass lining for road Moist, no 0.010”
moist and railway tunnels with running water (0.25 mm)
normal waterproofing in tunnel
requirements (excluding
portals)
3 AT-​3 Moist • One-​p ass lining Water 0.012”
without waterproofing dripping (0.30 mm)
requirements from
• Two-​p ass lining systems individual
spots
4 AT-​4 Wet • One-​p ass lining Water 0.012”
without waterproofing running (0.30 mm)
requirements in some
• Two-​p ass lining as a places
drained system
Source: Per ÖVBB 2011.

in segments based on the tunnel function and corresponding watertightness requirements,

which is shown in Table 3.15.

3.8.7 Verification of current crack width criteria

Bakhshi and Nasri (2017b) provides an analytical methodology for prediction of water
inflow into cracked concrete linings. The method is sensitive to all major physical factors
and focuses on the mechanism of water flow through the cracks by introduction of formulas
calibrated by several different experimental data. Best fits to experimental data were pre-
sented by introducing upper and lower bound factors to the water flow rate coefficient as
an important factor for determining water inflow rate. Effect of different types of reinforce-
ment, including conventional reinforcing bars or fiber reinforcement vs. plain unreinforced
concrete on controlling water inflow rate in cracked samples was introduced in a quantita-
tive manner.
To verify validity of current crack width criteria, water inflow rates corresponding to the
allowable crack widths were calculated for a midsize tunnel. Results show that the crack
width is not the only dominant factor impacting the water inflow rate. Water pressure
head, segment thickness, length and water temperature have also significant effects on the
water infiltration into tunnels. Results also imply that, without consideration of autogenous
healing, inflow rate may not be in agreement with the qualitative designations provided in
codes and guidelines such as ÖVBB 2011 by setting fixed allowable crack width ranges.
However, because autogenous healing in cracked concrete results in significant reduction
of water inflow with time, further studies are needed to verify the validity of current crack
width criteria.
228 Handbook of Precast Segmental Tunnel Lining Systems

On the other hand, results of parametric studies and application to a case of a mid-
size tunnel show a significant impact of fiber reinforcement in controlling the initial water
inflow. The proposed methodology indicates that in addition to the superior effect of fiber
reinforcement in controlling the cracks under service load, compared to conventional
reinforcement, fiber reinforcement also significantly improves the watertightness of cracked
concrete lining most likely due to change in morphology of cracks. The proposed approach
can be used to provide project-​based allowable crack width for concrete tunnel lining based
on parameters that are specific to the project including the type of reinforcement.

3.9 DESIGN FOR SUSTAINABILITY

Carbon footprint analysis is becoming more and more popular in every industry due to
increasing concerns about global warming and greenhouse gas (GHG) emissions. The con-
struction industry is a major producer of CO2 emissions. In the U.S., the construction pro-
cesses generate the third highest greenhouse gas (GHG) emissions among industrial sectors.
Among various construction processes, tunnel construction produces a significant amount
of carbon emissions arising from the embodied carbon in the primary construction mate-
rials –​cement and steel and utilizing various types of high energy-​consuming equipment.
Research shows that more than 80% of the CO2 emissions in the construction phase of a
tunnel are attributable to the construction materials cement and steel. Thus, reducing the
need for cement and steel can make a significant contribution to reducing CO2 emissions.
This comes along with a significant reduction in construction costs. Materials aside, the
embodied energy and emissions during construction processes are mainly attributable to the
vehicles and machinery used. To identify and mitigate such carbon emissions of a tunneling
project, it is required to reliably estimate the carbon footprints of a tunneling project. In the
next sections, the detail of this estimations is explained.

3.9.1 Embodied carbon emission of the material

Materials are responsible for a significant share of carbon emissions in the construction of
a tunnel project. Two of the largest contributors to the embodied carbon of a tunnel are
concrete and steel reinforcement used in tunnel lining. The CO2 emission of a material can
be calculated using the following formula:

Mass of CO2 − eq = CO2 − eq factor × material quantities (3.36)

where the CO2-​eq factor is the amount of equivalent CO2 per pound or kg of material that
comes from environmental product declarations (EPDs).
In the following, different ways of reducing CO2 emissions are presented.

(a) Reduction of cementitious materials or use supplementary cementitious materials

For concrete, 90% of CO2 in the mixture comes from the type of cement and the supple-
mentary cementitious materials (SCMs) in a mixture. Therefore, we can control 90% of our
CO2 issues by paying attention to how much and the type of cement and the type of SCM
we use. Table 3.16 shows the CO2 eq factor for different types of cement and SCMs.
Using Portland Lime Cement (PLC) instead of Portland Cement and high SCM can reduce
the CO2 emission by 7% and 45%, respectively (a mixture is considered high SCM when
slag is greater than 50% of total cementitious materials or fly ash is greater than 30% of
 Analysis and design of precast segmental linings 229

Table 3.16 CO 2-​e q factors.

CO 2-​e q factors for different types of cement and SCMs

Biner CO 2-​e q factor

Portland cement 0.92
Portland limestone cement 0.85
Slag 0.15
Fly ash 0.093
Silica fume 0.014

total cementitious materials or slag is greater than 40% and fly ash is greater than 20%
of total cementitious materials). High levels of cement replacement by SCMs will result in
lowering the early strength of concrete. High early-​strength concrete is required for demold-
ing segments within 5–​6 hours after casting (for example, 15 MPa after 5–​6 hrs). This is
the focal point of today’s industry research and so far, the only solution has been using an
alkali-​activated binder (AAB) based on sodium carbonate. The concern with the use of AAB
is the risk of the alkali–​silica reaction (ASR) and long-​term durability issues.

(b) Optimizing aggregate quantities

Traditionally aggregates are designed according to ASTM standards based on coarseness
factor charts, which were not reliable at all. Tarantula Curve is a new and effective tool
that was developed based on all workability tests including slump, the box test, ICAR
Rheometer, visual observation, and float test. This helps optimize the aggregate amount in
the concrete mixture and thus reduction of paste (total cementitious content and water).
The carbon emission can be reduced by up to 14%.

(c) Using fiber reinforcement instead of steel reinforcement

The steel fiber reinforcement can significantly reduce the embodied carbon in segmental lin-
ings, compared to bar reinforcement. Steel rebar mass in unit concrete volume of tunnel seg-
ments changes between 90 to 160 kg/​m3. Using Equation (3.36) and considering the CO2-​eq
factor of 1.85 for steel, carbon emission is about 166.5 to 296 kg/​m3. In comparison, the
mass of steel fibers (Dramix 4D 80/​60BGP) in the unit concrete volume of tunnel segments
is about 40 kg/​m3 and the CO2-​eq factor is 0.88, which gives carbon emission of 35.2 kg/​
m3. This means a 4.7 to 8.4 times reduction in CO2 emission.

(d) Using the carbon cure method

Carbon is performed by injecting captured carbon dioxide into concrete. It is a new tech-
nology that when easily accessible, can together with other technologies reduce unit CO2
emission of Concrete by as much as 71%

3.9.2 CO2 emission of the construction phase

There are two common sources of emission for equipment; the diesel fuel consumed to
move in and out of the tunnel in every cycle and the electricity consumed by the equipment
during operation. Explosives, which may be used during the drilling and blasting advancing
method, are another source of CO2 emission. In the following, the equations for calculating
230 Handbook of Precast Segmental Tunnel Lining Systems

these emissions are presented. These equations will be applied for all diesels driven and elec-
trical equipment that will be studied later.

3.9.2.1 Carbon emissions calculation equations

(a) CO2 emission from fuel consumption
For equipment speeds between 5 and 25 km/​h, the diesel consumption DCi (in grams) will
be the result of multiplying its weight Mi (in tons) by the distance the equipment goes back
and forth 2 × d (where d is the distance from the working face to the portal in km) and by
the fuel consumption per ton moved and km ran Cg (in g/​t × km):

DCi = Mi × 2 × d × C g (3.37)

where C g can be calculated using Equation (3.38).

C g ≈ 7 ( fR ± i ) (3.38)

A typical value for the friction coefficient in a civil work road is fR =​3.5%. The sign of the
slope will be negative when the equipment goes downwards and will be positive upwards.
The consumption in grams can be translated into liters, using the diesel density (833 g/​l).
The CO2 emissions ( DEj ) can be calculated multiplying the consumption of this equipment
by its conversion factor ( rD ):

DEi = ( DCi /833) × rD (3.39)

rD can vary depending on the region and data uses. A typical value of 3.25 kgCO2 per liter
of fuel for mining equipment can be assumed (Shillaber et al., 2016).
For equipment speed higher than 25 km/​h (for example, for transportation from the pro-
duction factories to the construction site), the consumption, in grams, will be given using
the following formula taking into consideration two different roads, the outside and the
inside of the tunnel:

(outside) Cg MatO = d0 × (mT + mL ) × (15 − Cgo ) + d0 × mT ×  15 −

V0 
 (3.40)
20 

 V 
(inside) Cg Mati ( )
= di × ( mT + mL ) × 15 − C gi + di × mT ×  15 − i 
 20 
(3.41)

where d0 is the distance travelled outside in km, v0 is the speed of the truck outside of the
tunnel (we can consider here the average speed of 60 km/​h), di is the distance travelled inside
in km, vi is the speed of the truck inside of the tunnel (10 km/​h), mT and mL are the weight
of the truck and the load respectively. Cgi is the amount of diesel consumed (grams) per ton
transported and km driven and can be calculated using the following formula:

vi
C gi ≈ 15 − (3.42)
20
 Analysis and design of precast segmental linings 231

Figure 3.154 CO 2 produced by muck pile transportation.

Source: Rodríguez et al., 2020.
Note: Accumulated CO 2 emissions versus advanced length graph for muck pile transportation.

To have an idea, the accumulated quantity of CO2 emission due to the articulated trucks
used to remove the rock from a tunnel is presented in Figure 3.154 (Rodriguez, 2020). The
tunnel taken as a reference is a double tube tunnel in Northwest of Spain with a slope of
about 1% upwards. The depth was 125 m in average.

(b) CO2 emission from electricity

The electrical energy consumption ( EC J ) in kWh can be calculated using the following
equation:

ECi = NP × LF × t (3.43)

where NP is the nominal power in kW, LF is the load factor (%), and t is the working time
in hours.
The emissions ( EEJ ) can be calculated multiplying the consumption of the electricity by
its conversion factor ( rE ):

EEi = ECi × rE (3.44)

A typical value for rE is 0.267 kg CO2 per kWh according to the Spanish institute IDAE
(2011). When using electric generator, it can be assumed as 0.66 kgCO2/​kWh. These
values depend on the specific regional characteristics, other authors suggest different
emissions rate: i.e., Shillaber et al. (2016) suggest a conversion factor 0.981 kgCO2/​kWh
for electric generators. To have an idea, the accumulated quantity of CO2 emission by
the mobile electric generators used in the reference tunnel is presented in Figure 3.155
(Rodriguez, 2021).
232 Handbook of Precast Segmental Tunnel Lining Systems

Figure 3.155 CO 2 produced by electric generators.

Source: Rodríguez et al., 2020.
Note: Accumulated CO 2 emissions versus advanced length graph for electricity generation.

(c) Emission produced during blasting

The CO2 produced during the blasting of the explosives will be calculated multiplying the
amount of explosive ( mexp ) used during the construction of the tunnel, by its conversion
factor rex.

Eexp = mexp × rex (3.45)

The conversion factor rex is about 0.258 kgCO2 per kg of explosive used (manufacturers
Maxam 2015). If emissions due to manufacturing should be included, this rate should be
2.0 kgCO2 per kg of explosive. In the case of tunnels with a cross sectional area bigger
than 60 m2, the amount of explosive needed according to different researchers (Cardu and
Seccatore, 2016) varies from 0.5 to 1.5 kg/​m3. To have an idea, the accumulated quan-
tity of CO2 emission by explosive detonation used in the reference tunnel is presented in
Figure 3.156 (Rodriguez, 2021)

3.9.2.2 Carbon emissions of different excavation methods

(a) Drilling and blasting method
The emissions will come from energy consumed by the equipment used for this task, the
jumbo (used to drill the blasting holes), TEJ and the platform (used to load the explo-
sives) TEPL , and the CO2 released during blasting Eexp (Rodríguez et al., 2017). The total
emissions produced during the advancing cycle with this method will be

TEDB = TEJ + TEPL + Eexp (3.46)

The jumbo has two sources of emissions. The first is the diesel fuel consumed to move in
and out this equipment of the tunnel in every cycle. Equation (3.39) can be used for esti-
mating this emission ( DEj ) . The second source of emissions is the electricity consumed
 Analysis and design of precast segmental linings 233

Figure 3.156 CO 2 produced by explosive detonation.

Source: Rodríguez et al., 2020.
Note: Accumulated CO 2 emissions versus advanced length graph for explosive detonation.

by the jumbo. It uses an electrical motor to drill the blasting hole. Equation (3.44) can
be used for this emission ( EEj ). The total emissions of the jumbo during its work will be
TEJ =​ DEJ +​ EEJ .
The platform to load the explosives uses diesel to move in and out of the tunnel and
to load the explosives into the blasting holes in every cycle. Its consumption ( DEPL ) will
be the same as Equation (3.39), and its emissions ( EEPL ) will be calculated as shown in
Equation (3.44). The total emissions of the platform during its work will be TEPL =​ DEPL +​
EEPL .
The CO2 produced during the blasting of the explosives will be calculated using
Equation 3.41.

(b) Excavation with road header

The road header uses electrical energy to perform its task. The cutting head is moved by
an electrical motor, and the energy consumption and the CO2 emissions will be given using
Equation (3.41). To calculate the load factor LF, we need to take into consideration the
RMR of the tunnel strata. For the working time t, in hours, we need to measure the real
timing that the equipment is running, and it will depend mainly on the quality of the rock,
if that is good we can spend most part of that working with the equipment, however if the
quality is bad, we will spend most part of the working time installing the support at the face
with a very few use of this equipment.

(c) Excavation with hydraulic break hammer

The hydraulic breaker hammer uses a Diesel engine to perform its work. Several studies
were done around its diesel consumption, and for the typical equipment of 1500 kg and
18 KW, on a 25t excavator its consumption is about 36 L/​hour (Rodríguez et al., 2017).
The emissions (DEHBH) will be calculated multiplying its consumption rate (36 L/​h), by the
number of working hours t and by the diesel conversion factor rD, as follows:

DEHBH = DCHBH × rD = 36 × t × rD (3.47)

234 Handbook of Precast Segmental Tunnel Lining Systems

Table 3.17 Summary of emission rates.

Emission rates for seven tunnels with different excavation methods (kgCO 2/ ​m)

Excavation method used (%)

Cross section Emissions rate
Tunnel RMR D&B RH HBH Length (m) (m 2) (kgCO 2/ ​m)
Somao 25–​4 5 10 0 90 466 142.5 15284
Fresno 25–​4 5 0 0 100 940 125 14417
Fabares 30–​5 5 18 72 10 1922 92.1 10276
Padrún 25–​5 5 98 0 2 1762 88.5 10261
Santiuste 25–​6 5 60 0 40 500 113.7 8710
Penaflor 21–​5 0 44 0 56 695 99.4 8260
Folledo 35–​7 0 100 0 0 2021 54 4293
Source: Rodríguez et al., 2020.

Table 3.18 Total power of TBM.

Total power of TBM for different tunnel diameter

No. Tunnel diameter Cutter head power Total power

1 11.5 ft 1,340 kW (1,836 hp) 2,010 kW
2 13 ft 2,100 kW (2,816 hp) 3,200 kW
3 14.1 ft 2,345 kW (3,143 hp) 3,517 kW
4 14.1 ft 960 kW (214.6 hp) 1,440 kW
5 15.1 ft 1,260 kW (1,690 hp) 1,890 kW
6 15.8 ft 2,275 kW (3,050 hp) 3,413 kW
Source: Adapted from Robbins, 2017.

For the working time, it will also depend on the quality of the rock as we mentioned above.
For each specific case, the working time needs to be measured and the emissions will be
evaluated.
Rodríguez et al. (2020) calculated the carbon emission of seven different tunnels exca-
vated in Northwest Spain. The summary of the ratio emissions for these tunnels (kg/​m) can
be seen in Table 3.17.

(d) Tunnel boring machine (TBM)

Some factors considered for design of TBM are machine diameter, length of the machine,
segment layout, quantity of segments in ring, ground conditions, minimum segment width
and thickness (Lovat, 2017).
Various types of TBMs have specific power requirements. Table 3.18 shows the total cut­
ter head power for different TBM diameters. The CO2 per kWh for electric power packs
using hydraulic motor depends on the area and it changes between 106 and 750 grams of
CO2 per kWh (EIA, 2017).
As an example, assuming TBM has an average production rate of 100 feet per 20-​hour
shift to allow 4 hours for maintenance and total CO2 emission for 100 feet advancement of
a 13 feet tunnel can be calculated as follows:
 Analysis and design of precast segmental linings 235

grCO2 hr CO2
3200 kW × (106 − 750) × 20 = 6, 780 − 48, 000 kg.
kWh 100ft 100ft
= 222 − 1575 kg CO2 /m

3.9.2.3 Emission released during loading and transportation

The emissions will come from the diesel consumed by the loader that is taking the muck pile
into the truck or dumper that is waiting to be loaded. Regarding the diesel consumed by the
loader, we will use 0.15 L/​h per kW of net equipment power (Salam et al., 2015). Thcen the
diesel consumption (DCMPC), in liters, will be

DCMPC = 0.15 × NP × t (3.48)

where NP is the net equipment power in kW, and t is the number of hours working.
Regarding the truck or dumper waiting on low motion to be loaded, can be negligible,
only 2.64 l/​h are used. Then, the consumption of the truck (in liters) in waiting mode
will be

DCTLM = 2.64 × t (3.49)

The CO2 emissions of this task DEL can be calculated using the conversion factor rD (a
typical value for rD is suggested in 3.9.2.1.

DEL = (DCMPC + DCTLM ) × rD (3.50)

3.9.2.4 Rock waste transportation

The emissions will come from the diesel consumed by the truck or dumper from the front
of the tunnel to the waste dump. The equipment used to transport the rock removed from
the face of the tunnel to the outside dump will use diesel. Their carbon emissions can be
calculated using the same formulas in 3.37 and 3.38 considering the weight of the truck and
the rocks going outside of the tunnel, and only the weight of the truck returning back to the
face of the tunnel. If the distance from the working face to the portal is d, the distance from
the portal to the out-​waste dump is d’, the truck weight is Mt and the rock weight is Mr ,
in tons, the diesel consumption in grams will be

DCT = (Mr + Mt ) × C g + Mt × ( d + d ) × C g (3.51)

We will consider the speed of 10 and 30 km/​h inside and outside the tunnel respectively. Cg
can be calculated based on equations presented in Section 3.9.2.1
The emission will be calculated multiplying the consumption by the CO2 conversion
factor rD :

DET = (DCT /833) × rD (3.52)

236 Handbook of Precast Segmental Tunnel Lining Systems

3.9.3 Example: carbon footprint emission saving in Montreal

Blue Line Extension
The PLB project consists of five stations and a TBM tunnel. The tunnel is approximately
6 km long. It is 15 m deep on average and will be excavated through rock. In line with the
commitment to integrating sustainability best practices, low-​carbon concrete and shotcrete
mixtures are proposed for the linings of TBM and SEM tunnels as well as stations and aux-
iliary structures.
In the baseline design, an arc shape section with cast-​in-​place (CIP) concrete lining using
ordinary portland cement (OPC) concrete mixture was proposed. To reduce carbon emis-
sions, an alternative case was suggested. In this case, while a similar section was assumed,
CIP lining was replaced with low-​carbon shotcrete lining.
TBM bored tunnel is another option instead of arc shape CIP section. For this case, an
external diameter of 9.3 m with 7+​1 precast segmental lining system and 1.8 m width was
proposed. For this case, changing from an OPC concrete mixture into a low-​carbon con-
crete mixture for prefabricated segments is suggested.
In the following, detail of concrete mixtures for baseline and low-​carbon case and cor-
responding CO2 emissions are presented.

3.9.3.1 CIP section-​baseline case

Figure 3.157 shows the tunnel geometry in the baseline design. It consists of three compo­
nents. 75 mm temporary shotcrete, 40 cm CIP lining, and 32 cm thick invert.
The baseline concrete mixture consists of ordinary Portland cement (OPC) with no slag
and silica fume. Table 3.19 presents the mixture design for temporary shotcrete, CIP arc

Figure 3.157 Tunnel geometry –​ CIP section.

Note: Tunnel geometry of CIP section consisting of arc shape section and invert.
 Analysis and design of precast segmental linings 237

Table 3.19 Concrete mixture designs –​ baseline case.

Concrete mixture designs for temporary shotcrete, CIP arc and invert-​b aseline case

Cementitious Portland Rebar Total

material cement Aggregate Water/​ (kg/​ Fiber–​ Admixtures (kg/​
(kg/​m 3) (kg/​m 3) (kg/​m 3) cement m 3) (kg/​m 3) (kg/​m 3) m 3)
Temporary 475 475 1430 0.38 20 –​ 4.5 2108.6
shotcrete
CIP arc 415 415 1800 0.41 –​ 62 3.94 2451.1
Invert 415 415 1786 0.41 –​ 76 3.94 2451.1

Table 3.20 CO 2 emission of concrete mixture designs –​ baseline case.

CO 2 emission of concrete mixture designs for temporary shotcrete, CIP arc and
invert –​ baseline case

Baseline temporary Baseline- ​C IP arc Baseline-​i nvert

shotcrete concrete concrete
CO 2-​e q Mass/​m 3 CO 2 eq/​ Mass/​m 3 CO 2 eq/​ Mass/​m 3 CO 2 eq/​
Components factor* (kg) m 3 (kg) (kg) m 3 (kg) (kg) m 3 (kg)
Portland 0.92 475 437 415 382 415 368
cement
Admixtures 1.67 4.5 7.5 3.94 6.6 3.94 6.3
Aggregate 0.006 1430 8.6 1800 1786
Steel bar 1.85 –​ –​ 62 114.7 76 140.6
Fiber (for 0.86 20 17.2 –​ –​ –​ –​
initial
support)
Total: 470.3 513.9 539.7

and invert. Using the presented method in Section 3.9.1, the CO2 emission per m3 of these
mixtures and the total CO2 can be calculated (Tables 3.20 and 3.21).

3.9.3.2 CIP section-​low carbon case

To reduce the carbon emission while having the same strength properties, Portland cement
was replaced with limestone Portland cement and 27% of the cementous material was
replaced with slag and silica fume (22% slag and 5% silica fume, which is considered mod-
erate supplementary cementitious materials or SCM). This mixture was proposed for both
shotcrete and invert. Table 3.22 presents this concrete mixture design. Using the presented
method in section 3.9.1, the CO2 emission per m3 of these mixtures and the total CO2 can
be calculated (Table 3.23 and Table 3.24).
Comparing the total CO2 emission of this case with the baseline, the total saving is 79%.

3.9.3.3 TBM tunnel with fabricated segmental lining

The baseline concrete mixture consists of ordinary Portland cement (OPC) with no slag
and silica fume. To reduce the carbon emission while having the same strength properties,
238 Handbook of Precast Segmental Tunnel Lining Systems

Table 3.21 Total embodied carbon footprint-​b aseline case.

Total embodied carbon footprint per 1 m length of the tunnel for temporary
shotcrete, CIP arc concrete, invert concrete –​ baseline case

Volume (m 3)/​1 CO 2 eq/​m 3 CO 2 eq/​1 m

Thickness (mm) m tunnel (kg) length (ton)
Temporary shotcrete 325 5.9 470.3 2.8
CIP arc concrete 400 7.03 513.9 3.61
Invert concrete 320 2.854 539.7 1.54
Total =​ 7.9

Portland cement was replaced with limestone Portland cement and 27% of the cementous
material was replaced with slag and silica fume (22% slag and 5% silica fume). Table 3.25
presents these concrete mixture designs.
The baseline backfill/​infill concrete consists of ordinary Portland cement (OPC) with no
slag and silica fume. In the carbon-​reduced case, 50% of the Portland cement was replaced
with slag. Table 3.26 presents the backfill/​infill concrete mixtures for these two cases.
The tail-​void backfill grout design is presented in Table 3.27. The same design is used both
for the baseline case and the carbon-​reduced case.
Table 3.28 and Table 3.29 summarize the detail of the total embodied carbon footprint
of the tunnel based on the geometry of the tunnel and unit CO2 emission presented in the
previous tables for baseline mixture design and SCM mixture design.
Based on Table 3.29, the total carbon footprint of the two cases, baseline design, and
carbon reduced design is calculated as below:

Total CO2 eq /1m tunnelbaseline design = 6.7 + 2.2 + 1.4 = 10.3 ton

Total CO2 eq /1m tunnelcarbon reduced design = 3.6 + 1.4 + 1.4 = 6.4 ton

As is seen, by using the carbon-​reduced concrete mixture for segments and backfill/​infill
concrete, the total carbon footprint per 1 m length of the tunnel is reduced by 38%.

BIBLIOGRAPHY

AUTHORITY DOCUMENTS
American Association of State Highway and Transportation Officials (AASHTO)
AASHTO DCRT-​1-​2010 Technical Manual for Design and Construction of Road Tunnels–​Civil
Elements
AASHTO LRFD-​1-​2017 LRFD Road Tunnel Design and Construction Guide Specifications
American Concrete Institute
ACI 207.1-​21 Mass Concrete –​Guide
ACI 224.1R-​07 Causes, Evaluation, and Repair of Cracks in Concrete Structures
ACI 318-​14 Building Code Requirements for Structural Concrete and Commentary
ACI 533.5R-​20 Guide for Precast Concrete Tunnel Segments
ACI 544.7R-​16 Report on Design and Construction of Fiber-​Reinforced Precast Concrete Tunnel
Segments
 newgenrtpdf
Table 3.22 Shotcrete mixture design –​ low carbon case.
SCM shotcrete mixture design for CIP section –​ low carbon case

Analysis and design of precast segmental linings

Cementitious Portland Silica Fiber-​D ramix
material Cement Slag fume (kg/​ Aggregate Water/​ 3D 65/​3 5BG Admixtures Total
(kg/​m 3) (kg/​m 3) (kg/​m 3) m 3) (kg/​m 3) cement Rebar (kg/​m 3) (kg/​m 3) (kg/​m 3) (kg/​m 3)
Shotcrete-​ 475 346.75 104.5 23.75 1430 0.38 –​ 40 4.51 2128.6
SCM

239
240 Handbook of Precast Segmental Tunnel Lining Systems

Table 3.23 CO 2 emission of shotcrete mixture –​ low carbon case.

CO 2 emission of SCM shotcrete mixture design for CIP section –​ low carbon case

CO 2-​e q CO 2 eq/​m 3 % Replacement

Components factor Mass/​m 3 (kg) (kg) by mass
Portland limestone cement 0.85 346.75 295
Slag 0.1466 104.5 15 22%
Fly ash 0.093   0.0 –​ 0%
Silica fume 0.014   23.75   0.3 5%
Admixtures 1.67     4.51   7.5 1%
Aggregate 0.006 1430   8.6
Steel bar 1.85 –​ –​
Fiber-​3 D 65/​3 5BG 0.86   40 34.4
Total 361

Table 3.24 Total embodied carbon footprint –​ low carbon case.

Total embodied carbon footprint per 1 m length of the tunnel for temporary
shotcrete, shotcrete lining and invert concrete –​ low carbon case

Thickness Volume (m 3) /​1 CO 2 eq/​m 3 CO 2 eq/​1 m length

(mm) m tunnel (kg) (ton)
Temporary shotcrete 160 1.8 361 0.65
+​ Shotcrete lining
Invert concrete 32 2.854 361 1.03
Total =​ 1.68
 newgenrtpdf
Table 3.25 Concrete mixture designs –​ TBM tunnel segments.
Concrete mixture designs of OPC and SCM segments –​ TBM tunnel

Portland Fiber-​
Cementitious Portland limestone Silica Dramix 4D

Analysis and design of precast segmental linings

material (kg/​ Cement cement Slag (kg/​ Fume Aggregate Water/​ Rebar 80/​6 0BGP Admixtures Total
m 3) (kg/​m 3) (kg/​m 3) m 3) (kg/​m 3) (kg/​m 3) cement (kg/​m 3) (kg/​m 3) (kg/​m 3) (kg/​m 3)
Segment-​ 475 475 –​ –​ –​ 1430 0.38 80 –​ 4.5125 2168.59
OPC
Segment-​ 475 –​ 346.75 104.5 23.75 1430 0.38 –​ 40 4.5125 2128.59
SCM

241
242 Handbook of Precast Segmental Tunnel Lining Systems

Table 3.26 Backfill/​infill concrete mixture designs –​ TBM tunnel.

Backfill/​infill mixture designs of OPC and carbon reduced concrete –​ TBM tunnel

Cementitious Portland Slag

material cement (kg/​ Aggregate Water/​ Water Admixtures Total
(kg/​m 3) (kg/​m 3) m 3) (kg/​m 3) cement (kg/​m 3) (kg/​m 3) (kg/​m 3)
Baseline 220 220 –​ 1950 0.8 176 2.09 2348.09
concrete
(OPC)
Carbon 220 110 110 1950 0.8 176 2.09 2348.09
reduced
concrete

Table 3.27 Tail-​v oid grout mixture design –​ TBM tunnel

Cement +​
bentonite+​ Portland Rebar
stabilizer cement Bentonite Stabilizer Water/​ (kg/​ Admixtures Total
(kg/​m 3) (kg/​m 3) (kg/​m 3) (kg/​m 3) cement m 3) (kg/​m 3) (kg/​m 3)
Baseline-​ 325 325 42.5 5 2.55 0 3.0875 1205.59
grout

Table 3.28 Total embodied carbon footprint segmental lining –​ TBM tunnel.

Total embodied carbon footprint of 40 cm thickness OPC rings and 35 cm thickness
SCM rings –​ TBM tunnel

Ring Tunnel Ring Total CO 2 CO 2 eq/​1

width length D-​e x D-​i n volume concrete eq/​m 3 m tunnel Total CO 2
(m) (m) (m) (m) (m 3) volume (m3) (kg) (ton) eq (ton)
40 cm 1.8 4086 9.3 8.5 20.1 45698.1 601 6.7 27,469.9
thickness
Rings-​O PC
35 cm 8.6 17.7 40210.5 363 3.6 14,592.6
thickness
Rings-​S CM

Table 3.29 Total embodied carbon footprint of the backfill/​infill and tail void backfill grout –​
TBM tunnel.
Total embodied carbon footprint of the OPC backfill/​infill, OPC backfill/​infill with
50% slag and tail void backfill grout –​ TBM tunnel

Volume (m 3/​ Total CO 2 eq/​1

1 m tunnel Tunnel volume CO 2 eq/​ m tunnel Total CO 2
length) length (m) (m 3) m 3 (kg) (ton) eq (ton)
Backfill/​i nfill with 10.25 4086 41881.5 218 2.2 9,113
OPC concrete
mixture
Backfill/​i nfill with 10.25 4086 41881.5 133 1.4 5,550
OPC concrete
mixture –​ 50% slag
Tail-​b oid backfill 4.45 4086 18195.8 315 1.4 5,725.5
grout-​b aseline
 Analysis and design of precast segmental linings 243

BIBLIOGRAPHY
Codes, Standards and Guidelines

American Society of Civil Engineers

American Society of Civil Guidelines for the seismic design of oil and gas pipeline
Engineers. Committee on Gas systems
and Liquid Fuel Lifelines.

ASTM International
ASTM A615/​A615M Standard Specification for Deformed and Plain Carbon-​Steel Bars for
Concrete Reinforcement
ASTM A706/​A706M Standard Specification for Deformed and Plain Low-​Alloy Steel Bars for
Concrete Reinforcement
ASTM A1064/​A1064M 
Standard Specification for Carbon-​ Steel Wire and Welded Wire
Reinforcement, Plain and Deformed, for Concrete
ASTM A820/​A820M Standard Specification for Steel Fibers for Fiber-​Reinforced Concrete
ASTM C39/​C39M Standard Test Method for Compressive Strength of Cylindrical Concrete
Specimens
ASTM C496  Standard Test Method for Splitting Tensile Strength of Cylindrical
Concrete Specimens
ASTM C618 Standard Specification for Coal Fly Ash and Raw or Calcined Natural
Pozzolan for Use in Concrete
ASTM C989 Standard Specification for Slag Cement for Use in Concrete and Mortars
ASTM C1609/​C1609M 
Standard Test Method for Flexural Performance of Fiber-​ Reinforced
Concrete (Using beam with Third-​Point Loading)

Austrian Society for Concrete and Construction Technology

ÖVBB 2011 Guideline for Concrete Segmental Lining Systems

Concrete Society
CS TR63 2007 Guidance for the Design of Steel-​Fibre-​Reinforced Concrete

French Ministry for the Economy and Finance–​Béton armé états limites
BAEL 91 Révisé 99:2007 Considerations on the Usual Methods of Tunnel Lining Design Working
Group No.7―Temporary Supports and Permanent Lining

French Tunneling and Underground Engineering Association

AFTES-​WG7:1993 Considerations on the Usual Methods of Tunnel Lining Design. Working
Group No.7―Temporary Supports and Permanent Lining
AFTES:2005  Recommendation for the Design, Sizing and Construction of Precast
Concrete Segments Installed at the Rear of a Tunnel Boring Machine (TBM)

Dutch Standardization Institute

NEN 6720:1995 Regulations for Concrete―Structural Requirements and Calculation Methods

European Committee for Standardization

EN 1992-​1-​1:2004 Eurocode 2: Design of Concrete Structures―Part 1-​1: General Rules and
Rules for Buildings
EN 1992-​1-​1:2004 Design of Concrete Structures–​Part 1-​2: General Rules–​Structural Fire Design
EN 14651:2005+​A1:2007 Test Method for Metallic Fibre Concrete. Measuring the Flexural Tensile
Strength (Limit of Proportionality (LOP), residual)
244 Handbook of Precast Segmental Tunnel Lining Systems

German Committee for Structural Concrete within German Institute for Standardization
DAfStB Stahlfaserbeton:2012-​11 Richtlinie Stahlfaserbeton/​Technical Guidelines for Steel Fiber
Reinforced Concrete, Part 1-​3

German Concrete Association

DBV:2001 Guide to Good Practice –​Steel Fibre Concrete

German Federal Ministry of Transport

ZTV-​
ING:2007  Zusätzliche Technische Vertragsbedingungen und Richtlinien für den Bau
von Straßentunneln (ZTV-​ING)―Teil 5, Tunnelbau, Abschnitt 3 Maschinelle
Schildvortriebsverfahren

German Tunneling Committee

DAUB:2005 Recommendations for Static Analysis of Shield Tunnelling Machines
DAUB:2013  Lining Segment Design: Recommendations for the Design, Production, and
Installation of Segmental Rings

Italian National Research Council

CNR DT 204/​2006:2007 Guidelines for the Design, Construction and Production Control of Fibre
Reinforced Concrete Structures

Italian Tunnelling Society

SIG WG N. 2:2019 Report N. 1–​Damages of Segmental Lining

International Federation for Structural Concrete (Fédération internationale du béton, fib)

Model Code 2010 Model Code for Concrete Structures 2010
fib Working Party 1.4.1 2017 fib Bulletin 83. Precast Tunnel Segments in Fibre-​Reinforced
Concrete

International Tunneling Association

ITA WG 2 2000 Guidelines for the Design of Shield Tunnel Lining
ITA WG 2 2016 Twenty Years of FRC Tunnel Segments Practice: Lessons Learnt and Proposed
Design Principles
ITA WG 2 2019 Guidelines for the Design of Segmental Tunnel Linings
ITA WG 6 2004 Guidelines for Structural Fire Resistance for Road Tunnels
ITA WG 7 1988 Guidelines for the Design of Tunnels. ITA Working Group on General Approaches
to the Design of Tunnels

International Union of Laboratories & Experts in Construction Materials, Systems &

Structures
RILEM TC 162-​TDF:2003 Test and Design Methods for Steel Fibre Reinforced Concrete. σ–​ε Design
Method: Final Recommendation

Japan Society of Civil Engineers

JSCE 2007 Standard Specifications for Tunneling: Shield Tunnels

Japanese Railway Technical Research Institute

RTRI 2008 Design Standards for Railway Structures and Commentary (Shield Tunnels)

Los Angeles County Metropolitan Transportation Authority

LACMTA 2013 Metro Rail Design Criteria―–​Section 5 Structural and Geotechnical
 Analysis and design of precast segmental linings 245

Singapore Land Transport Authority

LTA 2010 Civil Design Criteria for Road and Rail Transit Systems

Spanish Ministry of Public Works and Transport

EHE-​08 2010 Code on Structural Concrete, ANNEX 14―Recommendations for Using Concrete
with Fibres

National Fire Protection Association (United States)

NFPA 130 2017 Standard for Fixed Guideway Transit and Passenger Rail Systems
NFPA 502 2017 Standard for Road Tunnels, Bridges, and Other Limited Access Highways

United States Army Corps of Engineers

USACE EM 1110-​2-​2901:1997 Tunnels and Shafts in Rock

AUTHORED DOCUMENTS
Angerer, W., and Chappell, M., 2008, “Design of Steel Fibre Reinforced Segmental Lining for the Gold
Coast Desalination Tunnels,” Proceedings of 13th Australian Tunneling Conference, pp. 4–​7.
Bakhshi, M., and Nasri, V., 2013a., “Latest Developments in Design of Segmental Tunnel Linings,”
Proceedings of the Canadian Society for Civil Engineering General Conference (CSCE 2013),
Montréal, QB, Canada.
Bakhshi, M., and Nasri, V., 2013b, “Practical Aspects of Segmental Tunnel Lining Design,”
Underground –​The Way to the Future” Proceedings of the World Tunnel Congress, 2013., G.
Anagnostou and H. Ehrbar, editors, Geneva, Switzerland.
Bakhshi, M., and Nasri, V., 2013c, “Structural Design of Segmental Tunnel Linings,” Proceedings
of 3rd International Conference on Computational Methods in Tunneling and Subsurface
Engineering: EURO: TUN 2013, Ruhr University, Bochum, Germany, pp. 131–​138.
Bakhshi, M., and Nasri, V., 2014a, “Review of International Practice on Critical Aspects of Segmental
Tunnel Lining Design,” Proceedings of the 2014 North American Tunneling (NAT) Conference,
Los Angeles, CA, pp. 274–​282.
Bakhshi, M., and Nasri, V., 2014b, “Guidelines and Methods on Segmental Tunnel Lining Analysis
and Design –​Review and Best Practice Recommendation,” Proceedings of the World Tunnel
Congress, Iguassu Falls, Brazil.
Bakhshi, M., and Nasri, V., 2014c, “Developments in Design for Fiber Reinforced Concrete Segmental
Tunnel Lining,” 2nd FRC International Workshop (1st ACI–​fib Joint Workshop) on Fibre
Reinforced Concrete, Montreal, QB, Canada, pp. 441–​452.
Bakhshi, M., and Nasri, V., 2014d, “Design Considerations for Precast Tunnel Segments According
to International Recommendations, Guidelines and Standards,” TAC 2014, Vancouver, BC,
Canada.
Bakhshi, M., and Nasri, V., 2015a “Design of Segmental Tunnel Linings for Serviceability Limit
State,” ITA World Tunnel Congress (WTC) 2015, Dubrovnik, Croatia, May 22 to 28, 2015.
Bakhshi, M., and Nasri, V., 2015b, “Cracking Serviceability Limit State for Precast Concrete Tunnel
Segments,” Canadian Society of Civil Engineers (CSCE) 2015 General Conference, Regina,
Canada.
Bakhshi, M., and Nasri, V., 2017a, “Design of Steel Fiber-​Reinforced Concrete Segmental Lining for
the South Hartford CSO Tunnel,” Rapid Excavation Tunneling Conference (RETC) 2017, San
Diego, CA, June 4–​7, 2017, pp. 706–​717.
Bakhshi, M., and Nasri, V., 2017b, “Reduction of Water Inflow by Controlling Cracks in Tunnel
Segmental Linings Using Fiber Reinforcement,” SP-​319, Reduction of Crack Width with Fiber,
American Concrete Institute, Farmington Hills, MI (CD-​ROM).
246 Handbook of Precast Segmental Tunnel Lining Systems

Bakhshi, M., and Nasri, V., 2018, “Guide for Optimized Design of Tunnel Segmental Ring Geometry,”
Proceedings of the 2018 North American Tunneling (NAT) Conference, Los Angeles, CA,
pp. 274–​282.
Bakhshi, M., and Nasri, V., 2020, “Crack Width Simulation and Non-​Linear Finite Element Analysis
of Bursting and Spalling Stresses in Precast FRC Tunnel Segments Under TBM Thrust Jack
Forces,” RILEM-​fib International Symposium on Fibre Reinforced Concrete. Springer, Cham,
Switzerland, 2020.
Bakhshi, M., Barsby, C., and Mobasher, B., 2014, “Comparative Evaluation of Early Age Toughness
Parameters in Fiber Reinforced Concrete,” Materials and Structures, V. 47, No.5, pp. 853–​872.
Barros, J. A. O., Cunha, V. M. C. F., Ribeiro, A. F., and Antunes, J. A. B., 2005, “Post-​Cracking
Behaviour of Steel Fibre Reinforced Concrete,” Materials and Structures, V. 38, No. 1,
pp. 47–​56.
Barwart, C., Romualdi, P., and Barioffi, A., 2013, “Headrace Tunnel of the El Alto Hydropower
Project in Panama,” Geomechanics and Tunnelling, V. 6, No.4, pp. 301–​311. https://​doi.org/​
10.1002/​geot.201300​019
Beňo, J., and Hilar, M., 2013, “Steel Fibre Reinforced Concrete for Tunnel Lining –​Verification by
Extensive Laboratory Testing and Numerical Modelling,” Acta Polytechnica, V. 53, No. 4,
pp. 329–​337.
Blom, C. B. M., 2002, “Design Philosophy of Concrete Linings for Tunnels in Soft Soil,” Ph.D. disser-
tation, Delft University Press, December.
Burgers, R., Walraven, J., Plizzari, G.A., and Tiberti, G., 2007, “Structural Behavior of SFRC
Tunnel Segments During TBM Operations,” World Tunneling Congress (WTC) 2007,
pp. 1461–​1467.
Caan, C.P., Jansen, J. A. G., Heijmans, R. W. M. G., and van der Put, J. L., 1998, “High Speed Line–​
South: The Groene Hart Tunnel –​Lining Design,” Reference Design, Report No. 9G4 0001
981028, 52 pages.
Caratelli, A., Meda, A., and Rinaldi, Z., 2012, “Design According to MC2010 of a Fibre-​Reinforced
Concrete Tunnel in Monte Lirio, Panama,” Structural Concrete, V. 13, No. 3, pp.166–​173.
Cardu, M., and Seccatore, J., 2016, “Quantifying the Difficulty of Tunnelling by Drilling and Blasting,”
Tunnelling and Underground Space Technology V. 60, No. 2016, pp. 178–​182.
Caverzan, A., Colombo, M., Di Prisco, M., and Rivolta, B., 2015, “High Performance Steel Fibre
Reinforced Concrete: Residual Behaviour at High Temperature,” Materials and Structures,
V. 48, No.10, pp. 3317–​3329.
Chaallal, O., and Shahawy, M., 2000, “Performance of Fiber-​Reinforced Polymer-​Wrapped Reinforced
Concrete Column Under Combined Axial-​Flexural Loading,” ACI Structural Journal, V. 97,
pp. 659–​668. https://​doi.org/​10.14359/​7433
Chiaia, B, Fantilli, A.P., Vallini, P., 2007, “Evaluation of Minimum Reinforcement Ratio in FRC
Members and Application to Tunnel Linings,” Materials and Structures, V. 40, pp. 593–​604.
https://​doi.org/​10.1617/​s11​527-​006-​9166-​0
Çimentepe, A. G., 2010, “Evaluation of Structural Analysis Methods Used for the Design of TBM
Segmental Linings,” MSc Thesis, Middle East Technical University, Ankara, Turkey.
Colombo, M., Martinelli, P., and di Prisco, M., 2015, “A Design Approach for Tunnels Exposed to
Blast and Fire,” Structural Concrete, V. 16, No. 2, pp. 262–​272.
Curtis, D. J., Hay, M., and Croydon, A., 1976, “Discussion on ‘The Circular Tunnel in Elastic
Ground,’ ” Géotechnique, V. 26, No. 1, pp. 231–​237.
de la Fuente, A., Pujadas, P., Blanco, A., and Aguado, A., 2012, “Experiences in Barcelona with
the Use of Fibres in Segmental Linings,” Tunnel and Underground Space Technology, V. 27,
pp. 60–​71. https://​doi.org/​10.1016/​j.tust.2011.07.001
de Oliveira e Sousa, J. L. A., Gettu, R., and de Oliveira e Sousa, 2006, “Determining the Tensile Stress-​
Crack Opening Curve of Concrete by Inverse Analysis,” Journal of Engineering Mechanics,
V. 132, No. 2, pp. 141–​148. https://​doi.org/​10.1061/​(ASCE)0733-​9399(2006)132:2(141)
Deere, D. U., Peck, R. B., Monsees, J. E., and Schmidt, B., 1969, “Design of Tunnel Liners and Support
Systems,” Highway Research Record 889, Final Report to Urban Mass Transit Administration,
U.S. Department of Transportation, Washington, DC.
 Analysis and design of precast segmental linings 247

Dinsmore C.I., 1982,“Column Analysis with a Programmable Calculator,” Concrete International,

Vol 4, pp. 32–​36.
Duddeck, H., and Erdmann, J., 1982, “Structural Design Models for Tunnels,” Tunneling ’82-​
Proceedings of the Third International Symposium. Institution of Mining and Metallurgy,
London, UK, pp. 83–​91.
EIA (2017), U.S. Energy Information Administration, available at: www.eia.gov/​tools/​faqs/​faq.
cfm?id=​74&t=​11 (accessed Jan 10, 2017).
Foster, S.J., and Attard, M.M., 2001, “Strength and Ductility of Fiber-​Reinforced High-​Strength
Concrete Columns,” Journal of Structural Engineering, V. 127, pp. 28–​34. https://​doi.org/​
10.1061/​(ASCE)0733-​9445(2001)127:1(28)
Francis, O., and Mangione, M., 2012, “Developments in Joint Design for Steel Fibre Reinforced
Concrete Segmental Tunnel Linings,” Proceedings of the World Tunnelling Congress 2012,
Bangkok, Thailand.
Fukui, K., and Okubo, S., 2003, “TBM Cutting Forces with Particular Reference to Cutter and Tunnel
Diameters,” ISRM 2003–​Technology Roadmap for Rock Mechanics, South African Institute of
Mining and Metallurgy, Johannesburg, South Africa.
Groeneweg, T., 2007, “Shield Driven Tunnels in Ultra High Strength Concrete: Reduction of the
Tunnel Lining Thickness,” M.Sc. Thesis, Delft University of Technology, The Netherlands.
Guglielmetti, V., Grasso, P., Mahtab, A., and Xu, S., 2007, Mechanized Tunnelling in Urban
Areas: Design Methodology and Construction Control, CRC Press, 528 pages.
Harding, A., and Francis, O., 2013, “Designing at the Limit: Brisbane Airport Link Segmental
Lining,” Rapid Excavation Tunneling Conference (RETC) 2013, Washington, DC, June 23–​26,
2013, pp. 834–​843.
Höeg, K., 1968, “Stresses against Underground Structural Clinders,” Journal of the Soil Mechanics
and Foundations Division, V. 94, No. 4, pp. 833–​858.
Hoek, E., and Brown, E. T., 2018, “The Hoek–​Brown Failure Criterion and GSI; 2018 Edition,”
Journal of Rock Mechanics and Geotechnical Engineering, https://​doi.org/​10.1016/​
j.jrmge.2018.08.001.
Iyengar, K. T., 1962, “Two-​Dimensional Theories of Anchorage Zone Stresses in Post-​Tensioned
Beams,” ACI Journal, Proceedings V. 59, No. 10, Oct., pp. 1443–​1466.
Janssen, P., 1983, “Load Capacity of Segment Joints,” Ph.D. Dissertation, Braunschweig University of
Technology, Braunschweig, Germany.
Kitsutaka, Y., 1997, “Fracture Parameters by Polylinear Tension-​ Softening Analysis,” Journal
of Engineering Mechanics, V. 123, No. 5, pp. 444–​ 450. https://​doi.org/​10.1061/​
(ASCE)0733-​9399(1997)123:5(444).
Lee, T., and Choi, T., 2017, “Numerical Analysis of Cross-​passage Opening for TBM Tunnels,”
Proceedings of the 19th International Conference on Soil Mechanics and Geotechnical
Engineering, Seoul, South Korea.
Maidl, B., Herrenknecht, M., Maidl, U., and Wehrmeyer, G., 2012, Mechanised Shield Tunnelling,
2nd Edition, Ernst & Sohn, Berlin.
Mashimo, H., Isago, N., Yoshinaga, S., Shiroma, H., and Baba, K., 2002, “Experimental Investigation
on Load-​Carrying Capacity of Concrete Tunnel Lining,” Proceedings of AITES-​ITA 2002–​28th
General Assembly and World Tunnelling Congress 2002, Sydney, Australia, pp. 1–​10.
Maxam, 2015. Riodin Technical Brochure, Maxam Europe S.A., Madrid. www.maxam.net.
Meda, A, and Rinaldi, Z, 2017, “Technical Report Tests on Precast Tunnel Segments in Steel Fiber
Reinforcement with GFRP Rebars,” Established by Prof Alebrto Meda and Prof Zila Rinaldi
from TERC–​ TUNNELLING ENGINEERING RESEARCH CENTRE, University of Rome
“Tor Vergata for ATP and Bekaert.”
Mendez Lorenzo, M. G., 1998, “Reliability Analysis of a Steel Fibre Reinforced Concrete Tunnel
Lining,” M.Sc. Thesis, Delft University of Technology, Delft, The Netherlands.
Merritt, J. L., Monsees, J. E., and Hendron, A. J., Jr., 1985. “Seismic Design of Underground
Structures,” Proceedings of the 1985 Rapid Excavation Tunneling Conference, June 16–​20,
1985. New York. Edited by C.D. Mann and M.N. Kelley. The Society for Mining, Metallurgy
& Exploration (SME), New York. Vol. 1, pp. 104–​131.
248 Handbook of Precast Segmental Tunnel Lining Systems

Miliutenko, S., Akerman, J., and Bjorklund, A., 2012, “Energy Use and Greenhouse Gas Emissions
during the Life Cycle Stages of a Road Tunnel –​The Swedish Case Norra Lanken,” European
Journal of Transport and Infrastructure Research, V. 12, No. 1, pp. 39–​62.
Mobasher, B., Bakhshi, M., and Barsby, C., 2014, “Back Calculation of Residual Tensile Strength of
Regular and High Performance Fiber Reinforced,” Construction & Building Materials, V. 70,
pp. 243–​253.
Moccichino, M., Romualdi, P., Perruzza, P., Meda, A., and Rinaldi, Z., 2010, “Experimental Tests
on Tunnel Precast Segmental Lining with Fiber Reinforced Concrete,” ITA 2010 World Tunnel
Congress, Vancouver, Canada.
Morgan, H. D., 1961, “A Contribution to the Analysis of Stress in A Circular Tunnel,” Geotechnique,
V. 11, pp. 37–​46.
Muir Wood, A. M., 1975, “The Circular Tunnel in Elastic Ground,” Géotechnique, V. 25, No. 1,
pp. 115–​127.
Neun, E., 2012, “Structural Fire Safety Engineering. Eurocode Approach and Practical Application,”
Proceedings of the World Tunnelling Congress 2012. Bangkok, Thailand.
Newmark, N. M., 1965, “Effects of Earthquakes on Dams and Embankments,” Geotechnique, V. 15,
pp. 139–​159.
Ninić, J., and Meschke, G., 2017, “Simulation Based Evaluation of Time-​Variant Loadings Acting
on Tunnel Linings During Mechanized Tunnel Construction,” Engineering Structures, V. 135,
pp. 21–​40.
Olesen, J. F., 2001, “Fictitious Crack Propagation in Fiber Reinforced Concrete Beams,”
Journal of Engineering Mechanics, V. 127, No. 3, pp. 272–​ 280. https://​doi.org/​10.1061/​
(asce)0733-​9399(2001)127:3(272).
Owen, G. N., and Scholl, R. E., 1981, “Earthquake Engineering of Large Underground Structures,”
Final Report, Report No. FHWA/​ RD-​80/​
195, Federal Highway Administration, U.S.
Department of Transportation, Washington, DC.
Peck, R. B., 1969, “Deep Excavation and Tunnelling in Soft Ground, State of the Art Report,”
7th International Conference on Soil Mechanics and Foundation Engineering, Mexico City,
Mexico, pp. 225–​290.
Piek, L., Ramsey, M., Oginski, S., and Hariharan, M., 2017, “Tunnel Station Joint Solution on the
Crenshaw LAX Corridor,” North American Tunneling Journal, Dec. 2017–​Jan 2018, pp. 12–​18.
Plizzari, G., 2009, Construction Methodologies and Structural Performance of Tunnel
Linings: Optimisation of the Structural, Technological and Functional Performance, of
Construction Methodologies and Materials, in Tunnel Linings, Starrylink Editrice, Brescia,
Italy.
Poh, J., Tan, K. H., Peterson, G. L., and Wen, D., 2009, “Structural Testing of Steel Fibre Reinforced
Concrete (SFRC) Tunnel Lining Segments in Singapore,” Proceedings of the World Tunnelling
Congress, Budapest, Hungary.
Rodriguez, J.A., and Aristizabal-​Ochoa, J.D., 1999, “Biaxial Interaction Diagrams for Short RC
Columns of Any Cross-​Section,” Journal of Structural Engineering, V. 125, pp. 672–​683.
https://​doi.org/​10.1061/​(ASCE)0733-​9445(1999)125:6(672)
Rodríguez, R., and Pérez, F., 2020, “Carbon Foot Print Evaluation in Tunneling Construction using
Conventional Methods,” Tunnelling and Underground Space Technology, 108, p. 103704.
10.1016/​j.tust.2020.103704.
Rodríguez, R., Tosal, A., P´erez, F., Díaz-​ Aguado, M.B., 2017. “Advancing Rate and Carbon
Footprint Related to Different Excavation Methods in Conventional Tunnelling” (Fabares tun-
nel, Asturias, Spain). International Conference on NexGen Technoclogies for Mining and Fuel
Industries 2017 (NxGnMiFu 2017). February 15–​17, 2017, New Delhi, India.
Rostami, J., 2008, “Hard Rock TBM Cutterhead Modeling for Design and Performance Prediction,”
Geomechanics and Tunneling, V. 1, No. 1, pp. 18–​28.
Salam, A., Greberg, B., Skawina, B., Gustafson, A., 2015, “Analysing Energy Consumption and Gas
Emissions of Loading Equipment in Underground Mining,” Canadian Institute of Mining,
Metallurgy and Petroleum (CIM), V. 6, No. 4, pp. 179–​188.
 Analysis and design of precast segmental linings 249

Schwartz, C. W., Azzouz, A. S., and Einstein, H. H., 1980, “Improved Design of Tunnel Supports,”
UMTA-​MA-​06-​0100-​8 0-​5. U.S. Department of Transportation, Washington, DC.
Shillaber, C., Mitchell, J., and Dove, J., 2016, “Energy and Carbon Assessment of Ground Improvement
Works. 2016 II: Working Model and Example,” Journal of Geotechnical and Geoenvironmental
Engineering, V. 142, No. 3, pp. 04015083-​1–​04015083-​9.
Sinha, R. S., 1989, Underground Structures: Design and Instrumentation, Elsevier Science, New York,
500 pages.
Soranakom, C., and Mobasher, B., 2007, “Closed-​Form Solutions for Flexural Response of Fiber
Reinforced Concrete Beams,” Journal of Engineering Mechanics, V. 133, No. 8, pp. 933–​941.
St. John, C. M., and Zahrah, T. F., 1987, “Aseismic Design of Underground Structures,” Tunnelling
and Underground Space Technology, V. 2, No. 2, pp. 165–​197.
Steven, G., and Empelmann, M., 2014, “UHPFRC-​ Columns with High-​ Strength Longitudinal
Reinforcement,” Beton–​und Stahlbetonbau 109, 344–​354. https://​doi.org/​10.1002/​best.201300​090
Talmon, A. M., and Bezuijen, A., 2013, “Analytical Model for the Beam Action of a Tunnel Lining
During Construction,” International Journal for Numerical and Analytical Methods in
Geomechanics, V. 37, No. 2, pp. 181–​200.
Tiberti, G., 2009, “Concrete Tunnel Segments with Combined Traditional and Fiber Reinforcement:
Optimization of the Structural Behaviour and Design Aspects,” Ph.D. Dissertation, University
of Brescia.
Tiberti, G., 2014, “Concrete Tunnel Segments with Combined Traditional and Fiber Reinforcement:
Optimization of the Structural Behavior and Design Aspects,” Ph.D. Thesis, Department of
Civil, Environmental, Architectural Engineering and Mathematics, University of Brescia, June
2014, Aracne editrice s.r.l., Rome, Italy, ISBN: 978-​88-​548-​7005-​5, 396 p.
Tiberti, G., Plizzari, G. A., Walraven, J. C., and Blom, C. B. M., 2008, “Concrete Tunnel Segments with
Combined Traditional and Fiber Reinforcement,” Proceedings of the fib 2008 Symposium: Tailor
Made Concrete Structures, Amsterdam, the Netherlands, pp. 199–​205.
Tiberti, G., Conforti, A., and Plizzari, G.A., 2015, “Precast Segments Under TBM Hydraulic
Jacks: Experimental Investigation on the Local Splitting Behavior,” Tunnelling and Underground
Space Technology, V. 50, pp. 438–​450. https://​doi.org/​10.1016/​j.tust.2015.08.013
Vandewalle, L., 2000, “Cracking Behaviour of Concrete Beams Reinforced with a Combination of
Ordinary Reinforcement and Steel Fibers,” Materials and Structures, V. 33, No. 3, pp. 164–​170.
Vasdravellis G, Uy B, Tan EL, and Kirkland B., 2012, “Behaviour and Design of Composite Beams
Subjected to Negative Bending and Compression,” Journal of Constructional Steel Research,
V. 79, pp. 34–​47. https://​doi.org/​10.1016/​j.jcsr.2012.07.012
Vodička, J., Krátký, J., Hilar, M., and Ráček, V., 2012, “Structural SFRC for Precast Segments of
the Tunnel Lining,” 8th Central European Congress on Concrete Engineering, Durability of
Concrete Structures, Plitvice Lakes, Croatia.
Wang, J.-​ N., 1993, Seismic Design of Tunnels –​ A Simple State-​ of-​
the-​Art Design Approach,
Monograph No. 7, Parsons Brinckerhoff, New York.
Wells, D., and Coppersmith, K., 1994, “Updated Empirical Relationships among Magnitude, Rupture
Length, Rupture Area, and Surface Displacement,” Bulletin of the Seismological Society of
America, V. 84, pp. 974–​1002.
Yao, Y., Bakhshi, M., Nasri, V., and Mobasher, B., 2018a, “Interaction Diagrams for Design of Hybrid
Fiber-​Reinforced Tunnel Segments,” Materials and Structures, V. 51, No. 1, 17 pages.
Yao, Y., Bakhshi, M., Nasri, V., and Mobasher, B., 2018b, “Closed-​Form Solutions for Interaction
Diagrams of Hybrid Fiber-​Reinforced Tunnel Segments,” Proceedings of FRC2018–​3rd ACI-​
fib-​
RILEM on Fibre Reinforced Concrete: From Design to Structural Applications, Lake
Garda, Italy.
Zhang, J., and Stang, H., 1998, “Applications of Stress Crack Width Relationship in Predicting the
Flexural Behavior of Fibre Reinforced Concrete,” Cement and Concrete Research, V. 28, No. 3,
pp. 439–​452. https://​doi.org/​10.1016/​S0008-​8846(97)00275-​5.
Zhong, X.-​C., Liu, Q.-​W., and Zhao, H., 2011, “Study on the Grouting Pressure of Shield Tunnel,”
ASCE Geotechnical Special Publication, V. 215, pp. 183–​190.
250 Handbook of Precast Segmental Tunnel Lining Systems

NOTATION
A =​ effective tension area of concrete around rebar divided by number of steel bars,
in.2 (mm2)
Ad =​ load distribution area inside segment under thrust jack forces, in.2 (mm2)
Ag =​ gross area of concrete section, in.2 (mm2)
Aj =​ area of contact zone between jack shoes and the segment face, in.2 (mm2)
As =​ area of reinforcing bars, in.2 (mm2)
a =​ distance from edge of vacuum lift pad to edge of segment in the load case of
stripping (demolding), or dimension of final spreading surface under thrust jack
forces, in. (mm)
al =​ transverse length of contact zone between jack shoes and the segment face,
in. (mm)
at =​ transverse length of stress distribution zone at the centerline of segment under
thrust jack forces, in. (mm)
b =​ width of tunnel segment, or width of tested specimen, ft (m)
Cc =​ compression force in the concrete section, lbf (N)
Ct =​ tensile force in the section due to fiber reinforcement, lbf (N)
De =​ external diameters of the tunnel segmental lining, ft (m)
Di =​ internal diameter of the tunnel segmental lining, ft (m)
d =​ thickness of tested specimen, or total width of the segment cross section, in. (mm)
d1 =​ length of load transfer zone for the case of longitudinal joint bursting load,
in. (mm)
dburst =​ centroidal distance of bursting force from the face of section, in. (mm)
dk =​ width of the hinge joint or thickness of contact surface between segment joints
for the case of longitudinal joint bursting load, in. (mm)
dc =​ concrete cover over rebar, in. (mm)
ds =​ distributed width of stress block inside the segment for the case of longitudinal
joint bursting, in. (mm)
E =​ modulus of elasticity of concrete, psi (MPa)
Er =​ modulus of elasticity of surrounding ground, psi (MPa)
Es =​ stiffness modulus of the surrounding ground determined by oedometer test, psi
(MPa); or modulus of elasticity of rebar, psi (MPa)
EH =​ horizontal earth pressure, psi (MPa)
EV =​ vertical earth pressure, psi (MPa)
e =​ eccentricity, in. (mm)
eanc =​ eccentricity of jack pads with respect to the centroid of cross section, or max-
imum total eccentricity in longitudinal joints consisting of force eccentricity and
eccentricity of load transfer area, in. (mm)
F =​ forces acting on bottom segment due to self-​weight of segments positioned above
when segments are piled up within one stack during storage or transportation
phases, lbf (N)
Fsd =​ bursting tensile forces developed close to longitudinal joints, lbf (N)
Fsd,r =​ spalling tensile forces developed close to longitudinal joints, lbf (N)
Fsd,2 =​ secondary tensile forces developed close to longitudinal joints, lbf (N)
fy =​ yield stress of required reinforcing bars, psi (MPa)
f1 =​ first peak flexural strength, psi (MPa)
fbot =​ stress at the extreme bottom fiber of concrete section, psi (MPa)
f′c =​ specified compressive strength of concrete segment, psi (MPa)
 Analysis and design of precast segmental linings 251

fcd =​ concrete design compressive strength according to BS EN 1992-​ 1-​

1:2004,
psi (MPa)
fctd =​ fiber-​reinforced concrete design tensile strength, psi (MPa)
fct,eff =​ concrete tensile strength, psi (MPa)
f′co =​ compressive strength of partially loaded concrete surface, psi (MPa)
fD150 =​ residual flexural strength at net deflection of L/​150, psi (MPa)
fD600 =​ residual flexural strength at net deflection of L/​600, psi (MPa)
f′D150 =​ specified residual flexural strength at net deflection of L/​150, psi (MPa)
f′D600 =​ specified residual flexural strength at net deflection of L/​600, psi (MPa)
fD150r =​ required average residual flexural strength at net deflection of L/​ 150,
psi (MPa)
fD600r =​ required average residual flexural strength at net deflection of L/​ 600,
psi (MPa)
fFtu =​ fiber-​reinforced concrete tensile strength at ultimate limit state, psi (MPa)
fR3 =​ residual flexural strength of fiber-​reinforced concrete beam corresponding to
crack mouth opening displacement of 0.1 in. (2.5 mm), psi (MPa)
fR1 =​ residual flexural strength of FRC beam corresponding to crack mouth opening
displacement of 0.02 in. (0.5 mm), psi (MPa)
fs =​ stress in rebar, psi (MPa)
ft =​ specified splitting tensile strength, psi (MPa)
g =​ self-​weight of the segments per unit length, lbf/​in. (N/​mm)
H =​ overburden depth, ft (m)
Hw =​ groundwater depth, ft (m)
h =​ thickness of tunnel segment, in. (mm)
hanc =​ length of contact zone between jack shoes and the segment face, in. (mm)
I =​ moment of inertia of FRC segment, in.4 (mm4)
J =​ Tunnel Boring Machine (TBM) thrust jack forces, kips (kN)
k =​ coefficient of subgrade reaction or subgrade reaction modulus, lb/​ft3 (kg/​m3)
kjr =​ Janssen rotational spring stiffness in longitudinal joints, lb.in/​rad (N.mm/​rad)
kr =​ radial component of subgrade reaction modulus or stiffness of radial springs
simulating ground-​structure interaction, lb/​ft3 (kg/​m3)
kt =​ subgrade reaction modulus in the tangential direction, lb/​ft3 (kg/​m3); or in crack
width analysis, a factor depending on the duration of loading (0.6 for short and
0.4 for long term loading)
kθ =​ tangential component of subgrade reaction modulus or stiffness of tangential
springs simulating ground-​structure interaction, lb/​ft3 (kg/​m3)
L =​ distance between the supports, in. (mm)
lt =​ full length of contact area between segments in longitudinal joints, in. (mm)
Mdistortion =​ bending moment due to additional distortion effect, lbf.ft (N.m)
Mn =​ nominal resistance bending moment, lbf.ft (N.m)
N =​ axial hoop force in segments, lbf (N)
n =​ number of segments per ring excluding the key segment (n≥4); or number of
layers of tensile rebar in crack with analysis
NEd =​ maximum normal force due to permanent ground, groundwater and surcharge
loads, lbf (N)
P0 =​ surcharge load, lbf (N)
Pe1 =​ vertical earth pressure at crown of lining applied to the elastic equation method,
psi (MPa)
252 Handbook of Precast Segmental Tunnel Lining Systems

Pe2 =​ vertical earth pressure at invert of lining applied to the elastic equation
method, psi (MPa)
Pg =​ segment dead load, psi (MPa)
Pgr =​ Radial grouting pressure, psi (MPa)
Pw1 =​ vertical water pressure at crown of lining applied to the elastic equation
method, psi (MPa)
Pw2 =​ vertical water pressure at invert of lining applied to the elastic equation
method, psi (MPa)
Ppu =​ factored jacking force applied on each jack pad in circumferential joints, or
maximum factored normal force from the final service loads transferred in
longitudinal joints, psi (MPa)
qe1 =​ horizontal earth pressure at crown of lining applied to the elastic equation
method, psi (MPa)
qe2 =​ horizontal earth pressure at invert of lining applied to the elastic equation
method, psi (MPa)
qw1 =​ horizontal water pressure at crown of lining applied to the elastic equation
method, psi (MPa)
qw2 =​ horizontal water pressure at invert of lining applied to the elastic equation
method, psi (MPa)
R =​ radius from centerline of lining, ft (m)
ro =​ radius of excavated tunnel, ft (m)
s =​ maximum rebar spacing, in. (mm)
S =​ distance between stack supports and free edge of segments in the load case of
segment storage, ft (m)
ss =​ sample standard deviations of test results
sr,max =​ maximum crack spacing, mm
Tburst =​ bursting force, lbf (N)
WAp =​ groundwater pressure, psi (MPa)
w =​ segment self-​weight, plf (kg/​m); or maximum crack width, in. (mm)
yc =​ distance from extreme compression fiber to centroid of equivalent compres-
sion force in the section, in. (mm)
y =​ distance from extreme tension fiber to the neutral axis, in. (mm)
β =​ dimension of the loaded surface under thrust jack forces according to Iyengar
diagram, in. (mm); or in crack width analysis ratio of the distance between
neutral axis and tension face to the distance between neutral axis and cen-
troid of rebar
ΔPg, invert =​ vertical gradient of radial grout pressure between the crown and invert of
tunnel, psi (MPa)
δ =​ displacement of lining applied to the elastic equation method, in. (mm)
δd =​ diametrical distortion, in. (mm)
εtu =​ ultimate compressive strain
εcu =​ ultimate tensile strain
ε’csd =​ compressive strain due to shrinkage and creep equal to 150×10-​6
ϕ =​ strength reduction factor; or reinforcing bar diameter, in. (mm)
γ =​ material safety factor
λ =​ slenderness defined as the ratio between the developed segment lengths and
its thickness
 Analysis and design of precast segmental linings 253

θ =​ angle from crown in the elastic equation method, or rotation in the longitu-
dinal Janssen joint, radians
ρconcrete =​ specific weight of concrete, lb/​ft3 (kg/​m3)
ρeq =​ equivalent specific weight of grout, lb/​ft3 (kg/​m3)
σc,j =​ compressive stresses developed under jack pads because of axial effects of
thrust jack forces, psi (MPa)
σcm =​ fully spread compressive stress in method of the Iyengar diagram, psi (MPa)
σcx =​ bursting tensile stresses using the Iyengar diagram, psi (MPa)
σp =​ specified post-​crack residual tensile strength of FRC segment, psi (MPa)
τyield =​ shear yield strength of grout, psi (MPa)
Chapter 4

Precast concrete technologies

Lead authors: Barzin Mobasher, Jim Lindsay, and Louis Falco
Contributing authors: Stephen Farrington, Steve Schaef, and
Okan Duyar

4.1 INTRODUCTION
Portland cement industry is among the primary producers of greenhouse gases generating
between 5 to 8% of worldwide man-​made point-​source CO2 emissions. The nature of the
release of greenhouse gases from cement plants has led to pervasive discussions on the
carbon footprint of concrete materials [1]. Substitution with alternative construction mate­
rials is however a daunting task since cement and concrete contain the lowest embodied
energy of all man-​made industrial and construction materials [2].
The statistics are staggering. Every year an estimated weight of 25 billion tons of con-
crete is manufactured globally, represented by about 3.8 tons per person. Concrete is used
around the world more than the total of all other building materials, including wood,
steel and plastics. According to United States Environmental Protection Agency (US EPA)
estimates, at the end-​of-​use stage, the Construction and Demolition (C & D) debris con-
stituted 600 million tons in the United States in 2018, twice the amount of generated
MSW [3].
The United States cement industry developed a voluntary goal of reducing CO2 emissions
by 10% (based on 1990 levels) per ton of cementitious products by implementing quality
monitoring and control to make cement production more sustainable. We still face a multi-​
faceted set of challenges that include better design of structural materials by taking into
account the climate change, cement carbon footprint, the volume of materials used, envir-
onmental risk factors, durability and the need to incorporate serviceability, performance-​
based specifications, economy and hybrid design approaches. It is expected that we focus
on developing methods to better utilize, design and construct concrete structures while pur-
suing sustainable guidelines that include new materials and design approaches.
In recent years, a significant number of tunneling projects worth more than $100 bil-
lion are planned by municipalities in every state in the nation with more than $40 billion
in active bidding [4]. Given such infrastructure investments, the precast concrete industry
needs a more robust and systemic approach for design, analysis, testing, quality control,
tolerance control and qualifications of the tunnel lining segments that are among the most
highly engineered components of the civil works projects. Furthermore, with extremely high
tolerances and substantial costs of repair due to failure or deterioration, we must address
new construction or maintenance of aging infrastructure using innovative approaches devel-
oped recently in the concrete industry.
The use of precast segments by tunnel boring machines (TBMs) in underground con-
struction is becoming the dominant and cost-​effective approach for traffic, rail, metro,
water and wastewater tunnels systems [5]. The justifications for the use of precast concrete

DOI: 10.1201/9781032675541-4
 Precast concrete technologies 255

segments are based on the speed of construction, elimination of temporary support, serv-
ing as the reaction frame for boring, the combination of initial and final support, choice
support for TBM bored tunnels in soft ground and weak fractured rock [6]. This chapter
addresses the criteria and fundamental importance of the parameters that affect the
strength, ductility and durability of the precast concrete materials used in tunnel segments
using the TBM-​type machines. Fundamental contributions of the mix components, design
parameters, early age and long-​term properties, as well as the durability of the material,
are addressed.

4.2 DESIGN OF CONCRETE MATERIALS FOR SERVICEABILITY AND

SUSTAINABILITY REQUIREMENTS
In addition to highway and rail tunnels, areas of application have been extended to
the tunnels and pipes aimed at water and wastewater collection. This segment of the
nation's aging infrastructure has some of the most prevalent defects such as cracks and
broken pipe systems. In 2002, the US EPA Office of Water published a report titled
“Clean Water and Drinking Water Infrastructure Gap Analysis” (EPA-​816-​R-​02-​020)
[7]. The gap analysis report identified a critical shortfall in funding of the nation’s
water and wastewater infrastructure including a $270 billion gap for wastewater infra-
structure for the years 2000–​2019. The deferred approach to the operation and main-
tenance of the aging wastewater infrastructure is a paramount concern of the Agency
[8]. According to ASCE the operational efficiency of the water management system is
approximately at the 50% level. The distress due to the durability in these areas data
indicates that a rational approach for serviceability, durability and life cycle cost of
concrete tunnels is needed.
The problems facing the environmental structures present unique cases in comparison to
other applications. In general, the conventional structural design concentrates mainly on
strength and long-​term durability in terms of chlorides and frost resistance. In tunnel lining
and water treatment concrete structures, however, serviceability in terms of limited deflec-
tion and cracking, and low permeability assume equal importance. The use of technological
advances in the area of concrete technology will have a positive impact on the implementa-
tion of sustainable development and environmental safety of both the water and transpor-
tation systems.
A serviceability approach towards the design of reinforced concrete structures is a major
area of need that can optimally address the use of experimental tests and computational
modeling. Since the main parameters of concern are strength, ductility, aesthetics, dur-
ability and permeability, the goals are therefore multi-​faceted. Multiple structural design
objectives can be addressed using a hybrid reinforcement option using a mixture of trad-
itional continuous reinforcement (steel or FRP) and short fibers in a matrix that is designed
using the physical and chemical fundamentals of high-​ performance cement materials.
These objectives start with the first-​tier requirements in terms of economy, strength and
durability, however, they are broken down into tangible design parameters in terms of
second-​tier parameters. These include parameters such as ductility, crack growth resist-
ance, fracture toughness, crack width, stiffness, water permeability, chloride permeability,
corrosion resistance and other parameters that directly deal with the internal mix design
parameters.
ACI 350 provides recommendations for the structural material for the wastewater treat-
ment facilities [9]. Concrete is nearly impermeable in its uncracked state and extremely
256 Handbook of Precast Segmental Tunnel Lining Systems

resistant to corrosion, hence the stringent requirements of these structures require con-
crete quality and control of crack widths. The main design requirements of wastewater
treatment facilities are water tightness and durability. Tunnel facilities must be designed
to minimize fluids leakage. Accordingly, the design process should address the probability
distribution of maximum crack width under service loads while preventing the emergence
of wide cracks and other potential sources of leakage, such as joints. Shrinkage and thermal
dimensional changes must also be considered in the definition of joints. Criteria need to
be defined for the movement along slip lines and permission zones including barriers to
avoid the passage of fluids. Design analysis must be as accurate as possible, allowing the
perception of the structure’s true stress distribution and preventing cracking at the max-
imum stress zones.
In environmental applications, such as tunnel lining, one can anticipate advantages from
the use of FRC that include drastic reductions in reinforcement and a mix formulation
that permits self-​consolidation and improved segregation resistance. By increasing the cost-​
effectiveness and labor efficiency in construction, improved durability and minimized need
for maintenance and repair operations are expected.
The designer starts from the applied forces due to loading conditions, obtains the solu-
tion of equilibrium of forces using structural analysis or finite element models, and proceeds
to compute the limit states, failure mechanisms and deformations to obtain the structural
demand. After the choices of material properties are made, these will be interfaced with
optimization tools of analysis and design for a large subset of design parameters. Model
validation can be carried out through material testing, parameter estimation, experimental
full-​scale tests, failure mode analysis and load-​displacement experiments on full sections
that include TBM precast sections and other structural concerns, such as the durability that
apply to environmental structures.
The stages of loading can be determined in terms of production, handling, transporta-
tion and installation before the service life or ultimate state is considered. In these stages,
ductility and crack tolerance are important aspects of serviceability requirements. Due to
the many advantages offered to the tunnel sections, fiber reinforced concrete has received
much attention compared to conventional reinforced concrete (RC). In addition to the
economy and speed of construction, these include their ductility, durability, crack control,
as well as higher impact, fatigue, blast and fire resistance. In addition, fiber reinforcement
provides overall materials toughness and protections for localized cracking in the edges
and corners, which could easily occur during handling or installation. Special loading
cases are also considered in terms of dynamic loading, blast, impact and fire [10].
The primary mode of serviceability level loading of the circular tunnel linings is the com-
bined loading of normal forces and relatively low bending moments. Due to the low bend-
ing capacity required, oftentimes, the use of SFRC to satisfy the minimum reinforcement is
sufficient. SFRC segment applications have thus flourished as an integral part of the design
domain. SFRC also helps in areas with no continuous reinforcement to protect the edges
and corners of the elements. The benefits of FRC in areas such as fire and blast events in
the lining design have received limited attention. The energy density released in a confined
volume of a tunnel due to an explosion is much higher compared to a blast in open space.
Explosions are therefore the extreme consequence of fire accidents [11]. Design aspects such
as stiffness of the section after cracking, its ductility, and residual tensile capacity are dom-
inant design parameters. This indicates that the flexural strength or compressive strength
cannot be used as the primary acceptance metric and the full response of the sections under
flexural or tensile loading should be characterized.
 Precast concrete technologies 257

Key concerns in terms of design for serviceability in terms of permeability, crack width
and stiffness can be addressed by the use of fibers. Therefore, this has further promoted the
application of FRC due to its cost competitiveness at the structural level compared with
other traditional solutions which require a higher reinforcement ratio.

4.3 CONCRETE CONSTITUENTS AND CEMENT HYDRATION ASPECTS

Cement hydration starts as soon as the water is added to the mix and the hydration prod-
ucts result in the binding of the aggregates and other components of concrete. The hydra-
tion of cement is the most essential component of the process of concrete gaining strength
and durability. Water absorbs into the outer part of the cement grain, dissolving calcium
and hydroxide ions, which move out into the surrounding water. This leaves the cement
surface with a calcium-​depleted hydrosilicate layer carrying a negative charge. Some of the
positive calcium ions in the solution then adsorb back onto this silicate surface to give a
positive charge as measured by the Zeta potential. Water continues to diffuse into the grain,
releasing further calcium and hydroxide ions into the solution and increasing the thickness
of the hydrosilicate layer [12, 13].
The predominant reaction between the cement and the water mainly involves the tri-​and
di-​calcium silicates (C3S, C2S) phases of the cement. However, other phases are also present,
including tricalcium aluminate (C3A), which is very reactive although present at only about
10% of the cement weight. Sulfates, in the form of gypsum, are added to the cement to help
control the C3A reactivity by forming ettringite. Ettringite has a reactive surface that can
attract and absorb a disproportionate amount of some admixtures [14, 15]. The durability
and strength of concrete can be enhanced by the reaction of calcium silicates and compro-
mised by the early-​stage or delayed uncontrollable reactions of calcium aluminates. These
reactions that involve the primary ingredients affect the overall performance of various
blended mixtures and justify an understanding of the roles of various components in a
hydrating cementitious system discussed in the following sections.

4.3.1 Portland cement
Portland cement is the main binding agent in concrete that sets and hardens by the chemical
reactions of its constituents with water, hence categorized as hydraulic cement. Portland
cement clinker is made by calcination of raw natural minerals containing oxides of cal-
cium, silica, aluminum and iron elements. These are expressed in terms of the oxides (CaO,
SiO2, Al2O3, Fe2O3) and partial amounts of impurities. These oxides are intermixed and go
through the process of calcination in a kiln to produce the Portland cement clinker. At this
stage, the clinker is a hydraulic material with two primary components of tri-​calcium and
di-​calcium silicates (3CaO ⋅ SiO2 and 2CaO ⋅ SiO2), as that contributes to the hydration pro-
cess. The remaining compounds of calcium, aluminum and iron-​containing phases as well
as other compounds can be viewed as impurities.
Portland cement is produced by grinding the clinker to a powder with a high surface area
and then blending it with gypsum (calcium sulfate) to help control the reaction of the alu-
minate phases. Cement types are classified differently throughout the world. In the United
States, most Portland cements for commercial construction meet the requirements of ASTM
C150 [16]. ASTM C150 is a prescriptive specification with limits on both chemical and
physical characteristics. According to ASTM C150, cement types are classified based on the
relative proportions of the oxides present and labeled as type I through type V with those
meeting multiple categories such as types I/​II, and those including secondary components
258 Handbook of Precast Segmental Tunnel Lining Systems

as blended such as air-​entraining component with an A, such as type IIA. Blended cements
with pozzolans or slag are available and referred to with a suffix such as type IIP, IIS.
Portland-​cements in Canada are specified under a performance specification of CSA
A3001 [17]. Cement types are labeled as type GU, type MS, type MH, type HE, type LH, or
type HS, with an L added after the two-​letter code, such as GUL, designating that a higher
amount of limestone has been added.
In Europe, types of cement are classified in EN 197 by the composition of oxides into
many categories of different common types of cement, sulfate-​ resisting-​
cements, blast
furnace cement, and their constituents. Portland types of cement are labeled CEM I and
blended types of cement labeled CEM II, through CEM V [18]. Sulfate-​resisting Portland
types of cement are labeled with an SR designation. In addition, the cement is labeled with
a strength class of 32.5, 42.5, or 52.5, which is the minimum 28-​day strength of mortar, in
MPa, from standard testing. An example of a blended sulfate-​resisting cement from EN 197
is CEM III/​B-​SR 32.5. Desired concrete performance and local availability of specific classes
of cement will guide the selection of Portland cement for a project. New Belite cements are
also gaining popularity [19].

4.3.2 Supplementary cementitious materials

Supplementary cementitious materials (SCM) are an important category of products
addressing the sustainability of concrete [1]. Supplementary cementitious materials are a
category of inorganic powders with similar and compatible chemical and mineral charac-
teristics to conventional Portland cement and are referred to as mineral admixtures. They
are added to the Portland cement in a concrete mix or as a blended component. SCMs are
normally added to concrete on a weight basis and contribute to a percentage of the total
cementitious system based on the overall performance requirements. SCMs differ signifi-
cantly in their method and sources of production, physical and chemical properties, nature
of the reaction products, and the short and long-​term effects. However, they are identified
by the pozzolanic and/​or cementitious properties that they demonstrate. Figure 4.1 presents
the relative positions of Portland cement, fly ash, slag cement, silica fume, and metakaolin
on a ternary diagram (CaO–​SiO2–​Al2O3) [20]
An SCM may serve several functions in the hydration process such as reacting with the
products of Portland cement hydration via the pozzolanic reaction, provide a dense packing
of the particles from a physical aspect, or provide hydration nucleation sites for chemical
reactions that take place. The packing of the particles of small size in specific locations such
as the interface between the aggregates and cement paste affects the density and diffusivity
in the microstructure as well as changes to the chemical compositions and hydration reac-
tions. Certain SCMs may also be partially cementitious, reacting directly with water. This
partial replacement of Portland cement with pozzolanic materials results in improved mech-
anical and durability properties compared to ordinary concrete formulations [21]. A large
array of SCMs with different chemical compositions are categorized based on their origin
and include Class F and C fly ash, silica fume, slag cement, metakaolin, calcined clay or
shale, and natural pozzolans. Depending on the type and amount, SCMs serve several func-
tions by improving the fresh and hardened concrete in terms of the following:

(a) Rheology and workability by reducing the bleeding and segregation tendency
(b) Durability enhancements by reducing the permeability, and pore size, while increasing
the matrix and the interfacial transition zone packing and hydration products
 Precast concrete technologies 259

Figure 4.1 Ternary diagram of Portland cement and supplementary cementing materials.

Note: Ternary CaO-​S iO 2–​A l 2O 3 diagram situating the chemical constitution of the portland
cement, slag cement, fly ash, metkaolin and silica fume.
Source: [20]. 

(c) Resistance to ionic mobility including chlorides, sulfates, and alkalis responsible for
alkali-​aggregate expansion. Reduction of heat generation, shrinkage potential, plastic
shrinkage cracking, as well as resistance to thermal cracking,
(d) Increase early age and ultimate strength

The history of using blended cements to improve the strength and durability of concrete
goes back decades. The primary feature is the pozzolanic reactivity, which is generally iden-
tified with the reaction of SCM with portlandite to create additional calcium silicate hydrate
(C-​S-​H) [22]. Primarily the initial applications revolved around the effectiveness of blended
cements in reducing the potential sulfate attack damage [23]. Despite the wealth of studies
conducted, the role of the chemical composition of pozzolanic materials is affected by the
source due to the large degree of variation among the origin, type and manufacturing pro-
cess of pozzolans. The demand for blended cements and applications of pozzolanic materials
for ASR and sulfate attack mitigation has enhanced the scientific nature of characterization
tests. While the supplies are diminishing in certain categories, new sources of SCMs will
need to be identified and optimized for use in concrete construction [24].
The SCM materials used are typically specified in terms of their specific ASTM designa-
tions that include type I/​II ordinary Portland cement (OPC) conforming to ASTM C150,
class F, and C fly ash and metakaolin, or natural pozzolans conforming to ASTM C618, slag
conforming to ASTM C989, limestone powder conforming to ASTM C568, and microsilica
(silica fume) conforming to ASTM C1240. While these materials are significantly different
260 Handbook of Precast Segmental Tunnel Lining Systems

Figure 4.2 Particle size distribution curves for materials.

Note: Particle size distribution curves for Portland cement, fly ash, slag, metakaolin, and lime-
stone powders. The median size in microns is shown in parentheses.
Source: [26]. 

in terms of their chemical and physical properties, their particle size can be used as a basis
for comparison. The particle size distribution (PSD) curves of some of these materials are
shown in Figure 4.2. The chemical compositions and physical characteristics of these mate­
rials are available in the report [25].
The following sections describe some of the more common SCMs with the primary cat-
egories that include fly ash, silica fume, slag and metakaolin, which provide significant eco-
nomic and environmental benefits.

4.3.3 Silica fume (microsilica)

Silica fume is a very fine amorphous silica that is a by-​product of elemental silicon or silicon
alloy manufacturing (Malhotra et al. 1987). Due to the nature of the processing, the size
of a silica fume particle is an order of magnitude smaller than that of OPC, which results
in an extremely high surface area and the densification of the cement paste. The surface
area also provides an abundant number of nucleation sites for cement hydration, thereby
increasing the rate of reaction during the early stages of hydration. Silica fume is usually
added in a concrete mix at an addition rate of 5 to 15% by mass of cement and requires
an increased dosage of high-​range water-​reducing admixtures to ensure dispersion and
improve workability.
 Precast concrete technologies 261

Due to its fine particle size and challenges in handling and mixing, silica fume is used
under different forms of microsilica powder, condensed silica fume or slurry form. Research
has shown that silica fume particles tend to agglomerate resulting in extremely low work-
ability and therefore it is important to develop mixing methods to reduce the agglomera-
tions for a workable concrete mixture. The condensed silica fume is used quite commonly
due to ease of handling, and airborne particle safety.
Silica fume is pozzolanic and enhances the mechanical properties of the matrix due to
pore size and interface transition zone refinement as a result of the formation of secondary
hydration products. The reactivity of silica fume is attributed to the nature of highly reactive
silica and its small particle size, which results in faster property development compared to
other SCMs. The high reactivity is best shown by comparing the strength activity indices of
silica fume with fly ash in cementitious systems [27]. While the small particle size directly
increases the rate of reaction, a thorough and uniform dispersion of such fine particles is
challenging, thus a high dosage of superplasticizer is needed. The particle agglomeration
may reduce the expected benefits of mechanical properties or cause ASR-​type reactions due
to the agglomeration of condensed grains.

4.3.4 Fly ash
Fly ash is the finely divided residue that is airborne and collected in a bag-​house as a coal-​
burning by-​product of a steam-​generating power plant. Pulverized fuel ash is a less common
term for fly ash. Fly ash is primarily glassy (amorphous) spherical particles that react via the
pozzolanic reaction only, or have both pozzolanic and cementitious properties. Its chem-
ical nature is directly affected by the properties of the coal burned at the power plant. The
spherical shape of the particles improves the rheology, workability and packing density of
the mixture, thus improving the permeability [Figures 4.3(a) and (b)].
As the burning of pulverized coal generates heat, inorganic molten mineral residues
harden into bottom ash. Cooling of the flue gas causes the airborne particles to harden
as fly ash. Emission control equipment use a combination of electrostatic separators and/​
or filter fabric bag-​houses that remove the fly ash [28]. The economic and environmental
benefits of fly ash in concrete is enhanced by improving mechanical properties. Fly ash use

Figure 4.3 Scanning electron microscope image and x-​r ay diffraction patterns.

Note: Scanning electron microscope image and x-​r ay diffraction pattern for typical class F fly ash.
Source: [35]. 
262 Handbook of Precast Segmental Tunnel Lining Systems

reduces greenhouse gases, decreases life cycle costs, and increases the concrete’s durability.
Production of fly ash during the 1990s was about 100 million tons per year at a utilization
rate of 20% in applications such as blended cements, and road base stabilization. The avail-
ability however is reduced due to the switching of the power plants from coal to natural gas.
Depending on the source of coal, fly ash may be siliceous or calcareous, and while both
types have pozzolanic properties, higher calcium levels indicate potentials for hydraulic
properties. Siliceous fly ash consists of reactive silicon dioxide (SiO2) and aluminum oxide
(Al2O3) in the form of a spherical fine powder. The remainder contains iron oxide (Fe2O3)
with a low proportion of reactive calcium oxide (CaO) that is less than 10% by mass and
other compounds. In the United States, fly ash is classified primarily by the calcium content
into high-​calcium (class C) or low-​calcium (class F). Its chemical composition is per ASTM
C618. Due to the relatively high calcium content of up to about 20%, class C fly ashes typ-
ically have some cementitious nature. In Canada, class C fly ashes are further divided into
class CI, for intermediate calcium content, and class CH, for high calcium content [29]. Fly
ash is often used at a level of 15 to 30% of the total cementitious content of a concrete mix.
Class C ashes from sub-​bituminous coals contain calcium alumino-​sulfate glass, such as
quartz (Si O2), free lime (CaO), and high-​calcium content that is a key factor in mitigating
sulfate attack and ASR.
The reaction of calcium silicate phases (C3S and C2S) in the Portland cement generates
calcium silicate hydrate CSH gel and the extra calcium is released in the form of the calcium
hydroxide. As a pozzolanic material, fly ash reacts with the released calcium hydroxide to
form its CSH hydration product [30]. Due to the need for high pH necessary and sufficient
calcium ions present, this sequential reaction takes longer to accomplish than the hydration
of Portland cement. The additional CSH production ultimately improves concrete proper-
ties. Due to the extended time required for the hydration of fly ash, the rate of the heat of
hydration reduces. The net increase of the hydration products (C-​S-​H gel), improves the
mechanical properties and durability of concrete. The synergy of the use of fly ash with
other SCMs such as silica fume and metakaolin accelerates property development at early
ages [31]. The disadvantages include the slower rate of reaction of fly ash (in general, class
F), which may affect the setting times, as well as lower early-​age strength. The variability
of fly ash sources introduces uncertainty in the setting time, which affects the operation
schedule.
Class F fly ash replacement levels range from 10 to 30% by mass of cement, while class C
fly ash can be used in higher proportions. Early promotion of the use of fly ash can be
attributed to the significant research conducted during the 1980s and 1990s to address
concrete durability Aïtcin [32]. The utilization of mineral by-​products was correlated to
the worldwide growth of infrastructure development. Neville and Aitcin [33] reported on
high-​performance concrete containing fly ash, ground granulated blast furnace slag, and
silica fume where the dosage of fly ash in high volume of fly ash (HVFA) concrete can
exceed 50% [34]. Many studies addressed the physical, chemical (including trace elements),
mineralogical, and LOI (loss on ignition test) properties of fly ashes. The LOI tests deter-
mine the unburnt carbon content of fly ash as it affects the influence of carbon content on
air-​entraining admixture demand. The durability of concrete under various chemical and
physical environments has been studied in detail and with satisfactory resistance. Sulfate
resistance may be improved depending on the replacement level.
The chemistry of class C ashes is quite variable and less efficient at mitigating ASR com-
pared to class F. ASTM C1567 testing is required for comparison of different grades of
class C fly ash from class F so that a minimum dosage level is determined. The design for
 Precast concrete technologies 263

ASR resistance requires a reduction in the Portland cement to limit the alkali loading to
≤ 2.1 kg/​m3 (3.5 lb/​CY). ASTM C1567 testing is used to determine if reduced dosages,
other local class C ashes, or tertiary blends are acceptable. Figure 4.3(a) shows a scanning
electron microscope’s (SEM) micrograph of the small and characteristically round fly ash
particles. The X-​ray diffraction pattern of class F fly ash shows the general amorphous
nature of fly ash containing quartz and mullite as the few crystalline phases are shown in
Figure 4.3(b) showing present.
To address the risks from improper disposal and discharge of coal ash, new environmental
guidelines for coal ash disposal and additional controls on power plant water regulations
discharges were placed. Coal utilization in the energy sector has significantly reduced, lead-
ing to the decline in coal combustion products, therefore new fly ash sources are becoming
scarce. With the conversion of power generation plants to natural gas in recent years, the
supplies of conventional SCMs such as produced coal combustion fly ash or blast furnace
slags are diminishing [36]. Beneficiation of reclaimed and low-​quality fly ashes may also
provide large volumes of SCMs.

4.3.5 Slag cement
Slag cement, also called ground granulated blast furnace slag (GGBFS), is a hydraulic and
pozzolanic material that is produced by rapid quenching, granulating, and grinding of the
molten slag that is a by-​product of steel production in a blast furnace. Although some use
the terms slag and slag cement interchangeably, it should be noted that slag is the material
that is the by-​product of steel production, and it is only by rapid quenching that it becomes
reactive in cementitious systems.
Air-​cooled slag is non-​reactive and may be used as aggregate. Slag cement fume is usually
added in a concrete mix at a level of 30 to 50% of the total cementitious content of a con-
crete mix, but higher levels can be used. Slag cement may be used in concrete to improve
compressive strength, reduce permeability, improve sulfate resistance, improve resistance to
alkali-​silica reaction (ASR), and reduce thermal stress in mass concrete [37]. ASTM C989
specification addresses different grades of ground granulated blast-​furnace slag [38] that is
used for blending with Portland cement per specification of ASTM C595 or as a separate
ingredient in concrete, grout and mortars.

4.3.6 Metakaolin
Metakaolin is a calcined clay that is produced by low-​temperature calcination of kaolin clay
of high purity. It is a white powder that is often used as a substitute for silica fume when
light-​colored concrete is required. It is added in a concrete mix at a rate of 5 to 20% by
mass of cement to improve the durability of the concrete by reducing the permeability and
increasing the compressive strength. Concrete with metakaolin will almost always require
the use of a high-​range water-​reducing admixture to improve its workability and ensure the
dispersion of the powder.
Metakaolin (Al2Si2O7) has been commercially available since the mid-​1990s as a high-​
reactivity pozzolan for HPC applications. It is a thermally activated alumino-​silicate material
manufactured by calcination of kaolin clay in a temperature range of 1,300 to 1,560°F (700
to 850°C), which contains 50 to 55% SiO2 and 40 to 45% Al2O3 [39].
Metakaolin improves workability and constructability while improving the mechanical
properties. Brooks and co-​workers [40] reported a 30% reduction of slump when using
264 Handbook of Precast Segmental Tunnel Lining Systems

15% metakaolin in concrete followed by a 20% retardation of setting times. The mech-
anical and durability properties show as much as 20% increase in compressive strength,
a decrease of water penetration by 50%, gas permeability of 40% for 15% replacement
of the Portland cement by metakaolin. Both the water absorption potential decreased by
30%; while the ASR expansion decreased by 80% [41]. The use of metakaolin significantly
improved the durability performance and sulfate resistance of concrete.
Metakaolin is highly reactive with calcium hydroxide, resulting in a rapid strength gain
and accelerated cement hydration. Both silica and alumina in metakaolin increase the C/​S
ratio of the produced C-​S-​H gel compared to the C-​S-​H generated due to the hydration of
silica fume and fly ash. The CH consumption in the mixture was dependent on the level of
metakaolin replacement according to Wild and Khatib [42]. CH content after a year of cur­
ing reduced by 50% in pastes with 15% metakaolin indicating an increased C-​S-​H gel pro-
duction. The pore diameter of the paste after 28 days decreased by 66% (0.0122 microns)
for 20% metakaolin mixtures.

4.3.7 Limestone
The use of limestone powder is becoming an attractive cement replacement material based
on its ability to reduce the carbon footprint of the blended cement system through bypassing
the flow of limestone through the kiln and either co-​grinding it with the clinker or blending
it with the Portland cement after separate grinding. Portland limestone cements are rou-
tinely used in the United States. According to ASTM C595, a type IL cement may include
up to 15% limestone powder as an SCM [43].
Once fine limestone particles are specified and used in the micron range, the particle
packing, as well as enhancements in packing density are obtained. The small size of parti-
cles also serves as nucleation sites for the initiation of hydration reactions with aluminate-​
bearing materials. The chemical reaction of limestone powder with the aluminate phases
results in cement forming a carboaluminate phase. Research has shown that fine limestone,
in the presence of metakaolin or fly ash, enhances the amount of C-​S-​H gel formed as
early as 28 days, and has a beneficial impact on strength development. This has provided
the rationale for using fine limestone in ultra high performance concrete (UHPC) mixtures
in combination with other SCMs, [35, 44], which in turn improves the sustainability of
concrete.

4.4 AGGREGATES
The aggregate type and its source have a considerable influence on the compressive strength
of concrete since they occupy 60 to 75% of the concrete volume. As a general rule, an
uncrushed coarse aggregate (generally smooth and rounded) makes a concrete with a lower
strength than one with crushed coarse aggregate. Tunnel segment concretes are generally
very low in slump and designed to be produced in short production cycles considering
high early strength requirements, fine details on the surface for gaskets and other handling
details, and complex geometry of the segments.
Aggregates must conform to certain standards for project or owner-​specific properties
as well as national standards such as American Society for Testing Materials (ASTM) or
Canadian Standards Association (CSA). In general, clean and strong aggregates are expected
to be durable, free from adsorbed chemicals, and fine materials such as clay that may inter-
fere with cementitious materials hydration and bonding to aggregate surfaces. Geometrical
 Precast concrete technologies 265

Table 4.1 Range of proportions for normal weight steel fiber reinforced concrete

Size aggregate
Item Concrete ix parameter 3/​8 ” maximum 3/​4 ” maximum 1.5” maximum
1 Cement (lbs/​yd 3) 600 to 1,000 500 to 900 470 to 700
2 Water cement ratio (w/​c) 0.35 to 0.45 0.35 to 0.50 0.35 to 0.55
3 Percent of fine or coarse 45 to 60 45 to 55 40 to 55
aggregate
4 Entrained air content (%) 4 to 8 4 to 6 4 to 5
5 Fiber content (volume %)
5a Deformed fiber 0.4 to 1.0 0.4 to 0.8 0.2 to 0.7
5b Smooth fiber 0.8 to 2.0 0.6 to 1.6 0.4 to 1.4
Source: [46].

factors fine aggregates such as the fineness, maximum size, grading distribution, and particle
shape and surface texture, affect the compressive strength as well. Aggregate characteristics
indirectly influence the compressive strength through the water demand. Table 4.1 presents
the range of proportions for normal weight steel fiber reinforced concrete depending on
maximum aggregate size.

4.4.1 Coarse aggregates
A wide range of aggregate sources are used in the production of tunnel sections and opti-
mization of aggregate characteristics improves fresh and long-​term concrete properties
through optimal packing and reduced need for water, and chemical admixtures. Key char-
acteristics include but are not limited to shape, angularity and texture; gradation (including
maximum aggregate size), and fine particles. These characteristics may be considered based
on empirical, rheology-​based models and or amount of vibration to be used in production
to predict fresh concrete behavior and properties. The coarse aggregates are generally a
combination of gravels, or crushed stone particles larger than 5 mm (0.20 inch) or No. 4
sieve to the optimal size of 9 to 25 mm (0.375 to 1.0 inch) depending on cage openings,
cover, and 13 mm (0.50 inch) of the fiber length. Generally, tunnel segment concretes have
some combination of higher total aggregate content, the greater amount of fine aggregate
relative to coarse aggregate, and smaller maximum aggregate size ACI 533.5R) [45]
As a rule, aggregates including sand need to be washed and graded at the pit or plant.
Use of recycled aggregate is not recommended considering the complexity of the sourcing
and related durability concerns unless otherwise tested and proved to be applicable for the
project.

4.4.2 Fine aggregate
Fine aggregates are natural sand or produced from crushing aggregates smaller than 5 mm
(0.20 inches) or No. 4 sieve. Fine aggregates produced in crushing aggregates may not
need to be washed if the process is built to minimize the impurities in crushed sand. The
266 Handbook of Precast Segmental Tunnel Lining Systems

size and shape of fine aggregate particles affect the workability such that very fine sand
increases water demand for similar workability than using coarser sand. The interlocking
of angular fine aggregates affects the rheology and mobility of the aggregates during mixing
and consolidation.

4.4.3 Aggregate packing approaches

The primary goal of the mix design is developing optimal mixtures with low water to binder
ratio, reasonable cementitious material content to reduce the shrinkage cracking potential,
and a high microstructure packing density to reduce permeability and increase the strength.
The concept of aggregate packing becomes critical in this approach. It is known that the
heat of hydration and shrinkage potential of the overall cement content can be well regu-
lated by the use of SCMs. Understanding the design methods based on optimal packing
of aggregates is essential to achieve high efficiency since aggregate packing ensures proper
interlock, and a dense interparticle network for transmission of load, whereas powder pack-
ing, ensures a reduction of capillary porosity and inter-​particle voids.
Proper grading of aggregates is an important step of concrete design that has shited from
empirical approaches to optimization techniques based on effective and tangible scientific
metrics. However, as long as aggregates are handled in terms of coarse and fine terminolo-
gies with a wide range of possible size distributions, their combination can not address the
fundamental challenge of optimal packing. The combination of aggregates aims to pro-
vide optimum distribution of each range of aggregates from coarse to sand, however in
reality the barrier between the two categories is not defined. Intermediate aggregates 4 to
8 mm (0.16 to 0.31 inches) range are important to maintain the optimal range, and lack
of intermediate aggregates may increase shrinkage, or overuse of intermediate aggregates
may cause poor workability, increase in water demand, and poor finishability. Empirical
methods like power grading, fuller curve or particle packing methods can be used to develop
optimal gradation [47].
The combination of gradation and characteristics of aggregates indirectly affects the com-
pressive strength by influencing the water needed to meet the workability criteria, reducing
the cementitious materials for a given strength, and yet reducing possible fiber agglomer-
ation (if used in production). For the segment concretes over 50 MPa (7,250 psi) compres-
sive strength, it is generally necessary to obtain aggregates from select sources with higher
mechanical strengths and consistent quality. The correlation of the fiber length and the max-
imum aggregate size is important to provide uniform and effective fiber distribution. The
maximum aggregate size shall not be longer than 0.5 times the fiber length. Aggregate size
and proportions play an important role in determining the flexural properties and results.
The range of the size distribution of different supplementary cementitious products and
aggregates is about two orders of magnitude, requiring a unique approach to the blend-
ing optimization tasks. Several models are available at different levels of complexity [48].
Model predictions of the packing density of aggregates [49] produce a dense UHPC with
a binder content of about 650 kg/​m3 (1,095 lbs/​CY) The suspension viscosity model by
Mooney was modified by Stovall and de Larrard [50] to derive the linear model of solids
in a suspension with infinite viscosity. Using the virtual packing density of the particles a
random packing efficiency is calculated [51]. Using a maximization of packing density, each
particle is placed individually in a representative volume. While it is possible to compact the
particles to achieve a higher volume fraction, theoretical values are only obtained by indi-
vidual placement of the particles.
 Precast concrete technologies 267

Figure 4.4 The packing density of a poly-​d ispersed aggregate mixture.

Note: (a) Densely packed mixture of similar grains; (b) densely packed mixture of coarse and
fine grains showing loosening effect; (c) a densely packed mixture of fine grains disturbed in the
presence of a coarse grain illustrating wall effect; (d) binary aggregate mixture.

The compaction of real aggregates is challenging due to their irregular shape and the
random number of different sizes present. The compressible packing model (CPM) is widely
used for the prediction of the different size aggregate packing densities. The rheology of
fresh cementitious mixtures is affected by the solid packing fraction.

4.4.4 Compressible packing models

The packing density of mixtures with two or more aggregates sizes is calculated using CPM.
The packing density of a poly-​dispersed aggregate mixture can be expressed as a function
of the individual packing densities of each aggregate size present in the mixture as shown
in Figure 4.4(a). For an aggregate blend with several different sizes, the packing of any size
is affected by the presence of all other aggregate sizes and proportions. This introduces two
additional concepts of wall effect and loosening effect.
The loosening effect refers to the decrease in the packing density of a coarse grain blend
by introducing fine grains. For single-​sized coarse aggregate volume, the addition of rela-
tively smaller sizes disturbs the original particle packing shown in Figure 4.4(b). The wall
effect is due to introducing a coarse grain into a mixture of fine grains and is the opposite
of the loosening effect as shown in Figures 4.4(c) and (d). It lowers the packing density of
the mixture. The dominant aggregate size is the one size within an aggregate mixture with
the maximum volume fraction and depends on the aggregate quantity, shape and size in
addition to the individual packing density of each phase. The virtual packing density γ i is
the theoretical value for an aggregate mixture when an individual aggregate class ‘i’ is dom-
inant. These interactions define the parameters used for the loosening and wall effects [52].
By combining theoretical and experimental results CPM predictions of the packing
density show a high degree of accuracy, making it a preferred mix design approach. The
packing density obtained for a blend of aggregate size ranges can be calculated using the dry
rodded unit weight (DRUW) test. Additional information about the use of particle packing
models can be found in a recent report [53].

4.5 WATER
Water used in the production of tunnel segment concretes is expected to meet ASTM C1602
requirements. The use of recycled water is not recommended since impurities in batch water
268 Handbook of Precast Segmental Tunnel Lining Systems

may influence the set time, speed of hydration and impact the initiation and total duration
of curing time, and production cycles.

4.6 CHEMICAL ADMIXTURES
Admixtures are chemicals that are introduced into the fresh concrete or at the time of
mixing to modify in the fresh state or post-​hardening properties. These areas include
affecting the water demand through plasticizing or water-​reducing, changing the rate of
setting through accelerating/​retarding action, changing the yield strength and plastic vis-
cosity to control segregation, cohesion or the bleeding of the mix. These admixtures can
be a single or a blend of several chemicals mostly in aqueous solutions so that they can be
easy to measure, dispense and disperse accurately through the fresh concrete. The active
chemical in liquid admixtures is typically 35 to 40% but can reach as high as 100% in the
case of shrinkage-​reducing admixtures and as low as 2% for the synthetic air-​entraining
admixtures. Since the majority of admixtures are in the typical range of 0.3 to 1.5%,
a correction in water, cement or total weight is not required. The dosage is normally
expressed as liters or kg per 100 kg (220 lbs) of total supplementary cementitious prod-
ucts. The nature of chemical admixtures is through the various mechanisms that include
the following:

• The rate of reaction of the cementitious systems may be accelerated or retarded

through a chemical intervention in the basic cement hydration process (of one or
more of the cement phases). This can be accomplished by providing nucleation sites
to increase the rate of reaction, or prevention of nucleation to delay the precipitation
of hydration products. Finally, admixtures may change the hydration characteristics
by changing the pH of the basic chemistry.
• Plasticizing or superplasticizers act through homogenizing the distribution of the
cement particles through particle dispersion. The dispersion is due to the adsorption
of the admixtures onto the cement surfaces, results in the formation of a uniform sur-
face charge and thus particle repulsion and break up of agglomerations.
• Changes to the surface tension of the water would increase air entrainment.
• Changes to the mix rheology would result in an increased plastic viscosity or
cohesion.
• Special chemicals that affect unique properties such as steel rebars’ corrosion suscep-
tibility or repel water.

Admixtures are essential components of fresh concrete and benefit rheology, compac-
tion and finishing characteristics. The benefits to the hardened properties are in terms
of strength, stiffness, bond to the substrate, and permeability, which ultimately lead to
enhanced durability.
With the use of admixtures, there is usually a primary property for which the admixture
is being used that is also associated with secondary effects that are associated with the use.
For example, a water-​reducing admixture may be used to address durability concerns but
if the concrete mix lacks cohesion and bleeds, the use of a water-​reducing admixture that
increases cohesion would be more appropriate. The primary and secondary functions are
therefore a water reducer that improves cohesion. The secondary effects are often the prime
reason why an admixture works well with a specific source of cement and aggregates and is
completely ineffective for another cement type and aggregate source [12].
 Precast concrete technologies 269

4.6.1 A ir entrainment admixtures

The standard practice in improving the resistance of concrete to frost is air entrainment.
The entrained fine air bubbles in a fresh concrete state preserve their stability in the pres-
ence of water. Their distribution ensures that the distance between neighboring air spaces
is sufficiently small that pressures created by the freezing of water in the pores are reduced.
It is specified by the total percent volume of air in the mix, however other metrics such as
spacing factor or specific surface of the air are also used. It is usual to aim at an air content
of 5 to 6% by volume of concrete while the distance between individual air bubbles should
not exceed about 0.25 mm (0.01 inches). The mechanism of operation is through synthetic
surfactants that are polar molecules and affect the water’s surface tension using a charged
hydrophilic end and a hydrophobic tail. The nature of hydrophilicity causes air bubbles to
be trapped. Air entrainment protects against scaling and the deterioration of concrete due
to internal volume changes such as ASR and sulfate attack, thus increasing its durability
and service life.
Most tunnels are located well below the freeze–​thaw (heave) line and do not require
chemically entrained air unless required by the engineer and project specifications. When
the tunnel alignment crosses the freeze–​thaw (heave) line, the air entrainment value becomes
a critical component in the mix design. Air entrainment improves the properties of the fresh
state by increased cohesion, improved workability, and reduced bleeding and segregation.
In the hardened concrete, in addition to the resistance to freeze-​thaw action, improvements
in long-​term durability are associated with internal volume changes. The direct correlation
between the use of air entrainment with a loss of concrete strength results in typical strength
losses of about 5 to 6% per percentage of air entrainment, an issue that has been known for
more than 65 years [54]. Therefore, air content should be specified at the lowest possible
level to meet the durability objectives.
For the same total void space, large air voids are not nearly as effective as small, closely
spaced voids. Chemically entrained air bubbles are slightly smaller and spherical as com-
pared to entrapped air bubbles, which are usually a result of the folding and shearing actions
provided by the concrete mixer used when batching/​mixing the concrete ingredients. The
size difference between the entrained and entrapped air bubbles is “arbitrarily” set [at] a
boundary at 1.0 mm (0.04 inch). Industry-​standard requirements for normal air-​entrained
ranges vary from 4.5% to 8%, whereas the entrapped air ranges can vary from 1% to 4%
dependent on the concrete mixing equipment utilized.
Air entrainment is affected by several parameters including the mix constituents, mix-
ing, temperature, transport, and pumping distance, vertical drop, and pressure. In between
batch consistency can be difficult to maintain if these parameters are not well controlled.
For example, the variability of some SCMs such as fly ash due to the presence of unburnt
carbon affects the ability to tune the admixture dosage to meet the air content required. This
is due to the high pore system in the carbon particles, which is a justification for limiting the
loss on ignition (LOI) on the specifications that are directly related to the unburnt carbon
in the fly ash source.
Air entrainment affects both the plastic and the hardened concrete by the uniform distri-
bution of air bubbles. The mechanism of operation as defined in EN 934-​2, is due to the
changes they cause to the surface tension of water. The combination of a molecular hydro-
philic end and a hydrophobic tail, traps air bubbles and is sufficiently strong to maintain its
form during the mixing process. Traditionally, they were formulated using fatty acid salts or
vinsol-​resin but currently, blends of synthetic surfactants function better in terms of stability
and void characteristics.
270 Handbook of Precast Segmental Tunnel Lining Systems

Air entrainment is usually specified by the total volume of air voids, their size, specific
surface area or spacing. Typical dosage is relatively a small amount of the order of 0.2%
by cement weight and will result in significant changes, therefore, accurate and calibrated
dosing equipment is needed to obtain a consistent level of air. Since the strength of concrete
is affected by entrained air by as much as 5 to 6% reduction for each 1% of additional air,
air content should be limited to the lowest level necessary to achieve the required proper-
ties. However, in low cement content and harsh mixes, air entrainment may improve the
rheology and workability, allowing water reduction that may compensate for the strength
loss. The sensitivity of air entrainment to the mix design components, mixing, temperature,
transport and pumping from batch to batch diminishes the opportunity to specify a pre-
scriptive dosage for all situations.

4.6.2 Water-​reducing admixtures
Water-​reducing admixtures (WRA) in the concrete mix serve multiple advantages when
producing a precast segmental tunnel liner. They assist in increasing the workability of the
concrete mix, finish surface characteristics, increased strength (compressive, tensile, flex-
ural), allow for the reduction in water-​cementitious ratio, increased concrete durability,
and allows lesser quality cements to be used, thus promoting a lower carbon footprint and
subsequently can reduce overall costs.
Water reducing agents are a low viscosity liquid that mainly reduces the water content in
the mix design. The water reducers’ primary function is to disperse as many of the cemen-
titious particles as possible allowing more particles to be coated and hydrated by water
thus resulting in increased concrete strength [55]. There are two types of watering reducing
liquids presently in use in the industry, type A and type F.
Type A water reducing agent is considered mid-​range water-​reducing admixture (MRWR)
superplasticizers allowing water content reductions up to 20% dependent on superplasti-
cizer dosage rates applied. Dosage rates vary between 60 to 180 ml/​45 kg (2 to 6 fl oz per
100 lbs) of cementitious powder. Whereas, type F is high-​range water reducing admixture
(HRWR) from normal-​range to high-​range superplasticizers allowing water content reduc-
tions up to 35% to 45% dependent on the dosage rates applied. Dosage rates vary between
150 to 240 ml/​45 kg (5 to 8 fl oz per 100 lbs) of cementitious powder. The dosage rates for
the water-​reducing admixtures are dependent on the manufacturer’s dosage recommenda-
tion and testing during the trial mix design period as compared with the available products.
A low water cementitious ratio is most advantageous in defining concrete durability and
water-​reducing admixtures play a role in determining its success.

4.6.3 Retarding admixtures
A retarding admixture’s main function is to reduce/​delay the process of cement hydration
within the concrete mix, thus increasing the setting time of concrete. The primary concern
is to avoid rapid hardening, which can cause shrinkage cracks and reduce the concrete’s
overall durability. This is often subject to the manufacturing plant’s location or time when
hot-​weather concreting occurs. Other considerations are the manufacturing plant’s opera-
tions requiring sufficient time to access aggregates/​materials, batch, transport and cast con-
crete into precast segmental tunnel liners molds.
High compressive strength requirements with high cementitious powder may also require
retarding admixtures to ensure good-​quality products. When a retarding admixture is
 Precast concrete technologies 271

utilized, the proper dosage would be affected by the water-​cementitious ratio and the tri-​
calcium silicate (C3A) content.

4.6.4 Accelerating admixtures
There are primarily two reasons to consider the use of a set-​accelerating admixture in a
concrete mix design, the primary reason is to reduce the concrete setting time to allow the
beginning of accelerated curing and subsequently obtaining a higher early strength for the
product to be removed from its mold. The second consideration would be due to a predom-
inately cold-​weather environment that requires an accelerator in the concrete mix.

4.6.5 Specialty admixtures
As a minimum, the coarse and fine aggregates are required to be tested in conjunction with
the cementitious material for alkali-​silica reactivity according to ASTM C1567 standard. If
the results are not acceptable based on the standards, inhibiting admixtures can be included
in the concrete mix design. Other specialty admixtures include corrosion-​inhibiting and
shrinkage-​reducing admixtures.

4.6.6 Mixtures proportioning
Concrete mix proportions are made in consideration of workability, strength and durability.
The selection of mixture proportions requires an optimization of the initial, service level and
ultimate state design requirements for handling, transportation, installation and service life.
These topics are addressed under the criteria that include workability, appearance, finish-
ability, strength, durability and final surface features based on the economy. Other fresh
properties include pumpability, bleeding, density and heat generation.
The mix design procedures range from the local experience of the precast pro-
ducers mix plant operators using empirical guidelines and arbitrary proportions of
cement:sand:gravel:water up to much more sophisticated methods. The general approach
for selecting proportions for concrete made with hydraulic cement meeting ASTM C150/​
C150M, C595/​C595M, or C1157/​C1157M, SCM, normal or high-​density aggregates, and
chemical admixtures, with workability suitable for normal cast-​in-​place construction. The
next-​level mix designs are based on meeting minimum criteria for accomplishing the stated
minimum requirements of the project. Most commercially produced concrete today con-
tains a combination of chemical admixtures, and SCM, incorporated in proportioning the
mixture. A broad range of characteristics ranging from high strength to self-​consolidation
and flowable, from low permeability characteristics and applications are made possible with
the use of mineral and chemical admixtures and SCMs. The next level is extended to the first
principles such as packing model [56].
There are a wealth of resources available to troubleshoot the problems encountered in
the production cycle. These resources fall back on the fundamental aspects of the mecha-
nisms discussed in the previous sections. Variations in the strength for a given w/​cm may
be due to changes in placement or curing conditions, gradation, maximum size, texture,
shape, strength and stiffness of aggregates. Differences in cement types or sources; air con-
tent; and the use of chemical admixtures that affect the cement hydration process or that
develop cementitious properties themselves are also contributing factors. Water-​reducing
admixtures are used to reduce the mixing water requirements by (ACI 212.3R). Slump char-
acteristics for developing special concretes such as self-​consolidating concrete (ACI 237R),
272 Handbook of Precast Segmental Tunnel Lining Systems

or other applications are needed for close control of workability (ACI 29 238.1R). In add-
ition, admixtures such as set-​retarders, accelerators, and shrinkage-​reducers are used based
on various performance targets. (ACI 212.3R).
A low w/​cm will improve the transport properties by reducing the penetration of aggres-
sive ionic species as discussed. Resistance to freezing and thawing damage and deicing salts
are greatly improved by the incorporation of entrained air system (ACI 201.2R). The dur-
ability of concrete may be affected by exposure to seawater or sulfate-​containing soils,
which can be addressed by using sulfate-​resisting or slag cement, as well as other SCMs. The
aggregate-​alkali reaction (AAR) potential is mitigated by ASTM C1778.

4.7 FIBER REINFORCED CONCRETE

Plain concrete has low tensile strength and strain capacity at failure. As an alternative to
conventional reinforcement, fibers have been used as reinforcement. Concrete materials
produced with short, randomly distributed fibers may be superior to forms of reinforced
concrete using welded wire mesh or rebars. Both the tensile strength and toughness, espe-
cially the post-​crack strength are improved [57].
It has been shown that due to the reduced specific spacing, fibers strengthen the composite
at the micro-​level by bridging the microcracks before they reach the critical flaw size [58].
The small diameter of the individual fibers ensures a better and more uniform distribution
of reinforcement. In addition, the high surface area offers significant bond capability. Since
the bond strength of glass, steel or even polymeric fibers is far superior to reinforcing bars,
this increases the efficiency of reinforcement so that there is limited crack opening due to the
debonding and pull-​out of reinforcement. The fibers are distributed randomly, offering effi-
ciency in load transfer by the fiber phase. Finally, because the fibers that bridge the matrix
cracks are resilient and highly compliant, they can orient to carry the load across the crack
faces. This factor is expected to increase the durability of concrete substantially since the
crack width control affects long-​term durability.
The tensile capacity may be also supplied by short fibers since there is a history of
more than 40 years of working with fiber reinforcement [59]. Different fibers including
macro synthetic, steel or glass have their unique features. Steel fibers provide high levels
of strength and toughening, however, due to their bond strength, distribution and fiber
spacing, their effectiveness is observed after the matrix has fully cracked. Macro syn-
thetic fibers are alkali resistant and have high corrosion resistance, however, their stiff-
ness is low, and their fire performance has not been fully documented under long-​term
sustained creep effects. Recent developments have allowed for utilizing the strength and
stiffness of the fibers that bridge the tensile cracks for durability, serviceability and min-
imum reinforcement.
The characteristics of concrete mix design with fibers include the specifications, ingredi-
ents, testing, analysis and design of plain and hybrid use of fiber-​reinforced concrete (FRC)
for structural applications that deal with precast tunnel lining panels. Recent documents
published by ACI Committee 544 provide a wealth of information for the designers [60].
The guidelines of structural concrete design for Environmental Structures proposed by ACI
350 are intended to address serviceability limit states (short-​and long-​term deformations,
cracking, corrosion resistance, etc.) through the ultimate limit states. FRC can offer an eco-
nomical solution to high-​performance tunneling and environmental applications.
Several codes and guidelines allow for the use of fiber-​reinforced concrete (FRC) as a
structural material specially geared towards the use in tunneling applications MC 2010
 Precast concrete technologies 273

[62]. There are many field experiences with tunnels constructed with TBM in which FRC
is used as the main reinforcing component of the section. Several studies in the published
literature address the experiments, production methods, quality control testing, materials
property evaluation, and numerical modeling of precast concrete tunnel lining segments
[63, 64, 65, 66, 67].

4.8 CHARACTERIZATION OF DUCTILITY IN FIBER-​R EINFORCED

CONCRETE
Depending on the type, composition, geometry, length-​to-​diameter ratio, and dosages rate,
steel fibers added various levels of ductility to the segments. The distribution of the fiber
is related to the rheology of the mix as well as the maximum aggregate size, which affects
uniform and effective fiber distribution. Maximum fiber length must be per minimum
dimensions (thickness) of the segment and amount of reinforcement to ensure a uniform
distribution of the fibers in segments [68].
Discrete macro fibers added to concrete provides three-​dimensional reinforcement result-
ing in superior crack-​control, thus increased flexural capacity, fatigue and impact resistance.
The type, geometry, physical property and dosage of the macro fiber selected for reinforce-
ment have a significant influence on the flexural residual strength, ductility and crack con-
trol characteristics. ASTM C1116 is the standard specification for fiber-​reinforced concrete
(FRC) and classifies FRC by the material type of the fiber incorporated into the concrete
mixture. The correlation of the fiber length and the maximum aggregate size is important to
provide uniform and effective fiber distribution. The maximum aggregate size shall not be
longer than 0.5 times the fiber length. Maximum fiber length must be following minimum
dimensions (thickness) of the segment and amount of reinforcement this would provide a
complete and uniform distribution of the fibers in segments. ACI 544.5R-​10 provides a sig-
nificant set of discussions on the corrosion of steel fibers.
Proper addressing of the specifications, quality control testing, and design guides for
structural applications of concrete materials for tunnel lining applications. The intent is to
address fundamental aspects of materials selection and develop guidelines for the design
of mixtures. The testing for flexure is normally conducted following ASTM C1609-​12
Standard Test Method for Flexural Performance of Fiber-​Reinforced Concrete (Using Beam
with Third-​Point Loading) [69], which requires a closed-​loop testing system as shown in
Figure 4.5.
A parametric model for predicting flexural behavior of conventional reinforced concrete
and FRC is presented [70]. The approach applies to both serviceability limit state (SLS)
and ultimate limit state (ULS) criteria. Test methodology to characterize average residual
strength in fiber-​reinforced concrete has developed a calibration procedure that allows for
the design and verification of mechanical properties as well as any section required for a
given set of loading conditions. Using a system design approach where the analysis tools
(e.g. structural analysis) and the design optimization tools (e.g. non-​linear programming)
are closely linked to design requirements. The models can be implemented for both strain-​
softening and hardening FRC and extended to hybrid reinforced concrete (HRC), which
addresses structural members that combine continuous reinforcement with randomly dis-
tributed chopped fibers in the matrix [71].
The parameters for the constitutive models are obtained from either ASTM C1609/​
C1609M or BS EN 14651:2005 based on the recommendations of a parametric design
method as discussed in [61]. There are a minimum number of parameters for use in the ACI
274 Handbook of Precast Segmental Tunnel Lining Systems

Figure 4.5 Test set-​u p for flexural ASTM C1609 test.

Note: Arrangement to obtain net deflection by using a transducer (linear variable differential
transformer or LVDT) mounted on jig secured to specimen directly above supports.

Figure 4.6 Back-​c alculation parameters: (a) compressive; (b) tensile.

Note: (a) Bi-​l inear compression model; (b) tri-​l inear tension model for both strains softening and
strain hardening.

544-​8R model which include the ratio of compressive to tensile strength, the post crack
tensile stiffness and strain capacity in the post-​peak range, the post-​crack tensile residual
strength, and the ultimate allowable compressive strain as defined in Figures 4.6(a) and
(b) for tension and compression response. Using these parameters, the location of the neu-
tral axis, moment capacity and curvature of the section can be obtained at any imposed
 Precast concrete technologies 275

Figure 4.7 Experimental data with an average curve.

Note: Average, and average ±1.34 STD curves in comparison with simulation curves for (a) poly-
meric fiber and (b) steel fiber reinforcement.

strain. In the following curves, the results obtained from the back-​calculation process are
represented. In this procedure, the material properties of the samples are extracted from the
four point-​bending test results. These parameters enable one to predict the stress-​strain dia-
gram for the FRC materials for use in the design. A simulation of the average test results is
shown in Figures 4.7(a) and (b).

4.9 MATERIAL TESTING LEVELS AND BACKCALCULATION

APPROACHES
The compressive strength of tunnel segments is generally around 40 to 50 MPa (5,800 to
7,250 psi) range. Higher-​strength concretes may fail by brittle behavior at corners or uneven
surfaces during installation, transportation or handling process. Subsequently, repair of
damaged areas in very high strength concretes from spalling are often identified as weak
points in terms of durability. A dense concrete texture with low porosity should be targeted
for segments, especially in concrete cover regions. During the material selection process the
exposure conditions should be considered for durable service life.
Mechanical properties of fiber-​reinforced concrete (FRC) in tension and flexure can be
obtained using standard flexural tests such as ASTM C1609 [69] or EN14651 [72]. These
results can be incorporated to calculate the capacity of a section using simulations in order
to validate the experimental results or design factors of safety using analytical or finite
element solutions. Design guidelines for the section capacity of reinforced or hybrid sections
that substitute fibers for conventional rebars with both steel and macro-​synthetic fibers can
thus be developed. The hybrid design method improves the concrete post-​cracking behavior
considerably, reduces shrinkage crack width, improves fatigue and impact resistance and
therefore improves the durability [6, 73].
276 Handbook of Precast Segmental Tunnel Lining Systems

4.10 FULL-​SCALE TUNNEL SEGMENT TESTING

One of the main engineering objectives for designing tunnel segments is to develop better
predictive models and calibrate the proper structural models for the tunnel segments based
on the material and constitutive models that can predict the response from the material to
the precast section and the full ring test results. There is also a need to incorporate innova-
tive tools and methods for testing, analysis and design towards the serviceability-​based
design of tunnel segments under flexural or compression loads. The approach is also uti-
lized to validate the direct incorporation of material properties in the response of hybrid
reinforcing, or serviceability level requirements for precast segments. The areas of need
include serviceability measures that can address stiffness, strength, potential localized
cracking, long-​term durability and ductility under service loading as well as the ultimate
strength domains.
It is expected that by combining the state-​of-​the-​art full-​size experiments, analysis and
design methods for precast hybrid tunnel segments, one can validate the design parame-
ters. Serviceability-​based criteria utilize material properties such as residual strength
and stiffness computed from materials level tests. The experimental set up for the full-​
scale precast panels are shown in Figure 4.8. Closed-​form solutions based on strain com-
patibility analysis and hinge formation based on nonlinear moment-​ curvature and
optimization tools are used in the calculation of the load and deformation response.
Results have been validated in a state-​of-​the-​art laboratory recently constructed for
full-​scale tunnel lining sections. Load-​displacement results, failure modes, and strain
distributions measured by 3D digital image correlation (DIC) measurements will be
used for analysis [74].

Figure 4.8 Test set-​u p for segment full-​s cale flexural test.

Note: Line load conditions and testing arrangement to obtain deflection using transducers (linear
variable differential transformer or LVDT) to test full-​s cale tests for bending.
 Precast concrete technologies 277

4.11 APPLICATION OF STATISTICAL PROCESS CONTROL FOR

QUALITY CONTROL
Statistical process control procedures are applied in a range of industrial productions to
reduce variability and improve production efficiency. Quality control procedures are used
in monitoring the production, delivery schedule and construction process as an extremely
useful tool to apply to the historical data that is collected. Results are used to gain insight
into the operational procedures and allow for quick response time and decision making
regarding the modifications to the process; reduce the liabilities and benefits to the owners,
suppliers and clients if any of the design parameters fall below the minimum requirements.
The longer time it takes to detect discrepancies in the data, the more the penalty, project
delays and the higher the associated costs to the owners and suppliers.
Detailed analysis on the test results of flexural closed-​loop ASTM C1609 test for a pro-
duction run of more than 378 production samples for precast tunnel lining of a construc-
tion project. Statistical analysis was conducted on the test results of steel fiber reinforced
concrete mixtures used in the production of tunnel lining segments. Using the test results of
1 and 28 day flexural peak strength, and residual strength at various levels. The 28 day test
results flexural property results conducted as a part of the quality control requirements were
analyzed. The number of parameters evaluated includes flexural strength and deflection at
first crack, ultimate flexural strength and the residual flexural strength results at L/​600 and
L/​150. Figures 4.9(a) and (b) show the range of the values of the test results of a series of
quality control tests conducted over a 6 month project.
The statistical process control and the range of the data were studied in the context of
material properties as well as back-​calculation of the tensile parameters. Test data were also
used for the statistical analysis of the correlation of the parameters using a combination
of control charts for early detection of spurious shifts in the mean of the test data (out-​of-​
control signal). Results show that the cumulative sum (CUSUM) and exponentially weighted
moving average (EWMA) methods are able to address the variations and control residual
strength results at L/​600 and L/​150 [75]. Analysis of the combination of the control charts
identify periods of the production process where the process is in or out of control or cap-
able in segment fabrication. Early detection of shifts from CUSUM and EWMA charts with
run charts allow one to take the corrective actions in tunnel lining project that can improve
the quality control process [76].
Analysis of the combination of the control charts identify periods of the production pro-
cess where the process is in or out of control or capable in segment fabrication. Early
detection of shifts from CUSUM and EWMA charts with run charts allow one to take
the corrective actions in tunnel lining projects that can improve the quality control pro-
cess. These discussions however are beyond the scope of this chapter addressing how the
variations in the individual material data move towards the average results of the distribu-
tion curve.

4.12 FRESH STATE, RHEOLOGY TESTS AND WORKABILITY

The flow and rheology characteristics of a concrete mix are dependent on the materials and
the mix design proportions. Admixtures can be used to alter the rheology to address slump
and segregation issues. The use of rheometer tests in concrete technology addresses the work-
ability and fresh state performance through the terms such as yield stress, which represents
the slump value and a plastic viscosity that quantifies the cohesion aspects of the mixture.
Rheology measurements provide an insight into the effect of individual mix constituents and
278 Handbook of Precast Segmental Tunnel Lining Systems

Figure 4.9 Average and standard deviations of stress deflection.

Note: Average and standard deviations of stress-​d eflection plot of 271 individual samples at
28 days and flexural strength run chart.

quantify the parameters that affect the cohesion and flow. For example, adding water to a
mix reduces yield stress by increasing the slump but also reduces plastic viscosity (cohesion)
increasing the risk of bleeding and segregation [77]. The rheology tests are quite varied and
range from basic to empirical measurements of properties. They measure the amount of
shear stress to achieve the twisting torque required to impose a shearing strain rate on the
fresh concrete sample. As a measure of the flowing capability and adhesion of particles, the
shear stress is plotted vs. the rate of shear strain [78].
Various rheological tests are available in the testing of fiber reinforced concrete in terms
of flow characteristics as discussed in (ACI544-​PRC 2R). High cement contents and low
water content as well as the extra small SCM particles often result in the high plastic vis-
cosity and cohesiveness of high-​strength concrete. In areas such as underwater and pumped
concrete the added cohesiveness is valuable since they increase concrete’s resistance to seg-
regation. In self-​compacting concrete, the yield stress is required to be low while the plastic
viscosity is maintained to a level to meet the flow requirements. These requirements allow
for the selection of admixtures to achieve the required values of yield strength and plastic
viscosity.
The workability of the segment concrete and duration of vibration should be studied
during the mix design development phase. Tunnel segments can be produced in a carousel
system or in stationary production. The workability of a precast section can be low in
carousel systems and very high, even self-​consolidating in stationary systems. When there
is a sign of segregation present the potential for excessive vibration during consolidation
should be examined. [79]
When the concrete slump is 50 mm (2 inches) and below, it is placed with heavy vibra-
tion. The flow of concrete and its stability during vibration, as well as the surface finishing
properties, are important parameters to control. Some tunnel projects have very low water
 Precast concrete technologies 279

tightness and surface cracking requirements that may require none or minimal cracking and
no bug holes, especially around tunnel sealers.
When low slump concrete is used for the production of tunnel segments, its rheology is
very critical. Vibration is necessary to make the concrete fluid. Although the accuracy of its
measurements is not found to be accurate slump, Ve-​be (ASTM C1170) or DIN flow table
(EN 12350-​5) are the most common workability testing methods. Other rheology and test
methods that include vibration as a part of the testing procedure may be used.

4.13 DURABILITY PARAMETERS AND ASPECTS OF SERVICE LIFE

Environmental exposures of tunnel segments can be considered as a general exposure list
and involve the following concerns:

• Carbonation shrinkage and reduction of the corrosion resistance

• Sulfate-​containing water or soil with dissolving and leaking to the microstructure
• Chlorides due to de-​icing salts (e.g. tunnel access zones) and corrosion of reinforcing
steel
• Fire and fire extinguishing techniques
• Degradation from thermal or crystallization pressure due to climatic changes such as
freeze–​thaw.

The crack width and the water-​tightness are important factors in the design of tunnel sec-
tions. The use of fibers has been significantly helped with the reduction of the water perme-
ability to the effect of fibers in reducing the crack width as well as the increased tortuosity
of the crack path.
Frost damage results from the freezing and expansion of water within the capillary
system. The tensile forces may cause surface scaling or crack the concrete. Capillaries pores
are formed due to excess water present in capillary pores which were added for concrete
for workability enhancement. Most frost damage occurs in concrete due to the saturated
aggregates [77, 80].
Several mechanisms that include hydraulic and osmotic pressures, contribute to the dam-
age evolution due to freeze–​thaw damage in hydrated cement paste. Hydraulic pressure is
the source of damage in the aggregates due to water expansion due to freezing. Aggregates
with low or very high porosity are generally prone to failure by hydraulic pressure because
the amount of freezable water is so low, or the pore sizes are significantly large to withstand
the expansion volume. Aggregates with intermediate porosity are the least durable.
Many of the current structural design approaches lack objectivity in defining the important
parameters needed to characterize and define the durability failure criteria of concrete struc-
tures. At the same time, metrics and test methods for the determination of the acceptability
criteria are still under development and various approaches are not uniformly reached a
consensus approach. As we move towards serviceability-​based designs and performance-​
based specifications, we need to acknowledge that many durability measures are still defined
and related primarily to water to cement ratio and compressive strength of concrete [81].
Analysis of the existing durability assessment methods shows that these two metrics are not
proper indicators of the success of concrete structures. As long as the compressive strength
is used as the main governing parameter for the prescription of durability and service life,
the challenge of introducing indicators for better specifications remains a difficult goal to
accomplish.
280 Handbook of Precast Segmental Tunnel Lining Systems

Currently, there is little true durability-​based design applied to reinforced concrete struc-
tures. While many engineers are aware of the deterioration mechanisms, rational prescrip-
tion of economic solutions and decision support tools in terms of modeling of deterioration
rates are absent. These deficiencies significantly affect the confidence and reliability of the
prediction of service life.
An in-​depth discussion of serviceability-​based design and design for durability are avail-
able through RILEM Committee publications [82]. For example, European standard EN
206-​1 [83], addresses the specification, performance, production, and conformity for con­
crete construction via a prescriptive set of specifications. Lack of agreement on test methods
has resulted in the inability to move forward with the existing test methods. This accen-
tuates the need for a better definition of parameters that affect the rates of deterioration
and can be quantified under the production schedule. There are many test methods under
various stages of development that address appropriate test methods reviewed by RILEM
TC-​NEC [84]. These tests include chloride penetration, carbon dioxide using an accelerated
carbonation chamber, electrical resistivity, gas permeability, gas permeability, oxygen per-
meability and several others.
Several studies [85,86] indicate that implementation of performance-​based specifications
by proper prediction of service life design requires several key features and parameters
addressing the following.

• Llimit state criteria

• Defined service life
• Deterioration models
• Compliance tests
• Maintenance and repair strategies
• Quality control systems

Bentur and Mindess [57] define the steps in the design process into four steps, including the
following:

1. Fundamental mechanisms and quantification of models to address the physical or

chemical deterioration processes;
2. Definition of the limit states in terms of the critical points of acceleration of the deteri-
orating mechanisms such as time to corrosion, excessive cracking, delamination and
spalling of the concrete, collapse due to loss of cross-​section of the reinforcement, etc.;
3. Probabilistic approach to calculate the limit states of durability demand exceeding the
supply due to the material resistance; and
4. Definition of the type of limit state, and its consequences in terms of economy, service-
ability, or collapse or structural failure leading to the loss of life [i.e. a serviceability
limit state (SLS), or an ultimate limit state (ULS)] [86].

Transport properties play the dominant role in understanding the service life and durability
modeling of concrete. Research in this area has made tremendous strides, improvements
and contributions to a better understanding of the behavior of concrete materials [82].
The durability modeling challenges are not trivial. For example, the general governing phe-
nomena for mass transfer of harmful ions such as sulfates or chlorides through concrete is
modeled using conservation of mass type of partial differential equations that include diffu-
sion, and convection, in addition to chemical reactions, reactivation, binding, and sorption.
 Precast concrete technologies 281

Similar approaches also are operative for moisture diffusion into and out of concrete in the
case of free or restrained shrinkage.

4.14 DIFFUSIVITY-​B ASED APPROACHES FOR CONTROL OF

SERVICE LIFE
Starting with the diffusion of water through concrete, any accompanying ions such as chlo-
rides or sulfates become the main actors in the internal or external attack. Examples include
the reaction of sulfates to form ettringite, or binding of the chlorides with the aluminate
phases, as well as the critical concentration of alkalis from the cement phase, which con-
tribute to the formation of ASR gel and its subsequent expansion with water.

4.14.1 Transport properties
The major indicators of transport characteristics that affect the design life are the ionic
diffusion parameters. The diffusivity parameter, D, is the constant of proportionality. The
diffusion process is often modeled using Fick’s first and second laws, which use the concen-
tration gradient of the specific ions as the forcing function and the driving force to main-
tain the equilibrium. Fick’s first law states that the diffusive flux is linearly proportional to
the concentration gradient, however, it is only applicable under steady-​state conditions. To
describe the kinetics of diffusion, Fick’s second law is used, which is represented in terms of
space and time and predicts the diffusive flux to be only a function of concentration gradient
at the point and a function of time.
Fick’s second law can be converted to a diffusion-​reaction equation and accompanied by
a second term that is due to the reaction of the external ions with the cement constituents
or hydration products. A solution strategy for this class of problems was proposed [87].
The secondary reactions may also result in a reduction of the ions in the reaction front flux
and thus a delaying effect. Therefore, as the diffusion proceeds inward, a reaction front is
characterized by an effective ionic concentration that is conducive to the expansive reactions
that result in the decomposition of cement paste constituents, or the corrosion of rebars
due to their loss of passivation by chlorides. This leads to a positive feedback loop, where
the net consequence of the expansive product formation leads to microcracking of the bulk
paste, formation of new microcrack paths available for diffusion, and thus increasing the
diffusion constants for a higher ionic flux, often by an order of magnitude [88].
The coefficients of diffusion for various ionic species in concrete depend on the mixture
parameters such as aggregate size, type, volume, pozzolan type and dosage, degree of hydra-
tion, w/​cm, additives, etc.), as well as curing conditions (time, moisture, temperature, etc.).
The mathematical basis for the diffusion-​reaction equations can be used for various sources
of ions by changing the ionic species, and the coefficients of diffusion that are relevant to
the durability problem. Ionis species such as chlorides (Cl−) and sulfates (SO4-​2) have similar
diffusion rates in normal concrete [89].
The diffusivity parameter is affected by the internal capillary porosity as well as the inter-
facial transition zone (ITZ). The transport properties at the ITZ directly correlate with their
higher porosity, larger pore sizes and higher density of calcium hydroxide crystals in com-
parison to the bulk CSH. The diffusion coefficient of ITZ is reported to be 6 to 12 times
more than the bulk cement paste [88, 90]. The correlation between the transport properties
of cement paste with and without the SCM is a clear indication of durability improvements
due to the use of SCM materials.
282 Handbook of Precast Segmental Tunnel Lining Systems

4.14.2 Corrosion–​c hloride diffusion

The corrosion of reinforcing bars is a direct consequence of the chloride-​induced break-
down of the passive film that surrounds the rebar. Under the highly alkaline nature of the
hydration products, the passive film protects the rebar, however the migration of chlorides
to the rebar level reduced the environmental pH. Steel reinforcements are expected to carry
the tensile loads and their failure will affect a significant reduction of the structural cap-
acity. Deterioration of a metal due to corrosion is due to an electrochemical reaction with
the electrolyte.
There are four main four components of the process that include the anodic reaction
[which is the actual corrosion event (metal dissolution)], the cathodic reaction (which con-
sumes the electrons produced by the anodic reaction), the medium (electrolyte), and an
electronic path. As metal dissolves into ions and enters the solution, electrons are left behind
which accumulate. If there is no cathodic reaction, the electron density would slow down
the rate of metal dissolution.
The cathodic reaction is a major contributor to the corrosion process to complete the
loop by removing the electrons from the cathode. The typical anodic and cathodic reac-
tions for the dissolution of iron in cementitious systems within a highly alkaline medium
is: Fe →Fe2+​+​2e-​, as the oxidation component or anodic reaction and O2 +​2H2O +​ 4e-​ →
4OH-​, as the reduction or cathodic reaction. Angst [91] addresses the chloride threshold
value as an important parameter that determines the onset of corrosion in comparison to
the chloride ingress profile.

4.14.3 Diffusion-​based sulfate attack

Concrete materials may be subjected to sulfate attack which manifests itself as two types
of damage: loss of matrix strength due to degradation of hydration products such as
CSH, and cracking due to internal tensile stresses due to volumetric expansion [92].
This expansion is attributed to either gypsum or ettringite formation that leads to crack-
ing [93]. Loss of strength due to chemo-​mechanical reactions and decalcification of the
cement paste hydrates such as CSH, while cracking and expansion is attributed to the
formation of expansive compounds. Various durability models have been developed for
external sulfate attack [87, 94]. Many models use an empirical relationship to incorp­­
orate the expansion due to ettringite formation, often the expansive strain and ettring-
ite volume have a linear correlation. This approach was used in the 4SIGHT program
addressing the durability of concrete structures and the service life of modeling due to
transport mechanisms. [95]
The rate-​controlling step in the case of sulfates is the reaction term of the diffusion that
occurs based on an empirical linear relationship that links the depth of deterioration at a
given time to the tri-​calcium silicate (C3A) content and the concentration of magnesium and
sulfates in the original solutions. Solutions of diffusion-​reaction equations in both numerical
terms, analytical series solutions and finite element approach are available. The most
basic approach treats the problem as a first-​order chemical reaction to determine the sulfate
concentration as a function of time and space. The matrix is defined through its microstruc-
tural pore connectivity and transport parameters using the diffusion coefficient. Non-​
linear parameters of the diffusion coefficient can also be represented as a function of the
capillary porosity, packing density or particles, SCM content. The diffusion constants vary
with time since capillary pores fill up with the recently formed minerals through pore plug-
ging or increase due to microstructural damage and microcracking. A model based on the
 Precast concrete technologies 283

Figure 4.10 Concentration profiles for 1-​D diffusion.

Note: Concentration profiles for 1-​D diffusion with reaction (dashed lines) and without reaction
(solid lines). Lines refer to 1 to 5 years of diffusion for sulfate ions.
Source: [96]. 

diffusion-​reaction equations predicts the expansion of mortar bars in the 1-​D case [94] as
shown in Figure 4.10.
To simulate 1-​D diffusion, ASTM C1556 is used to measure the apparent coefficients
of diffusion for ions by wet chemical methods (e.g. NaCl for chlorides and Na2SO4 for
sulfates) from a singly exposed side [97]. The effective ionic diffusivity ‘D’ of cement paste
normalized to the diffusivity of ion in unconfined water, ‘D0’, can be related to the porosity
as described by Garboczi and Bentz [98]. Parameter D0 for sulfate ions was reported by
Hilsdorf in the order of 10−9 m2/​s [90]. This approach was used in the formulation devel­
oped by Bonakdar et al. [96]. The diffusion coefficients of sulfate ions (D) for various con­
crete mixtures are in the range of 10−14 to 10−11 m2/​s.
Figure 4.10 shows a semi-​infinite medium subjected to chemical ingress from one side as a
1st order reaction. The condition simulates a slab or a wall exposed to the initial condition
of C(x,0) =​ Ci and boundary condition of C(0,t) =​ C0 where D is the apparent coefficient
of diffusion and k is the chemical reaction coefficient. The solution by error function solu-
tion for a simplified case of Ci =​0. Typical concentration profiles considering the presence
or absence of the reaction term and assuming D =​10−12 m2/​s and k =​10−8 s−1 are shown
in Figure 4.10 and represent the fact the reaction term has a damping effect on the rate of
penetration assuming that the products of the reaction are stable.

4.14.4 Alkali-​silica reactions
Alkali-​silica reactivity is of major concern in the final precast concrete segmental tun-
nel liner product. Alkali-​silica reactivity is a two-​step process. First, there is a chemical
284 Handbook of Precast Segmental Tunnel Lining Systems

reaction between the alkali phases in the cement that are primarily considered as impur-
ities and reactive silica present in aggregates thus resulting in an alkali-​silica gel. Once
moisture encounters the gel, the results are internal pressures causing cracks within the
concrete product, which ultimately decreases the product’s strength, durability and imper-
meability [99].
Damage induced due to ASR reactions has been known for more than 60 years in the
field of concrete technology. Various methods are available based on the determination
of deleterious aggregates, alkali concentration in cement as well as the use of SCMs. The
damage mechanisms have been effectively addressed using specifications on the cement and
aggregates as well as the use of pozzolanic materials partly to reduce the calcium hydroxide
in the microstructure and change the diffusivity of the matrix.
The effect of blended cement on the morphology and chemical composition of reaction
products has addressed the change in the morphology of ASR gel that forms and its expan-
sion potential. Experimental data from ASTM C1567 [100] test method and microstruc­
tural studies including ESEM and quantitative EDS were used to develop a physio-​chemical
model of the different silicate glass structures [101]. Several fly ashes with various properties
were used to study the number of bridging and non-​bridging oxygens in the gel network
providing insight into ASR products.
Many chemical and mechanical models developed based on numerical and analytical
approaches address the diffusion of alkali ions (usually Na+​, K+​) into the cementitious
system, their reaction with amorphous silica from aggregates, and the formation of an
alkali-​silica gel [102]. These models have been compared and summarized through review
papers [103] and [104] to mention a few are based on gel permeation into the pores, swell­­
ing due to uptake of water, and generation of hydrostatic pressures. Micromechanics is
used to estimate the expansive stresses of spherical particles due to internal pressure. The
behavior of the silica gel is either elastic (solid-​state) or elasto-​plastic (liquid-​solid state) and
is considered as silicate glass whose structure depends on the alkali composition of sodium
and calcium phases.

4.15 FIRE PROTECTION AND DAMAGE CONTROL

Spalling mechanisms are attributed to the presence of moisture in the pore structure of the
hydrated cement paste, which creates pressure due to vaporization within the internal pores.
Without a path for pressure release, as the internal pressure exceeds the tensile strength,
explosive spalling occurs. The spalling damage can penetrate up to 150 mm (6 inches) into
the concrete as successive layers of concrete peel off as its face is exposed to fire. Spalling is
a major concern because the loss of concrete cover may compromise the load-​carrying cap-
acity of structural members. As the concrete cover spalls, the underlying reinforcing steel is
exposed to high temperatures and deteriorates rapidly.
Fibers improve the fire resistance of concrete by increasing the residual strength and
toughness to prevent explosive spalling. Short-​fiber reinforcement bridges the cracks to
maintain the integrity of the damaged material. Gambarove showed that steel fibers at a
high volume fraction of the order of 2% are needed to maintain residual strength after high-​
temperature loading [105] while Colombo et al. [11] found that steel fibers enhanced fire
resistance of concrete slabs by three to nine times longer than control specimens without fib-
ers. Synthetic microfibers (defined per ACI 544.3R), are quite effective in reducing the risk
of concrete spalling in tunnel linings subjected to fire. The melting of the polypropylene due
to the fire creates channels for the steam to escape, thus reducing the build-​up of pressure
when a moderately low dosage of 0.4 kg/​m3 (0.7 lb/​CY) is used.
 Precast concrete technologies 285

ACI 544-​5R discusses the fire resistance tunnel lining precast panels and means to remedy
the situation (ACI 544-​5R-​2010). The fire resistance of assemblies and systems are normally
determined using the ASTM E119 fire test standards which create a temperature rise of
1,093°C (2,000°F) over three hours. Hydrocarbon-​fueled fires may result in catastrophic
spalling failure and have a much faster heating rate. Polymeric fibers are quite effective in
the prevention of explosive spalling due to their low melting point [about 160°C (220°F)].
The effectiveness depends on the size, distribution, and volume content of the polypropylene
(PP) fibers.

4.16 PRE-​C ONSTRUCTION TESTING FOR SELECTING OPTIMUM

CONCRETE MIXTURE
The selection for an optimum concrete mix design is of primary consideration for any tun-
nel project. During the design phase, the engineers take into consideration its final function
in life both internally and externally with its surrounding ground conditions. The need for
pre-​construction testing is based on the project’s geographical location where materials,
cement type, supplementary cementitious materials, fine and coarse aggregates along with
concrete admixtures, will determine the concrete mix design’s performance. Considerations
to the availability of raw materials in the proximity of the precaster’s manufacturing plant
should also be realized with its economy of scale.
During the pre-​construction, pre-​manufacturing phase, the precaster is required to select
the concrete mix design, which will meet the client’s approval and performance based on
the specifications. The concrete mix design should also take into consideration the manu-
facturing process dictated by schedule demands. The manufacturing process can either be
a static mode or a carousel system. With both manufacturing systems being different, they
will still be required to achieve the same design requirements of compressive strengths,
residual flexural strengths when using fibers, stripping/​demolding strengths, and durability
testing parameters.
Based on the specific tunnel projects requirements, design engineers will require a min-
imum amount of cement and type supplemented by cementitious materials such as silica
fume, granulated ground blast furnace slag, or fly ash to meet the requirements set forth by
governing technical agencies such as ACI, BS 8110 or Eurocode 2.
Referencing the ACI 318-​14 Table 19.3.1.1 shown again in Table 4.2, the different cat­
egories and classes are considered by the design engineer during the design phase and the
precaster should perform preliminary testing to meet the engineer’s requirements.
The following testing of the concrete mix designs should be considered for the segmental
tunnel liners.

1. Compressive Cylinder Strength ASTM C39/​C 39M-​2 0

2. Stripping/​D emolding Strength ASTM C39/​C 39M-​2 0
3. Initial Set Times ASTM C403/​C 403M-​1 6
4. Residual Strength Using Flexural Beam EN 14651:2007
5. Split Tensile Strength ASTM C496/​C 496M-​1 7
6. Chloride Ion Penetrability ASTM C1152/​C 1152M-​2 0
7. Rapid Chloride Migration ASTM C1202
8. Relative Carbonation Resistance CEN/​T S 12390-​1 0
286 Handbook of Precast Segmental Tunnel Lining Systems

9. Standard Method of Test for Surface Resistivity AASHTO T358

Indication of Concrete’s Ability to Resist Chloride Ion
Penetration
10. Chloride Migration Coefficient from Non-​S teady State NT BUILD 492
Migration
11. Alkali–​A ggregate Reaction Resistance ASTM C1567-​1 3
12. Water Permeability CD Lawrence Method
13. Freeze-​T haw Resistance ASTM C666
14. Resistance to Degradation of Small-​S ize Coarse ASTM C131/​C 131M-​2 0
Aggregate by Abrasion and Impact in the Los Angeles
Machine

Table 4.2 Concrete exposure categories.

Summary of exposure categories for concrete by exposure class by per ACI 318-​1 9,
Table 19.3.1.1

Category Class Condition

Freezing and F0 Concrete not exposed to freezing-​a nd-​t hawing cycles
thawing (F)
F1 Concrete exposed to freezing-​a nd-​t hawing cycles with limited
exposure to water
F2 Concrete exposed to freezing-​a nd-​t hawing cycles with frequent
exposure to water
F3 Concrete exposed to freezing-​a nd-​t hawing cycles with frequent
exposure to water and exposure to deicing materials
Sulfate (S) Water-​s oluble sulfate (SO 42-​) Dissolved sulfate (SO 42-​)
in soil, percent by mass [1]‌ in water, ppm [2]‌
S0 SO 42-​ < 0.10 SO 42-​ < 150
S1 0.10 < SO 42-​ < 0.20 150 < SO 42-​ < 1,500 or
seawater
S2 0.20 < SO 42-​ < 2.00 1,500 < SO 42-​ < 10,000
S3 SO 42-​ > 2.00 SO 42-​ > 10,000
In contact W0 Concrete dry in service
with water Concrete in contact with water and low permeability is not
(W) required
W1 Concrete in contact with water and low permeability is required
Corrosion C0 Concrete dry or protected from moisture
protection of
reinforcement C1 Concrete exposed to moisture but not to an external sources of
(C) chlorides
C2 Concrete exposed to moisture and an external sources of
chlorides from deicing chemicals, salt, brackish water, seawater, or
spray from these sorces
Notes:
[1]‌ Percent sulfate by mass in soil shall be determined by ASTM C1580
[2]‌ Concentration of dissolved sulfates in water, in ppm, shall be determined by ASTM D516 or ASTM D4130
 Precast concrete technologies 287

These test results are taken into account in determining the concrete mix design’s durability
and the extent of the structure’s life serviceability. In summary, the concrete mix design’s
material proportions have to be taken into account when developing its final acceptable state.

REFERENCES
[1]‌ J. S. Damtoft, J. Lukasik, D. Herfort, D. Sorrentino, and E. M. Gartner, “Sustainable devel-
opment and climate change initiatives,” Cement and Concrete Research, vol. 38, no. 2, pp.
115–​127, Feb. 2008. doi: 10.1016/​j.cemconres.2007.09.008
[2]‌ G. P. Hammond, and C. I. Jones, “Embodied energy and carbon in construction materials,”
Proceedings of the Institution of Civil Engineers–​Energy, vol. 161, no. 2, pp. 87–​98, May
2008. doi: 10.1680/​ener.2008.161.2.87
[3]‌ A. D. L. Feeney, C. S., S. Thayer, M. Bonomo, and K. Martel, “White Paper on Condition
Assessment of Wastewater Collection Systems,” Washington, D.C., 2009. doi: https://​cfpub.
epa.gov/​si/​si_​p​ubli​c_​re​cord​_​Rep​ort.cfm?Lab=​NRMRL&dir​Entr​yId=​209​530
[4]‌ Tunnel Business Magazine, “https://​tunn​elin​gonl​ine.com/​upcom​ing-​tunnel​ing-​proje​cts/​.”
https://​tunn​elin​gonl​ine.com/​upcom​ing-​tunnel​ing-​proje​cts/​
[5]‌ “Water tunnel business ‘exploding’ as technology reduces risk and cost.” www.enr.com/​artic​
les/​44133-​cit​ies-​dig-​water-​tunn​els-​to-​meet-​long-​term-​dema​nds (accessed Nov. 30, 2018)
[6]‌ Bakhshi, M., and Nasri, V. “Cracking serviceability limit state for precast concrete tunnel
segments,” Canadian Society of Civil Engineers (CSCE) 2015 General Conference, Regina,
SK, Canada, 2016.
[7]‌ U.S. Environmental Protection Agency (USEPA), Clean Water and Drinking Water
Infrastructure Gap Analysis. Washington, District of Columbia, 2002.
[8]‌ A. D. L. Feeney, C. S., S. Thayer, M. Bonomo, and K. Martel, “White paper on condition
assessment of wastewater collection systems,” Washington, D.C., 2009. doi: https://​cfpub.
epa.gov/​si/​si_​p​ubli​c_​re​cord​_​Rep​ort.cfm?Lab=​NRMRL&dir​Entr​yId=​209​530
[9]‌ ACI Committee 350, Environmental Engineering Concrete Structures. American Concrete
Institute, 2003.
[10] M. di Prisco, G. Plizzari, and L. Vandewalle, “Fibre reinforced concrete: New design perspec-
tives,” Materials and Structures, vol. 42, no. 9, pp. 1261–​1281, Nov. 2009. doi: 10.1617/​
s11527-​009-​9529-​4
[11] M. Colombo, P. Martinelli, and M. di Prisco, “A design approach for tunnels exposed to
blast and fire,” Structural Concrete, vol. 16, no. 2, pp. 262–​272, Jun. 2015. doi: 10.1002/​
suco.201400052
[12] P. Hewlett, Lea’s Chemistry of Cement and Concrete. Elsevier, 2003. doi: 10.1016/​
B978-​0-​7506-​6256-​7.X5007-​3
[13] P. J. Jackson, “Portland cement: Classification and manufacture,” in P. Hewlett (ed.),
Lea’s Chemistry of Cement and Concrete, Elsevier, 1998, pp. 25–​94. doi: 10.1016/​B978-​
075066256-​7/​50014-​X
[14] W. C. P. Kosmatka, H. Steven, and B. Kerkhoff, Design and Control of Concrete Mixtures,
14th ed. Portland Cement Association, 2002.
[15] P. Hewlett, Lea’s Chemistry of Cement and Concrete. Elsevier, 2003.
[16] ASTM-​C150-​18, “Standard specification for portland cement,” 2018.
[17] “CSA A3000-​2018 cementitious materials compendium,” 2018.
[18] EN 197-​1, “EN 197-​1-​2011 Cement–​Part 1: Composition, specifications and conformity
criteria for common cements,” 2011.
[19] K. Scrivener, V. John, and E. Gartner, Eco-​Efficient Cements: Potential Economically Viable
Solutions for a Low-​CO2 Cement-​Based Materials Industry. 2016.
[20] D. K. Panesar, “Supplementary cementing materials,” in Developments in the Formulation
and Reinforcement of Concrete, Elsevier, 2019, pp. 55–​ 85. doi: 10.1016/​B978-​0-​08-​
102616-​8.00003-​4
288 Handbook of Precast Segmental Tunnel Lining Systems

[21] V. M. Malhotra, V. S. Ramachandran, R. F. Feldman, and P.-​C. Aïtcin, Condensed Silica

Fume in Concrete. CRC Press, 2018. doi: 10.1201/​9781351070843
[22] E. J. Bentz, and D., Garboczi, “Simulation studies of the effects of mineral admixtures on the
cement paste-​aggregate interface zone,” Aci Materials Journal, vol. 88, pp. 518–​529, 1991.
[23] G. L. Kalousek, L. C. Porter, and E. J. Benton, “Concrete for long-​time service in sul-
fate environment,” Cement and Concrete Research, vol. 2, no. 1, pp. 79–​89, Jan. 1972.
doi: 10.1016/​0008-​8846(72)90025-​7
[24] M. C. G. Juenger, R. Snellings, and S. A. Bernal, “Supplementary cementitious materials: New
sources, characterization, and performance insights,” Cement and Concrete Research,
vol. 122, pp. 257–​273, Aug. 2019. doi: 10.1016/​j.cemconres.2019.05.008
[25] M. Mobasher, B. Arora, A. Aguayo, N. Kianmofrad, F. Yao, and Y. Neithalath,
“Developing ultra-​high–​performance concrete mix designs for arizona bridge element con-
nections,” 2019.
[26] A. Arora et al., “Microstructural packing–​and rheology-​based binder selection and charac-
terization for Ultra-​high Performance Concrete (UHPC),” Cement and Concrete Research,
vol. 103, pp. 179–​190, Jan. 2018. doi: 10.1016/​j.cemconres.2017.10.013
[27] N. Neithalath, J. Persun, and A. Hossain, “Hydration in high-​performance cementitious
systems containing vitreous calcium aluminosilicate or silica fume,” Cement and Concrete
Research, vol. 39, no. 6, pp. 473–​481, Jun. 2009. doi: 10.1016/​j.cemconres.2009.03.006
[28] P. K. Malhotra, and V. M. Mehta, “High-​ performance, high-​ volume fly ash con-
crete: Materials, mixture proportioning, properties, construction practice, and case histories,”
in High-​performance, High-​volume Fly Ash Concrete: Materials, Mixture Proportioning,
Properties, Construction Practice, and Case Histories. Supplementary Cementing Materials
for Sustainable Development, Ottawa, Canada, 2005.
[29] Steven H. Kosmatka, “Design and control of concrete mixtures–​The guide to application,
methods, and materials,” 8th Canadi, 2016, pp. 411.
[30] I. G. Richardson, “The nature of C-​ H in hardened cements,” Cement and Concrete
S-​
Research, vol. 29, no. 8, pp. 1131–​1147, Aug. 1999. doi: 10.1016/​S0008-​8846(99)00168-​4
[31] K. Vance, M. Aguayo, T. Oey, G. Sant, and N. Neithalath, “Hydration and strength devel-
opment in ternary portland cement blends containing limestone and fly ash or metakao-
lin,” Cement and Concrete Composites, vol. 39, pp. 93–​ 103, May 2013. doi: 10.1016/​
j.cemconcomp.2013.03.028
[32] P. Aïtcin, High Performance Concrete, vol. 19980236. CRC Press, 1998. doi: 10.1201/​
9781420022636
[33] A. Neville, and P.-​C. Aïtcin, “High performance concrete—​An overview,” Materials and
Structures, vol. 31, no. 2, pp. 111–​117, Mar. 1998. doi: 10.1007/​BF02486473
[34] P. K. Malhotra, and V. M. Mehta, “High-​ performance, high-​ volume fly ash con-
crete: Materials, mixture proportioning, properties, construction practice, and case histories,”
in High-​performance, High-​volume Fly Ash Concrete: Materials, Mixture Proportioning,
Properties, Construction Practice, and Case Histories. Supplementary Cementing Materials
for Sustainable Development, Ottawa, Canada, 2005.
[35] A. Arora, B. Mobasher, and N. Neithalath, “A generalized multi-​scale approach for the
design of novel UHPC,” in International Interactive Symposium on Ultra-​High Performance
Concrete, vol. 2, no. 1, 2019.
[36] K. Scrivener, V. John, and E. Gartner, Eco-​Efficient Cements: Potential Economically Viable
Solutions for a Low-​CO2 Cement-​Based Materials Industry. Elsevier, 2016.
[37] ACI Committee 233, “Guide to the use of slag cement in concrete and Mortar,” 2017.
[38] ASTM C 989-​ 18, ASTM C 989–​Standard Specification for Ground Granulated Blast-​
Furnace Slag for Use in Concrete and Mortars. 2018.
[39] B. B. Sabir, S. Wild, and J. M. Khatib, “On the workability and strength development of
Metakaolin concrete,” in Concrete for Environmental Enhancement and Protection, E&FN
Spon, p. 651–​656, 1996.
 Precast concrete technologies 289

[40] J. J. Brooks, M. A. Megat Johari, and M. Mazloom, “Effect of admixtures on the setting
times of high-​strength concrete,” Cement and Concrete Composites, vol. 22, no. 4, pp. 293–​
301, Jan. 2000. doi: 10.1016/​S0958-​9465(00)00025-​1
[41] M. Shekarchi, A. Bonakdar, M. Bakhshi, A. Mirdamadi, and B. Mobasher, “Transport
properties in metakaolin blended concrete,” Construction and Building Materials, vol. 24,
no. 11, pp. 2217–​2223, Nov. 2010. doi: 10.1016/​j.conbuildmat.2010.04.035
[42] J. M. Khatib, and S. Wild, “Pore size distribution of metakaolin paste,” Cement and Concrete
Research, vol. 26, no. 10, pp. 1545–​1553, Oct. 1996. doi: 10.1016/​0008-​8846(96)00147-​0
[43] W. J. Tennis, P., Thomas, M.D.A, and Weiss, “State-​of-​the-​art report on use of limestone in
cements at levels of up to 15%,” Skokie, Illinois, USA, 2011.
[44] K. Vance, A. Arora, G. Sant, and N. Neithalath, “Rheological evaluations of interground
and blended cement–​limestone suspensions,” Construction and Building Materials, vol. 79,
pp. 65–​72, Mar. 2015. doi: 10.1016/​j.conbuildmat.2014.12.054
[45] A. C. 533, “Guide for precast concrete tunnel segments,” 2020.
[46] ACI 544.7R-​16, “ACI PRC-​544.7-​16 report on design and construction of fiber-​reinforced
precast concrete tunnel segments,” 2016.
[47] M. Glavind and E. J. Pedersen, “Packing calculations applied for concrete mix design,” 1999.
[48] M. R. Jones, L. Zheng, and M. D. Newlands, “Comparison of particle packing models for
proportioning concrete constituents for minimum voids ratio,” Materials and Structures /​
Mat6riaux et Constructions, vol. 35, pp. 301–​309, 2002. doi: 10.1007/​BF02482136
[49] R. Yu, P. Spiesz, and H. J. H. Brouwers, “Mix design and properties assessment of Ultra-​
High Performance Fibre Reinforced Concrete (UHPFRC),” Cement and Concrete Research,
2014. doi: 10.1016/​j.cemconres.2013.11.002
[50] T. Stovall, F. de Larrard, and M. Buil, “Linear packing density model of grain mixtures,”
Powder Technology, vol. 48, no. 1, pp. 1–​12, Sep. 1986. doi: 10.1016/​0032-​5910(86)80058-​4
[51] G. Fu, and W. Dekelbab, “3-​D random packing of polydisperse particles and concrete aggre-
gate grading,” Powder Technology, vol. 133, no. 1–​3, pp. 147–​155, Jul. 2003. doi: 10.1016/​
S0032-​5910(03)00082-​2
[52] F. De Larrard, Concrete Mixture Proportioning: A Scientific Approach (Modern Concrete
Technology). Taylor & Francis Ltd, 1999.
[53] B. Mobasher, A. Arora, M. Aguayo, F. Kianmofrad, Y. Yao, and N. Neithalath, “Developing
ultra high-​ performance concrete mix designs for Arizona bridge element connections,”
Arizona State University. School of Sustainable Engineering and the Built Environment,
FHWA-​AZ-​19-​745, 2019. https://​rosap.ntl.bts.gov/​view/​dot/​55498
[54] P. J. F. Wright, “Entrained air in concrete,” Proceedings of the Institution of Civil Engineers,
vol. 2, no. 3, pp. 337–​358, May 1953. doi: 10.1680/​iicep.1953.11041
[55] D. Sniderman, “Material matters: Water-​ reducing admixtures,” Precast Inc. Magazine,
vol. May/​June, pp. 14–​17, 2015.
[56] F. De Larrard, Concrete Mixture Proportioning: A Scientific Approach (Modern Concrete
Technology). Taylor & Francis, 1999.
[57] A. Bentur, and S. Mindess, “Mechanics of fibre reinforced cementitious composites,” in
Fibre Reinforced Cementitious Composites, 2nd ed., CRC Press, 2006, pp. 125–​ 192.
doi: 10.1201/​9781482267747-​12
[58] B. Mobasher, Mechanics of Fiber and Textile Reinforced Cement Composites. CRC Press,
2011. doi: 10.1201/​b11181
[59] B. Mobasher, “Single yarns in woven textiles: Characterization of geometry and length
effects,” in Mechanics of Fiber and Textile Reinforced Cement Composites, CRC Press,
2011, pp. 83–​96. doi: 10.1201/​b11181-​8
[60] ACI Committee 544, “ACI PRC-​544.7R-​16 Report on design and construction of fiber-​
reinforced precast concrete tunnel segments,” 2016.
[61] ACI Committee 544, “ACI 544. 8R-​16 Report on indirect method to obtain stress-​strain
response of Fiber-​Reinforced Concrete (FRC),” 2016.
290 Handbook of Precast Segmental Tunnel Lining Systems

[62] fib, fib Model Code for Concrete Structures 2010. Weinheim, Germany: Wiley-​VCH Verlag,
2013. doi: 10.1002/​9783433604090
[63] P. Nanakorn, and H. Horii, “Fracture mechanics-​based design method for SFRC tunnel
linings,” Proceedings of the Materials Engineering Conference, vol. 1, no. 1, pp. 819–​828,
1996. doi: 10.2208/​jscej.1996.532_​221
[64] L. Liao, A. de la Fuente, S. Cavalaro, and A. Aguado, “Design of FRC tunnel segments con-
sidering the ductility requirements of the Model Code 2010,” Tunnelling and Underground
Space Technology, vol. 47, pp. 200–​210, 2015. doi: 10.1016/​j.tust.2015.01.006
[65] L. Liao, A. de la Fuente, S. Cavalaro, and A. Aguado, “Design procedure and experimental
study on fibre reinforced concrete segmental rings for vertical shafts,” Materials and Design,
vol. 92, pp. 590–​601, 2016. doi: 10.1016/​j.matdes.2015.12.061
[66] G. A. Plizzari, and G. Tiberti, “Steel fibers as reinforcement for precast tunnel segments,”
Tunnelling and Underground Space Technology, vol. 21, no. 3–​4, pp. 438–​439, May 2006.
doi: 10.1016/​j.tust.2005.12.079
[67] M. L. Nehdi, S. Abbas, and A. M. Soliman, “Exploratory study of ultra-​high performance
fiber reinforced concrete tunnel lining segments with varying steel fiber lengths and dosages,”
Engineering Structures, vol. 101, pp. 733–​742, 2015. doi: 10.1016/​j.engstruct.2015.07.012
[68] CNR-​DT204, “Guidelines for the design, construction and production control of fibre rein-
forced concrete structures,” National Research Council of Italy, 2006.
[69] ASTM C1609 /​C1609M–​12, Standard Test Method for Flexural Performance of Fiber-​
Reinforced Concrete (Using Beam With Third-​Point Loading). 2012.
[70] B. Mobasher, Mechanical Testing and Characteristic Responses. CRC Press, 2011.
doi: 10.1201/​b11181-​8.
[71] Y. Yao, M. Bakhshi, V. Nasri, and B. Mobasher, “Interaction diagrams for design of hybrid
fiber-​reinforced tunnel segments,” Materials and Structures/​ Materiaux et Constructions,
vol. 51, no. 1, 2018. doi: 10.1617/​s11527-​018-​1159-​2
[72] DIN EN 14651, “Test method for metallic fibre concrete–​Measuring the flexural tensile
strength (limit of proportionality (LOP), residual),” 2007.
[73] M. Briffaut, F. Benboudjema, and L. D’Aloia, “Effect of fibres on early age cracking of con-
crete tunnel lining. Part II: Numerical simulations,” Tunnelling and Underground Space
Technology, vol. 59, pp. 221–​229, Oct. 2016. doi: 10.1016/​j.tust.2016.08.001
[74] D. Patel, C. Pleesudjai, Y. Yao, S. Schaef, and B. Mobasher, “Validation testing of precast
segments using polymeric fibers,” no. September, 2020.
[75] C. Pleesudjai, D. Patel, M. Bakhshi, V. Nasri, and B. Mobasher, “Applications of statistical
process control in the evaluation of QC test data for residual strength of FRC samples of
tunnel lining segments,” 2021, pp. 815–​826. doi: 10.1007/​978-​3-​030-​58482-​5_​72.
[76] C. Pleesudjai, D. Patel, M. Bakhshi, V. Nasri, and B. Mobasher, “Applications of statistical
process control in the evaluation of QC test data for residual strength of FRC samples of tun-
nel lining segments,” in RILEM-​fib International Symposium on Fibre Reinforced Concrete,
pp. 815–​826, 2020.
[77] J. Dransfield, “Admixtures for concrete, mortar and grout,” in Advanced Concrete
Technology, Elsevier, 2003, pp. 3–​36. doi: 10.1016/​B978-​075065686-​3/​50280-​9
[78] C. F. Ferraris, “Measurement of the rheological properties of high performance con-
crete: State of the art report,” Journal of Research of the National Institute of Standards and
Technology, vol. 104, no. 5, pp. 461, Sep. 1999. doi: 10.6028/​jres.104.028
[79] Walker, H. N., Lane, D. S., and Stutzman, P. E., “Petrographic methods of examining hard-
ened concrete: A petrographic manual,” 2004.
[80] M. Richardson, “Degradation of concrete in cold weather conditions,” Durability
of Concrete and Cement Composites, Elsevier, 2007, pp. 282–​ 315. doi: 10.1533/​
9781845693398.282
[81] M. Alexander, A. Bentur, and S. Mindess, Durability of Concrete. Boca Raton: CRC
Press, [2017] | Series: Modern concrete technology series: CRC Press, 2017. doi: 10.1201/​
9781315118413
 Precast concrete technologies 291

[82] M. Alexander, and H. Beushausen, “Durability, service life prediction, and modelling
for reinforced concrete structures–​review and critique,” Cement and Concrete Research,
vol. 122, pp. 17–​29, Aug. 2019. doi: 10.1016/​j.cemconres.2019.04.018
[83] E. 206-​1, Concrete-​Specification, Performance, Production and Conformity. 2013.
[84] H. Beushausen, and L. Fernandez Luco, Eds., Performance-​Based Specifications and Control
of Concrete Durability, vol. 18. Dordrecht: Springer Netherlands, 2016. doi: 10.1007/​
978-​94-​017-​7309-​6
[85] J. C. Walraven, “Design for service life: How should it be implemented in future codes,” in
Concrete Repair, Rehabilitation and Retrofitting II, II., P. M. Mark G. Alexander, Hans-​Dieter
Beushausen, Frank Dehn, Ed. CRC Press, 2008, pp. 21–​28. doi: 10.1201/​9781439828403-​5
[86] M. Alexander, A. Bentur, and S. Mindess, Durability of Concrete. Boca Raton : CRC
Press, [2017] | Series: Modern concrete technology series: CRC Press, 2017. doi: 10.1201/​
9781315118413
[87] R. Tixier, and B. Mobasher, “Modeling of damage in cement-​based materials subjected to
external sulfate attack. I: Formulation,” Journal of Materials in Civil Engineering, vol. 15,
no. 4, pp. 305–​313, Aug. 2003. doi: 10.1061/​(ASCE)0899-​1561(2003)15:4(305)
[88] A. Bonakdar, and B. Mobasher, “Multi-​parameter study of external sulfate attack in blended
cement materials,” Construction and Building Materials, vol. 24, no. 1, pp. 61–​70, Jan.
2010. doi: 10.1016/​j.conbuildmat.2009.08.009
[89] H. Hilsdorf, and J. Kropp, Eds., Performance Criteria for Concrete Durability. CRC Press,
1995. doi: 10.1201/​9781482271522
[90] H. Hilsdorf, and J. Kropp, Eds., Performance Criteria for Concrete Durability. CRC Press,
1995. doi: 10.1201/​9781482271522
[91] U. M. Angst, “Predicting the time to corrosion initiation in reinforced concrete structures
exposed to chlorides,” Cement and Concrete Research, vol. 115, pp. 559–​567, Jan. 2019.
doi: 10.1016/​j.cemconres.2018.08.007
[92] D. Krajcinovic, “Damage mechanics,” Mechanics of Materials, vol. 8, no. 2–​3, pp. 117–​
197, 1989.
[93] D. Krajcinovic, M. Basista, K. Mallick, and D. Sumarac, “Chemo-​micromechanics of brittle
solids,” Journal of the Mechanics and Physics of Solids, vol. 40, no. 5, pp. 965–​990, Jul.
1992. doi: 10.1016/​0022-​5096(92)90058-​A
[94] R. Tixier, and B. Mobasher, “Modeling of damage in cement-​ based materials sub-
jected to external sulfate attack. II: Comparison with experiments,” Journal of Materials
in Civil Engineering, vol. 15, no. 4, pp. 314–​ 322, 2003. doi: 10.1061/​(ASCE)0899-​
1561(2003)15:4(314)
[95] J. Bentz, D. Ehlen, M. Ferraris, and C. Winpigler, “Service life prediction based on sorptivity
for highway concrete exposed to sulfate attack and freeze-​thaw conditions,” 2002.
[96] A. Bonakdar, B. Mobasher, and N. Chawla, “Diffusivity and micro-​hardness of blended
cement materials exposed to external sulfate attack,” Cement and Concrete Composites, vol.
34, no. 1, pp. 76–​85, 2012. doi: 10.1016/​j.cemconcomp.2011.08.016
[97] ASTM C 1556, ASTM C 1556. Standard test method for determining the apparent chloride
diffusion coefficient of cementitious mixtures by bulk diffusion. 2004.
[98] E. J. Garboczi, and D. P. Bentz, “Computer simulation of the diffusivity of cement-​based
materials,” Journal of Materials Science, vol. 27, no. 8, pp. 2083–​2092, 1992. doi: 10.1007/​
BF01117921
[99] H. H. Powers, and T. C. Steinour, “An interpretation of some published researches on the
alkali-​aggregate reaction Part 1-​The chemical reactions and mechanism of expansion,” ACI
Journal Proceedings, vol. 51, no. 2, pp. 497–​516, 1955.
[100] A. C1567-​21, ASTM C1567-​21,Standard Test Method for Determining the Potential Alkali-​
Silica Reactivity of Combinations of Cementitious Materials and Aggregate (Accelerated
Mortar-​Bar Method). USA, 2021.
292 Handbook of Precast Segmental Tunnel Lining Systems

[101] A. Bonakdar, B. Mobasher, S. K. Dey, and R. Della M, “Correlation of reaction products and
expansion potential in alkali-​silica reaction for blended cement materials,” ACI Materials
Journal, vol. 107, no. 4, pp. 380–​386, 2010.
[102] S. Poyet et al., “Chemical modelling of Alkali Silica reaction: Influence of the reactive aggre-
gate size distribution,” Materials and Structures, vol. 40, no. 2, pp. 229–​239, Jan. 2007.
doi: 10.1617/​s11527-​006-​9139-​3
[103] A. Nielsen, F. Gottfredsen, and F. Thøgersen, “Development of stresses in concrete structures
with alkali-​silica reactions,” Materials and Structures, vol. 26, no. 3, pp. 152–​158, Apr.
1993. doi: 10.1007/​BF02472932
[104] R. Esposito, and M. A. N. Hendriks, “Literature review of modelling approaches for ASR
in concrete: A new perspective,” European Journal of Environmental and Civil Engineering,
vol. 23, no. 11, pp. 1311–​1331, Nov. 2019. doi: 10.1080/​19648189.2017.1347068
[105] P. Gambarova, B. Chiaia, A. P. Fantilli, C. Oggieri, C. Ronco, F. Minelli, G. Plizzari, G.
Tiberti, M. Aiello, and E. Vasanelli, Construction Methodologies and Structural Performance
of Tunnel Linings. 2009.
Chapter 5

Fiber reinforcement in precast concrete

segments
Bill Geers and Benoit De Rivaz

5.1 INTRODUCTION –​ FIBER REINFORCEMENT IN PRECAST

CONCRETE SEGMENTS
The Evolution of Fiber Reinforced Concrete (FRC)
The use of fibers as reinforcement has a history that spans thousands of years. The ancient
Egyptians used straw to reinforce mud bricks for use in the core walls of the pyramids.
During the first century AD, the Romans in the construction of the Coliseum incorporated
fibers in the form of horsehair to help prevent drying shrinkage cracking of their concrete.
As far back as 1911, an idea was patented to incorporate steel shavings into a concrete mix
to increase its strength and stability. However, it was not until the 1960s that the modern
development of steel fiber reinforced concrete (SFRC) began when it was realized that the
cracking tensile strength of concrete (one of its main drawbacks) could be significantly
increased by adding steel fibers. Employed initially for applications such as industrial floor-
ing, in the 1980s SFRC began to be used for underground applications, initially using shot-
crete (sprayed concrete or gunite) and then later concrete tunnel lining segments.
Reinforced concrete tunnels linings have traditionally been cast in-​place, but the develop-
ment of tunnel boring machines led to the innovation of precast concrete segmental lining
technology, which is one of the most promising applications of SFRC. The first use of fibers
to reinforce precast concrete segments in a tunnel was in 1982 for the Metrosud Subway
line in Naples, Italy, where steel fibers were the sole reinforcement in the segments. In the
last two decades, a large number of reference projects where the segments were reinforced
with either “fibers only”, or with a combination of fibers and conventional bar “combined
solution” have been successfully completed. The reported benefits of using SFRC in precast
segments include the following:

• Significantly improved post-​cracking behavior and crack control characteristics

• A more robust segment with improved handling and toughness characteristics
• Structural performance in service and ultimate limit state (SLS and ULS) reliably
achieved
• Cost and time saving in production
• Reduced damage of segments in handling and installation
• Impact resistance of SFRC up to 20 times higher than that for unreinforced concrete
• Sustainability advantages
• Superior durability and longer service life of completed tunnel
• Reduced carbon footprint

DOI: 10.1201/9781032675541-5
294 Handbook of Precast Segmental Tunnel Lining Systems

Over the last 20 years, the increased use around the world of SFRC in precast concrete seg-
ments was driven by many research studies and testing programs performed by industry,
owners, contractors and designers to gain its acceptance for use in specific projects. Testing
standards and protocols developed primarily in Europe and the USA over time were refined
to provide accurate and relevant fiber performance characterization to provide confidence
in the product’s use. These studies and test programs, as well as the many successful inter-
national project experiences, provided a continual better understanding of the performance
behavior of SFRC. This led to the development of valid constitutive laws for design with
SFRC, again primarily in Europe and the USA. The achieved body of knowledge resulted in
development of international codes, guidelines and standards, which have been published
on the design of fibers for use in structural reinforced concrete, such as the fib Model Code
2010. More recently design guide documents have been published on the specific use of
SRRC in precast tunnel segment, such as American Concrete Institute, ACI 544.7R-​16; fib
(fib =​Fédération Internationale du Béton) Bulletin 83; and in 2020, ACI 533.5R-​20. SFRC
used as “fiber only” or in a “combined solution” is now the standard of practice of the con-
crete in precast segments around the globe. Today, fib Bulletin 83, based on the 2010 Model
Code and ACI 544.7R-​16 are the most widely referenced documents for the design of SFRC
used in precast tunnel segments in the world.

5.1.1 Fibers for concrete

There are many different fiber types that are used in concrete for the various benefits they can
provide in a specific application. Fiber types include steel fibers, glass fibers, synthetic fibers
and natural fibers. Fibers are further classified as either macro-​or micro-​fibers. Fibers with
an equivalent diameter greater than 0.3mm (0.01”) are macro-​fibers and equivalent diameter
less than or equal to 0.3 mm (0.01”) are considered micro-​fibers. Choosing the appropriate
fiber for each specific application can be a confusing task. Please refer to Figure 5.1.

Figure 5.1 Various types of fibers in use today.

Note: Typical fibers for concrete reinforcement available in the market.
 Fiber reinforcement in precast concrete segments 295

However, for use in precast concrete segments, choosing appropriate fibers is relatively
straight forward. The three types of fibers that have been used in segments boil down to the
following:

• Steel fibers        Cold drawn wire type (State of Practice)

• Macro-​synthetic fibers Limited use in segments
• Micro-​synthetic fibers Passive fire protection; see Section 5.4

5.1.1.1 Steel vs. macro-​synthetic fibers

Macro-​synthetic fibers were developed by 3M in the USA in 1989, and this technology has
found use in applications, including slabs-​on-​ground, shotcrete and tunnel linings. The first
use of a 100% macro-​synthetic fiber reinforced concrete (MSFRC) segmental lining in the
USA was in 2019. The Blacksnake Creek storm water run-​off tunnel in St. Joseph, Missouri,
United States, is a 2.74m (108”) diameter and 2 km (6,648 LF) long tunnel lined with precast
concrete segments. This first use of MSFRC in a precast segment project was a milestone for
this type fiber. However, the large-​scale acceptance and use of MSFRC to replace SFRC state
of practice use in segmental linings is not eminent. To better understand why, let us look at
the structural performance requirements for a reinforced concrete precast segment.
A fiber concrete is a composite material made-​up of a cement mortar reinforced with a
matrix of fibers. In a fiber concrete, the fibers spread the strain across the cracks created in
the matrix. In other words, the fibers are only engaged if there are cracks in the concrete. No
cracks, no effect of the fibers. Cracks, however, can appear at different stages in the life of the
material. From the first moments, just after pouring the concrete, up to a very advanced age.
Once the concrete tensile strength is exceeded and then cracks to be considered as structural
reinforcement the fiber must provide additional tensile and flexural strength in the matrix.
The performance characteristics of a fiber reinforced concrete is dependent on the mech-
anical properties of the fiber type and the interaction of the materials within the concrete
matrix. Understanding the fundamental differences in mechanical properties of a steel vs. a
synthetic fiber, the post-​crack performance differences can be better understood. Table 5.1
provides Young’s modulus and tensile strengths of the concrete vs. the two types of fibers,
as well as the melting temperatures and creep temperatures of the fibers.
Comparing Young’s modulus of a steel vs. a synthetic fiber compared to the modulus of
concrete exhibit the fundamental differences in how the composites behave under tension.
The modulus of elasticity of hardened concrete is two to three times greater than the elastic
modulus of a synthetic fiber. What does this mean? While the concrete is in a plastic state,
synthetic fibers will provide plastic shrinkage crack control. However, as the concrete hard-
ens and its elastic modulus value exceeds the synthetic material’s modulus, the synthetic
fiber will not provide any tensile strength benefit until after the concrete cracks, and then
expand to a point where the synthetic fiber stretch sufficiently to transfer tensile strength to
the concrete matrix. Macro-​synthetic fibers require large crack widths before they develop
any useful stress in tension.
Steel fibers oppositely have a modulus of elasticity that is seven plus times greater than
the concrete. Steel fibers, unlike synthetics, do not provide any plastic shrinkage crack con-
trol. However, as soon as the concrete hardens, and cracks begin to form in the matrix, the
steel fibers immediately kick in and begin to absorb and then redistribute the tensile forces
within the matrix. Steel fibers arrest cracks as they begin to form and keep them from open-
ing further.
296 Handbook of Precast Segmental Tunnel Lining Systems

Table 5.1 Comparison of material properties.

Comparison of material properties between hardened concrete, steel fibers and
synthetic fibers

Hardened concrete Steel fibers Synthetic fibers

Item Values Metric Imperial Metric Imperial Metric Imperial
1 Young’s modulus 30 GPa 4,350 ksi 210 GPa 30,450 ksi 12 GPa 1,740 ksi
(up to)
2 Tensile strength 4 MPa 580 psi 2.3 GPa 330 ksi 0.65 GPa 96 ksi
(up to)
3 Material melting N/​A N/​A 1,500°C 2,732°F 165°C 329°F
point
4 Material creep N/​A N/​A 370°C 698°F >20°C >68°F
temperature

Controlling crack widths in tunnel linings by limiting the crack widths during design
and construction is critical. Wide crack widths can adversely impact the serviceability and
long-​term durability of the tunnel. Cracks in the lining are a major cause of leaks in a tun-
nel, both infiltration and exfiltration. Because steel fibers are distributed throughout the
entire cross-​section, steel fibers provide for reduced crack widths compared to convention-
ally reinforced concrete. The ability to control and reduce the width of cracks in a precast
segmental lining is a fundamental difference between a steel and synthetic fiber; steel fibers
are the preferred choice, including over conventional reinforcement. Section 5.5.3 of this
chapter will discuss durability of SFRC and the overall superior performance of it compared
to conventional reinforced concrete.
Comparing the melting and creep temperatures of a steel vs. a synthetic fiber explains the
fundamental differences in how the composite behaves under exposure to heat and fire and
under sustained loading. Section 5.4 of this chapter will discuss the use of micro-​synthetic
fibers for passive fire protection in the event of a fire in the tunnel. In this application the
fibers melt creating pathways to alleviate the pressure build-​up from the pore water in the
concrete turning to steam. However, because of the low melting temperature and creep
temperature of synthetic fibers it is not appropriate to use macro-​synthetics as a structural
reinforcement in any application subject to elevated temperatures or fire or where creep is
a concern. Table 5.2 provides a summary of the appropriate fiber types for the different
applications.
The differences in fiber types and properties will have a significant impact on the behavior
observed in different FRC characterization tests and in real-​life ground support systems. In
fact, in hyperstatic systems such as those found in ground support, an effective control of
the crack widths will promote a more efficient multi-​cracking process and, therefore, a more
ductile behavior of the structure, most probably due to the higher degree of hyper-​staticity
that is maintained.

5.1.1.2 Steel fibers for use in precast segments

Steel fiber reinforced concrete (SFRC) using fibers made from high tensile strength cold
drawn steel wire is considered standard of practice in precast segments today. Steel fibers
have been proven suitable for use in all type precast segmental applications either as “fiber
 Fiber reinforcement in precast concrete segments 297

Table 5.2 Appropriate fiber usages.

Steel, macro-​ and micro-​s ynthetic fibers –​ showing which one to use or not use in a
specific application

Fiber material type

Item Applications Steel Macro-​s ynthetic Micro-​s ynthetic
1 Fire protection No No Yes
2 Plastic shrinkage reinforcement No No Yes
3 Impact resistance Heavy Light No
4 Load bearing –​ SLS Yes No No
(crack width control)
5 Load bearing –​ ULS Structural Temporary (creep) No
6 Fatigue resistance Yes No No

Figure 5.2 A typical precast tunnel segmental lining.

Note: Exploded isometric view of the segments comprising a ring.

only” or as a “combined solution” and a better alternative to conventional rebar or wire

mesh. See Figure 5.2.
There are various types of steel fibers that are available in the market. ACI references
the ASTM A820/​A820M standard specification for steel fibers for use in concrete. ASTM
A820/​A820M provides classification for five general types of steel fibers, based primarily
on the product or process used in their manufacture: type I: cold-​drawn wire; type II: cut
sheet; type III: melt-​extracted; type IV: mill cut; and type V: modified cold-​drawn wire.
However, not all of the ASTM types of steel fibers are appropriate for use in this applica-
tion. The appropriate ASTM A820 fibers for use in precast segments are manufactured from
cold drawn wire, corresponding to a type I, or type V classification. The corresponding
appropriate European (BS-​EN) and International Standards (ISO) are the BS-​EN 14889-​1
standard with a CE marking System 1 (steel fiber for structural use) or the ISO 13270 (class
A) as a minimum. The fibers in the appropriate dosage are homogenously mixed into the
concrete and then provide a network of individual pieces that are oriented in multiple direc-
tions and reinforce the entire cross-​section of the segment. The performance of a steel fiber
298 Handbook of Precast Segmental Tunnel Lining Systems

Figure 5.3 Steel fiber attributes effecting performance.

Note: Illustration of glued steel fibers showing the key performance attributes.

in concrete is dependent on three primary attributes of the fiber: aspect ratio (L/​D), wire
tensile strength and anchorage mechanism as shown in Figure 5.3.
The following are the recommendations for steel fibers used in precast segments:

• Fibers out of drawn wire, with a tensile strength of steel wire >1,800 MPa (260 ksi)
minimum
• Precast segments typically have compressive strengths well in excess of the min-
imum that is required for their structural design: 35 to 40 MPa (5,000 to 6,000
psi)
• Because of high early strength requirements for production efficiency in the seg-
ment manufacturing plant, higher compressive strengths in the 65 to 70 MPa
(9,000 to 10,000 psi) are common
• To maintain ductility of the SFRC, higher tensile strength wire is recommended
• Hooked ends that have been optimized to ensure proper anchorage in the concrete
matrix
• Fiber length: in the range of 50 mm (2”) to 60 mm (2.4”)
• Aspect ratio (length/​diameter): a minimum of 65; 80 is recommended
• Minimum fiber length: 2.5 times the maximum coarse aggregate size
• Fibers glued in clips to promote homogenous distribution and eliminate balling of
fibers
• Introduction of fibers into mix using a fully integrated automatic dosing system

Dimensional tolerances are shown in Table 5.3

5.1.2 Characteristics of steel fiber reinforced concrete

SFRC is concrete with small diameter, large quantity of relatively short discrete pieces of
steel wire that are closely spaced, randomly orientated and homogenously mixed throughout
the concrete matrix. The fibers absorb tensile forces in all directions throughout the entire
cross-​section of the segment vs. conventional reinforcing, which provides strength in distinct
locations acting in one direction or plane. Because of the forces that a segment needs to
resist, from SFRC uniquely suited to this application because of the many benefits it provides
 Fiber reinforcement in precast concrete segments 299

Table 5.3 Steel fiber dimensional tolerances.

Requirements are per BS EN 14889-​1 :2006 –​ fibers for concrete. Steel fibers

Industry symbols, values and tolerances

Deviation of the Deviation of the
individual value average value
relative to the relative to the
Item Material property Symbol declared value declared value
1 Length and developed height l, l d +​/​–​1 0% +​/​–​5 %
>30 mm (1.2”) (if applicable) +​/​–​1 5 mm
≤30 mm (1.2”)
2 Equivalent diameter d +​/​–​1 0% +​/​–​5 %
>0.30 mm (0.01”) +​/​–​0 .015 mm
≤0.30 mm (0.01”)
3 Length/​d iameter ratio λ +​/​–​1 5% +​/​–​7 .5%
Notes:
1
The length shall be measured with a marking gauge (accuracy of 0.1 mm) (0.0039”). In an irregular cross-​
section, the developed length of the fiber shall also be determined to calculate the equivalent diameter.
2
Straightening of the fiber is necessary, it shall be done by hand or, if this is cnot possible, by hammering on a
level of wood, plastic material or copper using a hammer or a similar tool. While straightening, the cross-​s ection
must stay unchanged.
3
The diameter of the f iber shall be measured with a micrometer, in two directions, approximately at right
angles, to an accuracy of 0.01 mm (0.00039”). The f iber diameter shall be the mean of the two measured
diameters.

over conventionally reinforced concrete as shown in Figure 5.4. “Fiber only” SFRC is suit­
able for most tunnel linings but it has some performance limitations. Conventional reinfor-
cing steel can be added to provide a “combined solution” in areas with large, concentrated
bending moments or large isolated tensile forces in the concrete, to absorb these higher or
isolated stresses.
The SFRC mix design is determined by the same principles which apply to concrete mix
design. The prime factors controlling strength and quality are the water/​cementitious mate-
rials ratio, the grading of the aggregates and the degree of consolidation achieved. However,
there are a number of design considerations in which SFRC differs from conventional
structurally reinforced concrete. The major differences lie in the aggregate grading and the
cementitious content of the concrete design. The concrete composition must be engineered
in order to obtain homogeneous distribution of the steel fibers and good finishing ability.
The performance of SFRC is determined primarily by the following characteristics of the
material:

• Performance of the specific fiber in the specific concrete mix are related to the
following:
• Fiber tensile strength
• Aspect ratio (L/​D)
• Anchorage mechanism
• Fiber distribution
• Concrete composition (both freshly mixed and the hardened matrix properties)
• Concrete compressive strength
• Quantity of fibers (dosage)
300 Handbook of Precast Segmental Tunnel Lining Systems

Figure 5.4 Benefits of steel fiber reinforced concrete precast segmental tunnel linings.
Note: Benefits diagram of the advantages of the use of steel fiber in precast tunnel segments.

To ensure a sufficient distribution network of fibers is achieved, Table 5.4 provides a “sug­
gested” minimum dosages of fibers to meet spacing requirements based on fiber aspect ratio
and application. This minimum is applicable if dosage to meet the structural performance
requirements demonstrated by beam testing is less than these values. The minimum fiber
aspect ratio for segments is 65, and 80 is considered optimal. Any fibers over a 65 aspect
ratio must be provided glued in clips to prevent fiber balling and, therefore, restricting
proper distribution in the concrete.
SFRC was first used in precast segments for the modest level of post first crack flex-
ural capacity that it provides and the savings in costs due to efficiency and cost savings in
material and labor in the factory. As experience and SFRC technology has grown, other
very significant benefits that SFRC can provide in segments require further assessment of
the materials performance to assure realizing these added benefits. Therefore, when deter-
mining the appropriate dosage for a particular fiber, meeting a prescribed minimum flexural
performance is only one consideration.
SFRC with the appropriate performance characteristics, i.e. flexural hardening, provides
an exceptionally tough and durable segment to withstand high impact loading and provide
crack widths a fraction of that of conventional only reinforced segment. Eliminating dam-
age during handling and installation and providing a maximum crack width well within
or below allowable values specified by the codes, guidelines and recommendations in a
range of 0.15 to 0.3 mm (0.006” to 0.012”) are essential considerations. By increasing the
performance of SFRC to a level providing hardening characteristics the added benefits of
superior resistance are realized.
When characterizing SFRC performance in flexure, there are three basic material behav-
iors. Elastic and deflection softening, and deformation hardening plastic behaviors in bend-
ing. By definition, bending softening behavior in flexure is where the load capacity after first
crack is less than the load at first crack, and hardening is characterized as the load carrying
 Fiber reinforcement in precast concrete segments 301

Table 5.4 Recommended minimum dosages of steel fibers for “fibers only” or “combined
solution” design.
Minimum dosages in kg/​m 3 of steel fibers based on different aspect ratios and overlap
factors to achieve a sufficient network based on the McKee spacing theory

Item Description Recommended minimum dosages of steel fibers (kg/​m 3)

Aspect ratio (l f/​d f =​ y) 50 55 60 65 80 Application
1 Overlap factor =​ 1.66 >50 48 40 35 22 Combined solution
2 Overlap factor =​ 1.80 >70 60 50 43 30 Fibers only

Notes:
1
The steel fibers dosages recommended above are not fixed.
2
The design engineer may at any time choose to use a different steel fiber spacing which ensures a different
degree of network.
3
For precast segment linings, recommendation is to use an overlap factor =​ 1.66 for combined solutions with
conventional rebar and an overlap factor =​ 1.8 to 2.0 for applications 100% steel fiber reinforcement (structural
applications). The consideration of steel fiber spacing requirements ensures that there will be improvements in
concrete crack widths and impact resistance compared to wire mesh reinforce concrete.
4
Final decision on which dosage to use will be determined by residual strength minimum.

capacity of the SFRC segment in flexure after some deflections is greater than the first crack
in the post-​crack zone. As we will discuss in the following sections, fiber performance in the
small scale is typically characterized by performing flexural testing on beams and measuring
the load deflection response of the material. Figure 5.5 shows graphically the load deflection
response of SFRC exhibiting hardening vs. softening behavior.
Most SFRC in the past that contained approximately 0.5% by volume of the steel fibers,
typically provided flexural softening behavior. Today, with the advancements in fiber and
concrete technology, obtaining deflection hardening response in flexure is achieved using
the same volume of the appropriate fibers in an optimized concrete mixture. Higher per-
formance levels of SFRC provide more applications for “fiber only” reinforcement solutions
in segments. The SFRC for segments, “fiber only” or “combined solution”, to optimize
the performance requires more considerations than just structural capacity of the segment.
Tunnels being built today typically require 100 to 120 plus year design service life to be
demonstrated. Properly designed higher performance SFRC in segments is a powerful tool
in achieving a robust, durable and watertight tunnel lining with a long service life as illus-
trated in Figure 5.6.

5.2 PRECONSTRUCTION TESTING
Prior to any production of segments on a project testing is performed to determine the
appropriate fiber dosage and optimization of the concrete mix design to assure that
the SFRC to be used meets all of the requirements of the design. In general, most of the
302 Handbook of Precast Segmental Tunnel Lining Systems

Figure 5.5 Steel fiber reinforced concrete –​ softening vs. hardening behavior.

Note: Classification of load and deflection performances for fiber reinforced concrete.

Figure 5.6 High-​p erformance SFRC hardening behavior for segments (red circles).
Note: Characterization of softening vs. hardening performance of fiber reinforced concrete.

engineering properties of SFRC are primarily related to its concrete matrix properties and
thus test methods developed for plain concrete in both fresh and hardened state can be used
in the case of SFRC. These include the conventional concrete tests at early-​age to char-
acterize workability, plastic shrinkage and, at later ages once it is hardened, tests such as
compression, direct tensile (splitting) and modulus of elasticity. The significant difference in
the performance of SFRC vs. plain concrete is its post-​crack tensile performance of SFRC
 Fiber reinforcement in precast concrete segments 303

that marks the significant difference of FRC’s behavior in relation to that of plain concrete.
The parameters that impact the post-​cracking tensile strength performance of SFRC were
discussed in Section 5.1.2. In the design of the segmental SFRC tunnel linings, the relevant
performance characteristic to be quantified is the tensile stress–​strain behavior in its hard-
ened state.
The post-​crack tensile properties of SFRC can be determined by the following types
of tests:

• Direct uniaxial tension test

• This type of test is very difficult to perform and is not being used in testing of
SFRC in segments.
• Indirect “special” tension test
• Double punch test (also known as the “Barcelona indirect tensile strength test”)
is an example. This type of testing is very specialized, not currently validated by
international standards organizations. Not to be used as stand-​alone in precon-
struction testing, however, they do have potential for use for quality control test-
ing during segment production.
• Small-​scale flexural testing of beams
• This type testing is the most common and accepted type testing to determine the
post-​crack tensile strength in flexure. They involve beams tested in bending accord-
ing to ASTM or EN standards. The beam bending tests include three-​point or
four-​point (or “third point”) tests at the beam aspect ratio (length: depth) greater
than 3:1. The main advantage of these types of tests is that they are well estab-
lished, relatively easy to carry-​out and there is an extensive history of their use.
  There are two basic types of beams with specific loading configurations that are
used for characterization of the FRC. Two types are “notched beam” [EN 14651,
Réunion Internationale des Laboratoires et Experts des Matériaux, systèmes de
construction et ouvrages (RILEM) and un-​notched beam (ASTM C1609)]. These
two types of tests will be discussed in detail later in this chapter.
• Full-​scale testing of a segment
• Bending tests are performed on an individual segment in order to check the bear-
ing capacity under flexure, which occurs during demolding, handling, storage,
transport, installation and in place (primarily due to unsymmetrical earth pres-
sures). Also, point-​load tests simulating the TBM ram thrust forces are performed
to determine the ability of the SFRC segment to resist these stresses.
  ACI 544.7R, “Design and Construction of Fiber-​Reinforced Precast Concrete
Tunnel Segments” and fib Bulletin 83, “Precast Tunnel Segments in Fiber-​Reinforced
Concrete” both have guidance and recommendations on performing this testing
to validate material performance and conformance with design methodology and
requirements. This type of testing will also be discussed later in this section.

5.2.1 Testing procedures and performance criteria

Currently, the widely accepted test methods to characterize the behavior of the SFRC in
flexure and determine the direct tensile strength provided by the fibers in the segment is
through flexural testing with beams. The determination of the tensile strength provided
is done indirectly and then used to determine conformance with design requirements. See
Figure 5.7, which provides an overview of the SFRC segment design process using beams for
304 Handbook of Precast Segmental Tunnel Lining Systems

Figure 5.7 Overview steel fiber reinforced concrete design with beam testing.
Note: Design flowchart of the key design stages for fiber reinforced concrete.

material characterization. The two main standardized flexural (beam) test methods being
used around the globe are either the ASTM C1609 or the EN 14651 (same as RILEM beam)
methods. Primary differences between the two test methods are how the load is applied to
the beam [notched beam is single-​center point loaded and un-​notched is loaded at the third
points (two points) of the beam] and the EN beam is notched vs. ASTM is unnotched.
A notch in the bottom of the beam at mid-​span controls the location of the crack when
loaded.
The ASTM standard has been referenced predominately in areas where the ACI standards
and guides are referenced and the EN standard is used predominately where the fib 2010
Model Code is the reference design document. There are perceived advantages and disad-
vantages of both types of tests. The European fib Model Code 2010 prescribes the use of the
notched beam whereas the ACI 544.7R design guide prescribes the ASTM C1609. Because
the results of the two types of tests are markedly different, it is essential that the design
method and test method are consistent. The beam type specified by the designer cannot be
changed without the designer’s confirmation.
As discussed earlier in this chapter, historically SFRC typically exhibits strain softening
characteristics in flexure. Now, thanks to advancements in steel fiber and concrete tech-
nology, SFRC that exhibits post-​crack deflection hardening is readily achievable at reason-
able dosages (0.5% by volume). The benefits in tunnel linings for reduced crack widths,
durability and increased toughness means that SFRC that exhibits hardening in flexure is
the state of the art. In the past, the engineer typically only specified the “average” residual
strength that the FRC provides from the beam test. Today with high performance SFRC,
the engineer should specify minimum fiber performance requirements at first crack, SLS and
ULS load deflection levels. Now engineers are looking at design for serviceability as well as
ultimate strength levels due to this higher level of SFRC performance.
Serviceability in SFRC is a performance characterized by the residual strength at lower
levels in the deflection or crack opening response in the beam (L/​600 deflection in ASTM
 Fiber reinforcement in precast concrete segments 305

C1609 or CMOD1 in EN14651). Ultimate strength performance of the SFRC is determined

by the residual strength at a higher level of deflection or crack opening in the beam test
(L/​150 in ASTM C1609 or CMOD3 in EN14651). Hardening by definition means that
as the load increases after initial cracking the SFRC provides for higher loads and stress
redistribution.
It has been observed that because of the notch in the EN type beam, as the load increases
the location of the crack formation and its propagation is localized at the notch (single
plane). However, in an un-​notched ASTM type beam, as the load increases the initial crack
forms at the weakest plane in the beam, then as the material exhibits hardening character-
istics, additional crack locations may form. Once this was observed as an issue, the fiber
manufacturer performed third party testing to evaluate the ASTM vs. EN beam protocols
in characterizing performance of hardening materials. This testing found that EN 14651
protocol is suitable for characterizing a SFRC material since it reduces the structural effects
on the tests. A four-​point bending test on un-​notched specimen (ASTM C1609) introduces
structural effects (e.g. the monitored section is not defined): these effects can modify the
response of the tests when materials properties change (e.g. moving from softening to hard-
ening behavior). This results in the ASTM C1609 beam excessively under-​estimating the
residual strength at the higher load and deflection as listed below in Table 5.5 for a concrete
mix with 62 MPa compressive strength.
It is recommended that for conformance with design, a notched type beam protocol be
used for SFRC, which exhibits deflection hardening. Designers of SFRC segments in accord-
ance with constitutive laws per ACI 544.7R, where hardening behavior is desired, need
to be aware of this issue. It is recommended that they consider converting performance
requirements of ASTM C1609 test to values applicable to notched three-​point bending
beam (EN 14651) test and use this protocol for preconstruction testing. Typically, when

Table 5.5 Comparing EN 14651 to ASTM C1609 beam test results in a deflection hardening
SFRC.
Test results in hardening steel fiber reinforced concrete (SFRC) show that the
ASTM beam test results are underestimated at higher load/​d eflection values due to
structural effects (multiple cracks forming) in the beam

f r1 vs. f D600 f r3 vs. f D150

Item Testing standard SLS Metric Imperial ULS Metric Imperial
1 EN 14651 –​ test f R1 5.09 736 psi f R3 7.90 1,146
method for metallic fibre Average MPa Average MPa psi
concrete –​ measuring the
flexural tensile strength
[limit of proportionality
(LOP)], residual
2 ASTM C1609 –​ standard f D600 6.02 873 psi f D150 5.41 785 psi
test method for flexural Average MPa Average MPa
performance of fiber-​
reinforced concrete (using
beam with third-​p oint
loading)
3 Ratios f D600 Average /​ f r1 f D150 Average /​ f r3
Average =​ 1.18 Average =​ 0.68
306 Handbook of Precast Segmental Tunnel Lining Systems

correlating the results of an ASTM C1609 beam to EN 14651 beam test, the ASTM L/​600
corresponds to EN fR1 (SLS), and the L/​150 corresponds to the fR3 (ULS) value.
The execution of preconstruction testing using beams should be in a laboratory con-
trolled environment. Regardless of the testing standard used, the casting of the SFRC and
test beams are of critical importance to obtain accuracy and the lowest variation in the
results. The following are the leading steel fiber manufacturers’ recommendations for mak-
ing the preconstruction test beams.
Preferably, the specimens are made under laboratory conditions. In this case, a pan
mixer shall be used to mix the concrete, and all the ingredients must be weighed accur-
ately. Aggregates, sand and cement are added to the mixer (in this order) and mixed for 60
seconds. Then the water is added, and the concrete is mixed for another 60 seconds. After
this step, the fibers are added, and mixed for 270 seconds to make sure that all fibers are
separated and homogeneously distributed. The mixer can be stopped now, and a visual
inspection of the distribution of the fibers needs to be performed. Filling the molds is done
in one lift, up to a height of 110% of the mold. The concrete is then vibrated on a vibrating
table (unless it is self-​compacting concrete) and levelled off during vibration.
Full compaction is achieved when there is no further appearance of large air bubbles on
the concrete surface, and the surface is becoming relatively smooth with a glazed appear-
ance, without excessive segregation. Internal vibration is not recommended, in order not to
disturb the three-​dimensional homogeneous distribution of the fibers. If a vibration needle
is used, the zone around the vibration needle will contain less fibers than the rest of the
beam. The specimens should be left in the molds for at least 16 hours, but no longer than
three days, protected against shocks, vibration and dehydration. The specimens are cured
at a temperature of 20±5°C (68±9°F) for a minimum of 3 days, either in water at a tem-
perature of 20±2°C (68±4°F), or in a chamber at 20±2°C (68±4°F) and a relative humidity
of minimum 95%. Regular checks should be made that the surfaces of the specimens in the
chamber are continuously wet. Loss of moisture and deviations from the required tempera-
ture should be avoided.
Bending tests must be performed on a minimum number of specimens and the variation of
the results in terms of residual strength at SLS value, must not exceed the following values:

• Characterization tests: 12 each, residual strength variation ≤25%

• Control tests: 9 each, test beams and residual strength variation ≤25%

The values derived from the suggested 12 minimum beam characterization tests are not
“mean” values but need to be expressed as “characteristic” values (denoted as a subscript
“k”–​in the EN 14651 beam testing values; fR1k, fR2k, fR3k, etc.). Figure 5.8 below listed the
information from the fib Model Code 2010 to calculate characteristic values.

5.2.1.1 ASTM C1609 –​ standard test method for flexural performance

of fiber-​reinforced concrete
ASTM C1609 testing standard is most widely used in North America and anywhere the
American Concrete Institute’s codes and design guides are used. The test method evaluates
the flexural performance of fiber-​reinforced concrete using parameters gained from the
load-​deflection curve obtained by testing a simply supported beam under third-​point load-
ing using a closed-​loop, servo-​controlled test set-​up. The standard beam dimensions used
for SFRC precast segments 150 × 150 × 500 mm (6” × 6” × 20”) tested on a 450 mm (18”)
 Fiber reinforcement in precast concrete segments 307

Figure 5.8 Characteristic value computation.

Note: Characteristic value computation from beam test values. S x denotes the standard deviation.

Figure 5.9 ASTM C1609 four-​p oint beam bending test –​ general arrangements.
Note: Typical arrangement and calculations for residual strengths of FRC.

span as shown in Figure 5.9. The design method in ACI 544.7R, “Design and Construction
of Fiber-​Reinforced Precast Concrete Tunnel Segments” guide is based on the results from
this testing method. This method provides reliable and accurate results for SFRC exhibiting
flexural softening or elastic characteristics.
For SFRC that exhibits flexural hardening, the residual strength results at higher deflec-
tions have been found to have wide variations and to significantly understate the residual
strength values in the beam. ASTM currently has a work item in process (ASTM-​WK68392)
which states in its rationale, “As more designers are using crack width as a design param-
eter for FRC, it is essential and useful to include a notched beam version as an alterna-
tive in ASTM C1609”. Until a notched alternative is available in ASTM, the EN 14651
beam protocol should be used, requiring the engineer’s approval and needing to provide the
applicable performance requirements for using this alternative method.
308 Handbook of Precast Segmental Tunnel Lining Systems

Figure 5.10 ASTM C1609 beam testing apparatus.

Note: Typical laboratory load testing set-​u p equipment for beam team.

The loading is performed on a closed-​loop servo-​controlled testing system with roller

supports that are free to rotate about their axis. The testing machine must be capable of
measuring the net deflection of the center of the beam and control the rate of increase of
deflection. Testing machines that are stroke displacement or load controlled are not appro-
priate for this testing method. The rate of increase in net deflection up to a net deflection of
L/​900 is at a specified slower rate and then may be increased at a higher rate beyond this
deflection. The determination of first-​peak and peak loads and the corresponding stresses
calculated by inserting them in the formula for modulus of rupture is called for. Residual
strengths are calculated based on loads obtained at specified deflections (L/​600, L/​150).
The ASTM C1609 test parameters used to summarize performance are defined graph-
ically in Figures 5.10 and 5.11. Performance is primarily assessed at three points: the first
peak in the load-​deflection curve, which is used to calculate the modulus of rupture; the
load at 0.75 mm (0.030”) central deflection, which is used for performance assessment
under service conditions for service limit state (SLS); and the load at 3.0 mm (0.12”) cen-
tral deflection, which is used for ultimate strength estimates for ultimate limit state (ULS)
design. These points correspond to the load at first crack and deflections of span/​600 and
span/​150. The standard also calls for determination of specimen toughness based on the
area under the load-​deflection curve up to a prescribed deflection and the corresponding
equivalent flexural strength ratio.

5.2.1.2 EN 14651 –​test method for metallic fiber concrete–​measuring the flexural
tensile strength
EN 14651 is the reference standard for the European Union CE label for steel and polymer
fibers and has been adopted by a number of fiber manufacturers and designers especially
in Europe and the Middle East regions. The tensile behavior of SFRC is evaluated in terms
 Fiber reinforcement in precast concrete segments 309

Figure 5.11 ASTM C1609 four-​p oint beam bending testing parameters.

Note: Typical representation of testing results from the four-​p oint beam tests.

of residual flexural tensile strength values, determined from the load-​crack mouth opening
displacement curve, or load-​deflection curve, obtained by applying a center-​point load
on a simply supported notched beam. Notch dimension is 25 mm long × 5 mm wide (1”
long × 0.25” wide); and beam dimension is 550 × 150 × 150 mm (22” long × 6” thick ×
6” wide). The test results are expressed in terms of the limit of proportionality (LOP) and
the residual flexural strength. Testing of the specimens is done in a three-​point bending test
configuration but can be performed in two ways. In the first method, the crack mouth open-
ing displacement (CMOD) placed at the notch is measured, and a displacement transducer
is mounted along the longitudinal axis at the mid-​width of the test specimen. The distance
between the bottom of the specimen and the line of measurement shall be less than 5 mm
(0.25”) as shown in Figure 5.12.
A second possibility is to measure the deflection instead of the CMOD. In that case, a
displacement transducer shall be mounted on a rigid frame that is fixed to the test spe-
cimen at mid-​height over the supports. One end of the frame should be fixed to the spe-
cimen with a sliding fixture, and the other end with a rotating fixture. A thin plate fixed at
one end can be placed at mid-​width across the notch mouth at the point of measurement.
The tests are preferably deflection controlled. To control the test with a CMOD, knives
need to be glued next to the notch. It is possible that they come loose during the test due
to a bad interlayer between the knives and the concrete. It is easier to mechanically fix the
deflection transducer to the concrete specimens. Less test specimens and test results will
be lost in this way.
The testing machine should be capable of operating in a controlled manner, producing a
constant rate of displacement (CMOD or deflection), and have a sufficient stiffness to avoid
unstable zones in the load–​CMOD curve or the load deflection curve. A total stiffness of
the system of 200 kN/​mm (1,132 lbf/​in) (including frame, load cell, loading device and sup-
ports) is advised. All rollers should be made of steel and have a circular cross-​section with
310 Handbook of Precast Segmental Tunnel Lining Systems

Figure 5.12 EN14651 –​ crack mouth opening displacement (CMOD) displacement measured.

Note: Typical arrangement of three-​p oint notched beam test.

a diameter of 30 mm ±1 mm (1.18” ±0.04”). Two of the rollers, including the upper one,
shall be capable of rotating freely around their axis and of being inclined in a plain perpen-
dicular to the longitudinal axis of the test specimen. The distance between the centers of the
supporting rollers shall be equal to 500 ±2 mm (19.7” ±0.08”). The load measuring device
needs an accuracy of 0.1 kN (0.011 ton) and the linear displacement transducer needs an
accuracy of 0.01 mm (0.0004”). The data recording system should be able to record load
and displacement at a rate not less than 5 Hz.
In the case of a testing machine controlling the rate of CMOD increase, the machine
should operate from the start of the test with a CMOD increase of 0.05 mm/​min (0.002”/​
min) and a data logging minimum of 5 Hz. When CMOD is 0.1 mm (0.004”), the machine
should operate at a CMOD increase of 0.2 mm/​min (0.008”/​min) and a data logging min-
imum of 1 Hz. The test should not be terminated before a CMOD value of 4 mm (0.157”).
In the case of controlling the deflection increase, the machine should start the test with a
deflection increase of 0.08 mm/​min (0.003”/​min) with a data logging minimum of 5 Hz.
When the deflection reaches 0.125 mm (0.005”), the deflection increase should change to
0.21 mm/​min (0.008”/​min) until a final deflection of 3.5 mm (0.139”), and a data logging
minimum of 1 Hz. If the crack starts outside the notch, the test result should be rejected.
Before starting any segment production, compressive and bending tests in accordance
with EN 14651 have to be performed in order to control the fulfilment of the character-
istic values defined in the design. Note that the “k” subscript fLk, fR1k, fR3k denote the use of
“characteristic”, not mean values as discussed in Section 5.2.1.
The test results, which need to be expressed, are the limit of proportionality (LOP) and
the residual flexural strengths as shown in Figure 5.12.
The limit of proportionality f fct,L is calculated as

3 l
fctf ,L = .FL . 2 (5.1)
2 bh

where FL is the maximum load between CMOD of 0.00 and 0.05 mm (0.0020”) or deflec-
tion of 0.00 and 0.08 mm (0.00315”). The residual flexural strength fR,x needs to be evalu-
ated at four different displacements.
 Fiber reinforcement in precast concrete segments 311

3 l
f R, i = .FR,i . 2 (5.2)
2 bh

where FR,i is the residual load at

i =​1: CMOD =​0.5 mm (0.020”) or deflection 0.47 mm (0.018”)
i =​2: CMOD =​1.5 mm (0.060”) or deflection 1.32 mm (0.052”)
i =​3: CMOD =​2.5 mm (0.098”) or deflection 2.17 mm (0.085”)
i =​4: CMOD =​3.5 mm (0.137”) or deflection 3.02 mm (0.119”)

and
l =​the span between the supports; nominal distance 500 mm (19.6”)
b =​the width of the concrete sample; nominal value 150 mm (6”)
h =​the residual height of the concrete sample; nominal value 125 mm (5”)

Value fR1 (CMOD =​0.5 mm) (0.020”) is used for the service limit state (SLS). Value fR3
(CMOD =​2.5 mm) (0.098”) is used for the ultimate limit state (ULS). FRC toughness can
be classified by using a couple of parameters: the first one is a number representing the
SFRC compressive strength while the second one is a letter representing the ratio fR3k/​fR1k.
The strength interval for fR1k is defined by two subsequent numbers in the series: 1.0; 1.5;
2.0; 2.5; 3.0; 4.0; 5.0; 6.0; 7.0; 8.0 MPa (145, 217, 290, 362, 435, 580, 725, 870, 1,015,
1,160 psi) and,
The fR3k /​ fR1k ratio can be represented with letters a, b, c, d, e, corresponding to the ranges:

“a” if 0.5 ≤ fR3k /​ fR1k ≤ 0.7

“b” if 0.7 ≤ fR3k /​ fR1k ≤ 0.9
“c” if 0.9 ≤ fR3k /​ fR1k ≤ 1.1
“d” if 1.1 ≤ fR3k /​ fR1k ≤ 1.3
“e” if 1.3 ≤ fR3k /​ fR1k

The designer should specify the residual strength class and the fR3k /​ fR1k ratio as well as the
type of fiber and its material properties such as fiber strength. Materials with fR1k ranging
from 4.0 MPa to 5.0 MPa (580 to 725 psi) are commonly used for precast tunnel segments

Figure 5.13 EN14651 three-​p oint beam bending testing parameters.

Note: Typical representation of testing results from three-​p oint beam with notch tests.
312 Handbook of Precast Segmental Tunnel Lining Systems

without any bar reinforcement, combined with a fR3k /​ fR1k ratio in the ranges 0.9 < fR3k /​
fR1k < 1.1 or 1.1 < fR3k /​ fR1k < 1.3 (class C or class D, respectively, according to the Model
Code 2010 definition). Note that flexural hardening is indicated by a fR3k /​ fR1k ratio > 1.0.

5.2.2 Full-​scale testing
Proof testing (aka, proof-​of-​design) of precast concrete elements is routinely done, espe-
cially true for repetitive units, where large numbers are required to meet a particular pur-
pose such as with the manufacturing of segments. Load testing can be used to determine the
ability of a structural element to validate performance, to gain knowledge on the behavior
of a structure, and to supplement, validate or refine analytical work models. Both ACI
544.7R and Bulletin 83 recommend full-​scale testing during preconstruction to “prove” the
performance of the segment is adequate.
SFRC precast segments are subjected to various loads; from manufacturing, to installa-
tion, and then in situ conditions. Two different types of full-​scale tests have been performed
on segments in specific projects in Europe to successfully assist in the design process to val-
idate the EN 14651 beam test in characterizing the SFRC material performance in conform-
ance with fib Model Code 2010 methodology and constitutive laws in design.
Testing performed included the following:

• Full-​scale bending tests aiming to evaluate the bearing capacity under flexure, that
can occur during transitional stages; i.e. demolding, storage and moving phase, and
in field due to the asymmetrical ground pressure)
• Thrust tests able to reproduce the TBM action on the segment during the excavation
process.

5.2.2.1 Full-​scale bending test

Three tests should be performed with the loading done in a closed-​loop, by adopting a
1,000 kN (112 tons) hydraulic jack, with a proportional integral derivative (PID) con-
troller and by imposing a stroke speed of 10μm/​sec (0.000393”/​sec). The segments should
be placed on cylindrical support with a known span and the load, applied at midspan, was
transversally distributed by adopting a steel beam. See Figures 5.14 and 5.15.
During the test, the following measures were continuously registered:

• The load F, measured by means of a 1,000 kN (112 tons) load cell with a precision
of 0.2%
• The midspan displacement measured by means of multiple potentiometer wire trans-
ducers placed along the transverse line
• The crack opening at midspan, measured by means of two linear variable differential
transformers (LVDTs)

Furthermore, the crack pattern should be recorded at each different step, with the help of a
grid plotted on the intrados surface 100 mm × 100mm (4” × 4”). See Figure 5.16.

5.2.2.2 Full-​scale thrust load testing

Point load tests to replicate the thrust loading from the rams of the TBM should be per-
formed by applying two point loads at the segment. The loading on the segment should be
 Fiber reinforcement in precast concrete segments 313

Figure 5.14 Segment flexural testing –​ apparatus.

Note: Typical arrangement of the full-​
s cale flexural test apparatus at the Arizona State
University lab.

Figure 5.15 Segment flexural testing diagram.

Note: Typical arrangement of the full-​
s cale flexural test apparatus at the Arizona State
University lab.

applied by adopting the same steel plates used by the TBM machine as shown in Figures 5.17
and 5.18. A uniform support is considered, as the segment is placed on a stiff beam suitably
designed. Two 2,000 kN (225 ton) hydraulic jacks were used for every steel plate.
The load was continuously measured by pressure transducers. Four wire transducers (two
located at the intrados and two at the extrados) measure the shoes displacement, while
314 Handbook of Precast Segmental Tunnel Lining Systems

Figure 5.16 Example of crack patterns recorded at different loads.

Note: Typical graphic illustration of crack pattern for deflection hardening using SFRC.

Figure 5.17 Thrust load testing apparatus schematic.

Note: Typical arrangement of the full-​s cale thrust load testing apparatus at the University of
Arizona laboratory.

one linear variable differential transformer (LVDT) transducers is applied between the load
shoes, to measure the crack openings.

5.3 QUALITY CONTROL DURING CONSTRUCTION

Quality of products is generally defined to be the consistent conformance with require-
ments. All quality assurance and quality control (QA/​QC) procedures and conformance
 Fiber reinforcement in precast concrete segments 315

Figure 5.18 Thrust load test apparatus.

Note: Typical arrangement of the full-​s cale thrust load testing apparatus at the Arizona State
University laboratory.

testing done in a typical precast concrete manufacturing plant are applicable to the produc-
tion of a precast concrete segment that incorporate fibers into the concrete mix. There
are two quality control reference documents commonly used by structural precast concrete
manufacturers in North America that are applicable to the manufacturing of precast seg-
ments utilizing structural fibers. The documents are as follows:

• The Precast Concrete Institutes (PCI) MNL-​116, Manual for Quality Control: Structural
Precast Concrete Products
• National Precast Concrete Association’s NPCA Quality Control Manual for Precast
Concrete Plants.

When fibers are used as structural reinforcement there are areas where additional test-
ing and controls need to be in-​place in the precast plant to assure the conformance with
the design and specifications for the segments are being met. These areas include the
following.

5.3.1 A ssurance fibers meet material standards and specification

The producer/​supplier of the structural steel fibers and if applicable, micro-​PP fibers for
passive fire protection, should supply with every shipment certification that fibers provided
within the shipment meets the applicable material standard specification. Each individual
package of the product should contain a label identifying the material corresponding to the
certification of compliance and showing origin of product manufacturer. The precast plant
must establish a procedure to maintain fiber identification and documenting/​track the spe-
cific fibers used in any segment as part of their quality control plan.
 n

316 Handbook of Precast Segmental Tunnel Lining Systems

5.3.1.1 Steel fiber material compliance requirements

Steel fibers appropriate for use in precast concrete segments in North America shall conform
to ASTM A820/​A820M, deformed fiber, type I, produced from cold-​drawn wire rod. With
every shipment of fibers to the project the producer/​supplier of the fibers should provide a
certificate of compliance showing that the fibers being provided were manufactured, sam-
pled, tested and inspected with this specification (including the year of issue) and any other
requirements as designated in the contract requirements.
Certification identifying the place of manufacturing of the fiber and also the specific wire
rod used to produce the fiber is particularly important for infrastructure type projects in the
United States. Typically, these types of projects will require the fibers supplied be manufac-
tured in the USA from wire drawn from domestically produced wire rod. Not complying
with these requirements or obtaining a waiver to the requirements can result in the owner’s
loss of all federal funding to a project, fines or even criminal prosecution. Table 5.6 shows
an example certificate of compliance provided by the supplier/​producer with a shipment of
fibers for a project requiring the fibers to be made and melted in the USA. All individual
packages of fibers should have labels that clearly identify the material contained within the
package and correspond to information referenced in the provided material certification for
the shipment.

5.3.1.2 Micro-​synthetic PP fiber material compliance requirements

If micro-​synthetic polypropylene (micro-​PP) fibers are required to be provided in the
concrete mix design for the segments, the manufacturer/​supplier of the fibers should
provide certification that the supplied fibers are specifically intended for this application.
Fibers used for explosive spalling resistance should be PP micro-​fibers (100% virgin
polypropylene fibers, containing no reprocessed olefin materials) conforming to EN
14889-​2:2006 Class 1a and specifically engineered and manufactured in an ISO 9001
certified facility for use as concrete secondary reinforcement. The properties of the fibers
should be constant.
Micro-​PP fibers should have been tested and proven in accredited fire testing labora-
tories to show their effectiveness in reducing explosive spalling and suppliers should dem-
onstrate a track-​record of usage in tunnel lining applications. For the tests, the range of
variation of properties have to be taken into account. Certification should include fire
testing results and the recommended dosage used to achieve the required results. Fire
testing in Europe concluded that the optimum fiber for this application is a shorter and
thinner micro synthetic mono-​filament-​PP fiber. A length of 6 mm (0.25”) and 18 µm
(0.000708”) in diameter performed the best. For more information see Section 5.4 of this
chapter.

5.3.2 Assuring a homogenous fiber distribution

Obtaining a homogenous and multi-​directional distribution of fibers inside a precast seg-
ment is crucial for achieving the desired performance. Fiber balls, clumps and areas of
higher concentration of fibers should be avoided, whether they are due to improper batch-
ing and mixing, vibration and placement, or inadequate workability of the mixture. The
fiber-​reinforced precast segment manufacturer needs to design the mix, implement proce-
dures, install equipment and make modifications that facilitate the dosing of the fibers into
the mixer.
 newgenrtpdf
Table 5.6 Steel fiber “Made in the USA” certificate of compliance example.
This is an example material certification showing that the steel fibers were produced in United States from wire rod made and
melted in United States; meeting “Buy America/​A merican Requirements”

Item Description Product data and properties

A Product data and specifications
1 Product Steel fibers Date 00 00 0000
2 Specification AS-​0 0007-​B CQA-​E N of June 2017 Order 3900676363, BOL 2867
3 Conforms to EN 14889-​1 Gross Wt 19,040 kg 41,976 lbs
ASTM A 820 Net Wt 18,700 kg 41,226 lbs
ISO 13270–​C lass A No. Units 17 each

Fiber reinforcement in precast concrete segments

B Chemical %C %Mn %Si %P %S
composition
1 Proportions 0.20% 0.64 0.19 0.008 0.020
C Properties Diameter d Length L A/​R (L/​D ) Tensile strength Rm
1 Individual values 0.75 ±0.02 mm 0.029 ±0.00008” 61 ±3 mm 2.40 ±0.125” 81 ±15% 1,800 ±15% N/​ 261,068 ±15%
mm 2 psi
2 Average values 0.75 mm ±5% 0.29 in ±5% 61 mm ±5% 2.40 in ±5% 81 ±7.5% 1,800 ±7.5% N/​ 261,068 ±7.5%
mm 2 psi
Notes:
We herewith certify that the product described above
• Conforms to the requirements of specifications:
• AS-​0 0007-​B CQA-​E N and EN 14889-​1
• ASTM A 820 –​ type I
• ISO 13270 –​ class A
• This steel product is produced in the United States in accordance with America Iron and Steel (AIS) requirements
• Meets Buy America and Buy American Act requirements:
• All steel is melted and manufactured in the United States
• All manufacturing processes occurred in the United States

317
318 Handbook of Precast Segmental Tunnel Lining Systems

5.3.2.1 Use of collated fibers

As mentioned earlier in this chapter, steel fibers used in precast concrete segments have an
optimum aspect ratio of 80. Fibers with an aspect ratio over 65 must be provided collated
(glued in clips using water soluble glue) to prevent fiber balling as shown in Figure 5.19.

5.3.2.2 Use of automatic fiber dosing equipment

The use of automatic fiber dosing systems directly linked to the control panel of the batching
plant is recommended for incorporating the fibers into the mixer as shown in Figure 5.20.

Figure 5.19 Collated (glued) vs. loose fibers.

Note: Collated (glued) fibers prevent balling during mixing.

Figure 5.20 Automatic fiber dispensing systems –​ steel and micro-​P P fibers.

Note: General arrangement of bulk fiber dispensing equipment in a precast facility.
 Fiber reinforcement in precast concrete segments 319

This equipment helps to assure that the correct dosage of the applicable fiber is added into
the mixer. The fibers are introduced into the mix at the appropriate stage of loading the
materials into the mixer. Fibers are not to be the first item loaded in the mixer. Loading
the prescribed quantity of the right fibers into the mixer is step number one in obtaining a
quality segment. Followed by sufficient mixing, then placing and vibration of the mix into
the formwork will result in a homogenous fiber distribution network throughout the seg-
ment with performance that meets or exceeds the specifications.

5.3.2.3 Placing and finishing segments

While casting the segments, special attention needs to be taken to the placing and vibration
of the fiber reinforced concrete (FRC) into the segment molds. Inspecting that there are no
fiber balls present and that the fibers appear to be homogenously distributed. Unlike plain
concrete, which can be internally vibrated, all FRC, test specimens and segment molds,
should only use external vibration. Care must be taken that the FRC is not over vibrated
to help ensure the fibers are randomly oriented and homogenously distributed throughout
the entire segment.
The exposed concrete surface will first be screeded-​off, and then smooth finished with
steel trowels. Fibers should not protrude from exposed surfaces and should be covered with
a layer of paste with a minimum thickness of 0.2 mm (0.008”) to prevent any surface cor-
rosion. Unlike conventional reinforcing steel bars or wire mesh, steel fibers require minimal
cover to provide long term durability. Steel fibers compared to conventional reinforcement
are very thin, discrete pieces of steel distributed throughout the matrix, that even if corro-
sion is initiated in a fiber near the surface, it remains isolated and is not detrimental to the
durability of structural performance of the segment.

5.3.3 Material sampling and testing

The American Concrete Institute (ACI) through its 544 Committee on Fiber Reinforced
Concrete, has developed and published two documents that are must haves for any
FRC segment producer. The first document is entitled, ACI 544.2R–​17, “Report on the
Measurement of Fresh State Properties and Fiber Dispersion of Fiber-​Reinforced Concrete”.
ACI 544.2R-​17 supersedes ACI 544.2R-​89 and was adopted and published May 2017. The
second document is entitled, ACI 544 9R-​17, “Report on Measuring Mechanical Properties
of Hardened Fiber-​Reinforced Concrete”, ACI 544.9R-​17 was adopted and first published
January 2017. These two documents are excellent resources for understanding the proper
methods of making specimens (beams, cylinders, etc.) and performing conformance testing
of all FRC.

5.3.3.1 Fresh state sampling and testing

Once the FRC is mixed and ready to be placed into the molds, the typical fresh state prop-
erties testing done on plain concrete; i.e. slump, air content, unit weight, etc., are required
to be performed on the FRC as well. In addition, FRC requires additional sampling of the
fresh mix to perform testing such as determining that the fiber content is within an accept-
able range of what is required. Typically, this is done through a wash-​out test. Also prior
to placing the concrete, a sufficient quantity of fresh FRC needs to be obtained for making
testing specimens for testing the hardened properties of the mix. In addition to the typical
320 Handbook of Precast Segmental Tunnel Lining Systems

cylinders for doing concrete compressive strength testing, specific test specimens such as
beams and cylinders also need to be made to determine the structural design performance
requirements of the hardened SFRC. (residual flexural strength)

5.3.3.2 Hardened state testing in the precast plant

Steel fibers in the volumes (typically about 0.5%) added in precast segment mixes do not
impact the compressive strength of the hardened concrete. Standard compressive strength
testing per the ASTM C31 standard are performed in the precast plant typically at the time
of demolding and then 7 and 28 days. The splitting tensile strength test per the ASTM
C496 standard on a test cylinder is also routinely a part of the QC testing protocol. The
splitting tensile strength of the concrete is also not impacted by fibers as this is a measure
of the modulus of rupture (MOR) of the concrete, which is a function only of the concrete
compressive strength. Steel fibers, like conventional steel reinforcing, do not impart any ten-
sile strength until after the concrete has initially cracked. MOR of the concrete is typically
about 10% of its compressive strength. Regardless, the primary difference in the perform-
ance of these tests is that when making the cylinders the concrete is not to be consolidated
by rodding. For that matter, all SFRC cast testing specimens are to be filled in a single lift
and externally vibrated, regardless of the test.
Testing to achieve residual flexural tensile strength requirements of the SFRC are of crit-
ical importance. Typically, multiple sets of beam test samples are produced from the fresh
SFRC mix, cured and stored in controlled environment tested at prescribed intervals; typic-
ally tested in groups of three beams at a time. For the reasons outlined in Section 5.2.1.1 of
this chapter, until a notched alternative is available in ASTM, the EN 14651 beam protocol
should be employed in the quality control testing of the beams. Other testing protocol that
has recently gained limited acceptance as a residual tensile strength QC testing protocol is
the test known as the double punching test (DPT) or sometimes called the Barcelona test.
This type test is useful only for QC purposes in the precast plant. The DPT test provides
information on the “overall” residual flexural strength and toughness of a specific mix. This
test does not provide information relating to the characterization of the FRC behavior.
From a design conformance standpoint, the relationship between the loading and deflec-
tion in a beam are critical parameters that provide crucial information of how the composite
performs. SLS and ULS design values are of key importance to the structural design of pre-
cast segments. Post-​crack deflection hardening or softening behavior of the composite can-
not be determined using the DPT test. Because of interest in using this simplified test method
compared to performing beam testing in the precast plant, ASTM C09.42 subcommittee on
fiber reinforced concrete currently has a task group and work item that is developing an
ASTM testing standard for this type test. If this standard becomes available, the test will
become a viable quality control only test. Because of the limitations it will not be a substi-
tute for performing beam flexural testing and/​or full scale segment testing for determining
mix conformance with the structural design.
There are also instances where quality control testing is required to be performed on cores
or beams cut from actual SFRC segments. Core samples can be taken from a production
segment and used for random evaluation of fiber distribution in a segment, compressive
strength, splitting tensile strength or even DPT testing of the cored sample. As mentioned
earlier in this section, ACI 544.2R-​17 and ACI-​544.9R-​17 provide more detailed informa-
tion and should be part of the precast producers’ library that produces any FRC precast
elements.
 Fiber reinforcement in precast concrete segments 321

5.4 PASSIVE FIRE PROTECTION (PFP) WITH MICRO-​P P FIBERS

5.4.1 History on the use of micro-​PP fibers for PFP

In the 1980’s and 90’s in Europe there were several major fires in tunnels in Europe that
brought to the forefront the problem of explosive spalling of the concrete in tunnel lin-
ings when subjected to fires. This included a major fire in the Channel Tunnel between
France and the United Kingdom in 1996 as shown in Figure 5.21. It was reported that a
significant loss of cross-​sectional area in the precast segmental tunnel lining had occurred
due to severe spalling of the high strength concrete that had been exposed to very high
temperatures. In some sections, the loss of concrete was so great that the embedded
reinforcement steel had become so exposed that the structural integrity of the tunnel had
been placed at risk. This spalling cannot only become a structural issue, but it is also a
life safety issue.
Explosive spalling of the concrete is a life safety issue to anyone that is caught in the
tunnel needing to escape or to fire fighters or emergency medical services (EMS) workers
that need to enter the tunnel. Explosive spalling can result in large pieces of concrete to be
violently ejected over significant distances away from the lining. As the lining spalls away a
fresh concrete face is presented to the fire, progressive explosive spalling can continue deep
into the concrete section, threatening the structural integrity of the construction by reducing
the cross-​section of concrete and exposing the reinforcement to fire.
The Channel Tunnel Rail Link (CTRL) project in the United Kingdom which began in
2000 was a new high-​speed line to traverse 108 km (68 miles) between the Channel Tunnel

Figure 5.21 Channel Tunnel Rail Link fire in 1996.

Note: View of the explosive spalling of the precast segmental concrete lining. Later designs
included the micro-​s ynthetic fibers.
322 Handbook of Precast Segmental Tunnel Lining Systems

and St. Pancras Station in central London. Because of the extensive damage from the fire in
the Channel Tunnel in 1996, CTRL decided to do an extensive fire testing at the University
of Delft in the Netherlands to evaluate different concrete mixes including monofilament
and fibrillated polypropylene fibers. This testing found that by inclusion of 1.0 kg/​m3 (1.7
lbs/​CY) of monofilament polypropylene fibers in the high strength, low permeability mixes
significantly the risk of explosive spalling when exposed to a severe hydrocarbon fire was
reduced. The CTRL project was the first project to use the recommended 1.0 kg/​m3 (1.7
lbs/​CY) of micro-​PP fibers for controlling explosive spalling. In this project 30 kg/​m3 (51
lbs/​CY) of steel fiber was used as the only reinforcement along with the micro PP fibers as
passive fire protection (PFP).

5.4.2 Basic concepts of explosive spalling

Explosive spalling occurs due to steam pressure build-​up within the pores of the concrete
when subject to temperatures above 250˚°C. The build-​up of steam pressure produces
stresses, which the material is unable to resist. This stress is estimated to be about 8 MPa
(1,200 psi) for a temperature of 300°C (572°F) and 17 MPa (2,500 psi) for a temperature
of about 350°C (662°F). When you consider the tensile strength of the concrete is typically
much less than these values, the spalling behavior of the material is understandable.
The primary role of the short and very thin micro-​PP fibers is to create a network of fib-
ers, that then when exposed to heat will melt and create pathways for the vapor pressure
(steam) to escape, thus preventing the explosive spalling. The concept of explosive spalling
and prevention using micro-​synthetic fibers is shown in Figure 5.22.

Figure 5.22 Basic concepts of using micro-​s ynthetic polypropylene (PP) fibers for PFP.
Note: General arrangement mechanism illustrating the failure of tunnel lining in compression.
 Fiber reinforcement in precast concrete segments 323

5.4.3 Fiber and dosage recommendations

The determination of the optimum dosage for a specific micro-​PP fiber that will provide
explosive spalling resistance for a specific project, can be established by large-​scale fire test-
ing of the actual concrete mix with the specific fiber to be used on the project. Large scale
means to do the test under the same load, stress and fixity conditions as the concrete will
have at the construction site. This is a costly exercise and an expense that many projects
would like to avoid. For many projects, engineers rely on information provided from the
suppliers and manufacturers of the fiber on fire testing and certifications previously done
on their fiber.
Fire resistance tests on SFRC segments have recently been carried-​out in Europe. This
showed that the use of steel fibers also provides a benefit over conventionally reinforced
concrete in reducing explosive spalling. This is due to the thermal conductivity of the con-
ventional reinforcements, which can also contribute to rapid bursting of the exposed con-
crete. Testing has shown the ability to reduce the dosage of micro-​PP fibers when used with
SFRC. Here is an example specification of a micro-​PP fiber that has been tested/​certified for
this application:

• Virgin polypropylene Mono-​f ilament

• Nominal tensile strength R M ≥250 N/​m m 2 (≥36,260 psi)
• Melting point 165°C (329°F)
• Density <950 kg/​m 3 (<1,600 lbs/​C Y)
• Elastic modulus ≥3,500 N/​m m 2 (≥507,630 psi)
• Length 5 mm to 7 mm (0.197” to 0.276”)
Testing showed 6 mm (0.25”) performed better than 12 mm
(0.47”) long
• Diameter <20 μm (0.000787”)
Testing showed 18 μm (0.000708”) performed better than
32 μm (0.001259”)
• Dosage 1.0 to 2.0 kg/​m 3 (1.7 to 3.4 lbs/​C Y)
Testing showed dosage could be lowered if SFRC used

The dosing of micro-​PP fibers into the mix is a relatively simple process. For smaller pro-
jects, micro-​PP fibers are typically supplied in fully degradable paper packaging that enables
the desired dosage per unit volume to be simply added directly into the concrete truck or
mixer. The packaging is designed to rapidly break down allowing uniform distribution of
the fibers into the concrete. In smaller projects this is often the most cost-​effective method
to adopt, with packaging available in 1.0 kg (1.7 lbs/​CY) or 2.0 kg (3.4 lbs/​CY) bags. On
larger precast segment project applications that involve significant quantities of micro-​PP
fibers the use of a separate automatic dosing equipment can be employed. See Figure 5.20
for example.

5.5 SUSTAINABILITY
Today green is not just a color! The modern environmental movement or “green revolu-
tion” as it is sometimes called had its roots in the 1960’s in the USA. On 22 June 69, the
324 Handbook of Precast Segmental Tunnel Lining Systems

Cuyahoga River in Cleveland, Ohio, burst into flames five stories high as a result from long
time dumping of oil and chemical pollution into the river. This event became what is now
considered a defining moment in the modern environmental movement. In 1969, the US fed-
eral government enacted legislation called The National Environmental Policy Act (NEPA).
This act established the US Environmental Protection Agency (EPA) in 1970 that continues
to exist today. For the last 50 years the United States has been committed to a national
policy “to create and maintain conditions under which humans and nature can exist in
productive harmony, that permit fulfilling the social, economic and other requirements of
present and future generations”.
Today the terms sustainability and sustainable development are embraced globally. The
concept of sustainability is one that has been around for as long as humans have: a concern
for the future of our resources. The word itself, however, has more recent origins. A term first
used in Germany, Nachhaltigkeit, meaning “sustained yield”, appeared in a forestry hand-
book published in 1713. The term was used to say that to be “sustainable” in forestry you
should never harvest more trees than the forest can regenerate. An example of sustainable
use today in the context of fisheries is that fishing limits of a particular species are established
to assure that species population does not decline over time because of over-​fishing practices.
In the 1980s, the term sustainability began to be used in terms of how humans live on the
planet. Today, the concept of sustainability when used in terms of a construction project
being sustainable, has roots in the 1980’s. The United Nations convened a group in 1987,
known as the World Commission on Environment and Development (WCED), also called
as The Brundtland Commission, that was first to use the term “sustainable development”.
The commission’s report defined sustainable development as “development that meets the
needs of the present generation without compromising the ability of future generations to
meet their own needs”.
The designer of a reinforced concrete precast tunnel segment tasked with providing a
sustainable TBM bored tunnel lining must consider not only environmental impacts of the
design, but structural requirements must also be met as well as meeting minimum service life
requirements through providing increased durability. When the geology requires the tunnel
to be lined, the use of SFRC in a precast tunnel segment, either in combination with bar
reinforcement or alone, has been found to provide a more sustainable solution over other
options.

5.5.1 Environmental impacts of greenhouse gas emissions

Scientists have shown that human generated emissions of greenhouse gases (primarily CO2)
from fossil fueled power supplied to industrial processes and fossil fueled vehicles, are the
primary contributors to the dramatic rise in Earth’s temperature since the 1950’s. Model
simulations by the Intergovernmental Panel on Climate Change estimate that Earth will
warm between two and six degrees Celsius over the next century, depending on the rate of
carbon dioxide emissions from the use of fossil fuels. When global warming has happened
in the past two million years (the end of the last ice age), geological records indicate that it
took the planet about 5,000 years to warm five degrees Celsius. This current predicted rate
of warming for the next century is at least 20 times faster!
From National Aeronautics and Space Administration (NASA)’s website on Global
Climate Change:

The consequences of changing the natural atmospheric greenhouse are difficult to pre-
dict, but some effects seem likely: On average, Earth will become warmer. Some regions
 Fiber reinforcement in precast concrete segments 325

may welcome warmer temperatures, but others may not. Warmer conditions will prob-
ably lead to more evaporation and precipitation overall, but individual regions will vary,
some becoming wetter and others dryer.
A stronger greenhouse effect will warm the ocean and partially melt glaciers and
ice sheets, increasing sea level. Ocean water also will expand if it warms, contrib-
uting further to sea level rise. Outside of a greenhouse, higher atmospheric carbon
dioxide (CO2) levels can have both positive and negative effects on crop yields. Some
laboratory experiments suggest that elevated CO2 levels can increase plant growth.
However, other factors, such as changing temperatures, ozone, and water and nutrient
constraints, may more than counteract any potential increase in yield. If optimal tem-
perature ranges for some crops are exceeded, earlier possible gains in yield may be
reduced or reversed altogether. Climate extremes, such as droughts, floods and extreme
temperatures, can lead to crop losses and threaten the livelihoods of agricultural pro-
ducers and the food security of communities worldwide. Depending on the crop and
ecosystem, weeds, pests, and fungi can also thrive under warmer temperatures, wet-
ter climates, and increased CO2 levels, and climate change will likely increase weeds
and pests.
Finally, although rising CO2 can stimulate plant growth, research has shown that it
can also reduce the nutritional value of most food crops by reducing the concentrations
of protein and essential minerals in most plant species. Climate change can cause new
patterns of pests and diseases to emerge, affecting plants, animals and humans, and pos-
ing new risks for food security, food safety and human health.

Although the consequences of climate change is difficult to predict, the rapid increase in
Earth’s temperature will bring about detrimental impacts to future generations. It is increas-
ingly evident that we must act to reduce manmade levels of CO2 emissions into our atmos-
phere. To be considered a sustainable project today, design engineers are required to make
conscious efforts to reduce the carbon footprint as well as provide for a minimum service
life of structures they are designing. Considerations on selection of design and construction
process are no longer based on the lowest first cost. This section of the chapter will look at
how the use of fibers to replace all or a portion of the conventional reinforcement in a pre-
cast tunnel segment provides for a significant reduction of the carbon footprint of the tunnel
and high-​performance steel fibers provide smaller crack widths and improved durability and
increased service life.

5.5.2 Sustainability assessments
Determining the environmental impacts of one construction process or alternative to
another needs to be objectively assessed. The engineer must also consider that reducing
the environmental impact of a structure is only one consideration. The structural design
of a tunnel lining segment is primarily dependent upon the load conditions applied to it
(ground and groundwater loads, handling, transportation, and erection, as well as live
loads). The benefits of reduced embodied CO2 can only be realized when the selected
reinforcement alternative also meets all the design requirements (structural, safety and
durability).
fib Bulletin 83, “Precast Tunnel Segments in Fibre-​Reinforced Concrete”, contains a
chapter entitled, “10.1 Sustainability Index”. The chapter proposes a method to compare
precast concrete segment alternatives taking into account the economic, environmental
326 Handbook of Precast Segmental Tunnel Lining Systems

and social requirements involved in the production of the precast segments. The com-
parison is carried-​out on three different precast segment alternatives by means of the
sustainability and mechanical indexes assessed for each alternative. The chapter provides
a detailed assessment on an example precast concrete tunnel segment using three differ-
ent alternative reinforcing solutions. This chapter is a good reference for an engineer that
needs to perform a sustainability assessment on precast concrete segmental tunnel lining
alternatives.

5.5.3 Reduced carbon footprint

The use of steel and synthetic fibers to replace all or a part of conventional reinforcement
has been demonstrated to lower the embodied CO2 of the segmental lining. Macro-​synthetic
fibers are shown to provide a lower embodied CO2 content than steel, however, as addressed
in Section 5.1.1.2 Steel vs. macro-​synthetic fibers, are unable to meet all of the structural/​
mechanical performance requirements of a precast segment, with limited exceptions.
Therefore, this section will focus on precast segment lining using SFRC to replace rebar/​
mesh. While it is possible to significantly reduce the embodied CO2 of a concrete mixture for
segment production by replacing a portion of its cement content with alternative cementi-
tious materials, there is little or no difference between the cementitious blends and contents
required to produce fiber reinforced or conventionally reinforced concrete segments for
tunnel linings.
Figure 5.23 is data from a paper entitled, “Consultant’s View on Service Life Design”,
by Carola Edvardsen of COWI Denmark, and provides a comparison of embodied CO2 for
different types of binder and steel reinforcement used for major infrastructure projects in
Europe and the Middle East. On a per pound (kg) basis, the embodied CO2 of conventional

Figure 5.23 Embodied CO 2 for different types of binder and steel reinforcement.

Note: Comparison of embodied CO 2 for different types of binder and steel reinforcement used
for major infrastructure projects.
 Fiber reinforcement in precast concrete segments 327

rebar and steel fibers is assumed the same. This is a generalization assuming the wire rod
that the fiber is produced from, and the rebar have similar percentage recycled material
content and similar steel production methods. In a precast concrete segment design the per-
centage of reduction of steel is dependent on the structural and serviceability requirements
of the segment, the fiber’s attributes, and performance in the specific concrete mix. Based on
information from a leading steel fiber producer, a typical reduction percentage in steel when
using steel fibers in a precast concrete segment is typically 60 to 65%. The actual reduction
in embodied CO2 of the lining will depend on the contribution of the reinforcement plus
the concrete.
From the descriptions of the reference projects listed above, it is noted that advances with
regard to durability are carried on from either previous projects from the region. Moreover,
as will be presented in the following, the advances in terms of durability are associated with
a gain in terms of sustainability, but it is stressed that the advances in terms of sustain-
ability, e.g. reduction of CO₂ emission, have been made without sacrificing the durability
of the concrete structures. The key to the innovations and the corresponding gains regard-
ing durability and sustainability lies in a construction material and the overarching design.
Some examples presenting selected trends regarding concrete durability design of bored,
segmental linings in the Middle East are presented along with the benefits from a sustain-
ability perspective are given in the list below.

• Steel fiber reinforced concrete

• Less maintenance
• Increased durability
• Reduced steel content and CO₂ emission
• Use of supplementary cementitious materials, i.e. GGBS, FA, and/​or MS
• Required to achieve denser concrete (reduced diffusion)
• Reduced CO₂ emission
• Omission of epoxy coatings on segment surface
• Reduced cost

To illustrate the sustainability-​advances made in recent years by using supplementary

cementitious materials such as GGBS and FA as partial replacement for OPC, as well as
using steel fibers instead of traditional bar reinforcement, Figure 5.23 presents schematic­
ally the calculated CO₂-​eq. emission from production per m³ concrete from projects in the
Middle East, along with other, European reference projects.

5.5.4 Economic and durability benefits

The environmental, economic and the increased durability benefits provided using high-​
performance steel fibers to replace all or a portion of conventional reinforcement in a pre-
cast segment is substantial. A durable tunnel, one that lasts a long time, provides a long
period of time to amortize the environmental and economic costs that were incurred in
building it. Hypothetically, if you double the life of a tunnel, you halve the environmental
impacts of it being built. A tunnel designed and built to last 100 years offers significant
resource advantages over a comparable tunnel that will last just 50 years. The improved
durability of SFRC segments coupled with the fact that the use of SFRC precast segments
provides substantial first cost savings over conventional only reinforced segments magnify
the economic benefits and sustainability of the tunnel.
328 Handbook of Precast Segmental Tunnel Lining Systems

5.5.4.1 Economic benefits of using SFRC

The economic benefits of steel fibers replacing all or part of the conventional reinforcing in a
precast segment is a fundamental reason that the use of SFRC in segments is now considered
the state-​of-​the-​art. The first cost savings are realized in both manufacturing and installa-
tion of the precast segments. The cost savings in the manufacturing of SFRC segments are
derived from the following:

• High-​performance steel fibers typically provide a 40 to 60% reduction in quantity of

steel required to meet the design structural requirements. Even if a combined solution
is required to meet design requirements –​less steel overall is required.
• Reduction in labor costs associated with the elimination or vast reduction of conven-
tional reinforcing.
• Increased efficiencies in segment production per shift are realized.
• Use of bulk packaging of fibers and automatic dosing systems simplify addition of fib-
ers saving time, money and assuring correct amount of fibers per batch.
• SFRC greatly increases the impact resistance and toughness of the segment. These
characteristics reduce damages incurred during demolding, handling, storage and
transport of the segments. Segments damaged during manufacturing require either
costly repairs or rejection of segments.

The cost savings in the installation of SFRC segments is due to significant reduction of dam-
age to the segment caused by the high, concentrated jacking forces on the segment that drive
the excavation. The thrust loads applied by the jacking rams acting in the cross section of
the tunnel induce spalling and bursting stresses in the circumferential joints of the tunnel
segments. Conventional reinforcing steel requires cover whereas steel fibers can reinforce
all the way to the edges. When spalling occurs, the segment requires costly repairs or even
replacement, which can then delay the excavation. The damaged edges also compromise the
durability of the concrete and performance of the gasket in the damaged area. Experience
has shown that the use of SFRC in segments substantially reduces this type of damage dur-
ing installation. See Figure 5.24.
A study undertaken in the United Kingdom in 2004, by the Client Manager for the High
Speed 1, Channel Tunnel Rail Link (CTRL) Project. The CTRL is the high-​speed Euro-​
Star train line running for 108 km (68 miles) between the Channel Tunnel and St. Pancras
Station in central London. This project was a real break-​through, with more than 260,000
segments with SFRC only, using 30 kg/​m3 (50.5 lbs/​CY) of steel fibers along with 1 kg/​m3
(1.70 lbs/​CY) of micro-​PP fibers for passive fire protection. The study concluded that using
SFRC significantly improved the handling capacity and surface finish of the segments.
Segment damage requiring repair or replacement during manufacturing and installation
was minimized compared to rebar reinforced segments. The study concluded that the use of
steel fibers to replace the conventional rebar in the segments ended up providing more than
a 10% savings on the estimated installation costs of the tunnel lining. This level of savings
by replacing conventional reinforcement with steel fibers has become a consistent outcome
throughout the tunneling industry.

5.5.4.2 Durability benefits of using SFRC

A key component of a sustainable tunnel design is designing durability into the structure. It
is universally agreed that SFRC has improved resistance to chloride and CO2 induced cor-
rosion in concrete, compared to conventional reinforcing steel.
 Fiber reinforcement in precast concrete segments 329

Figure 5.24 SFRC segment reinforcing and performance.

Note: SFRC concrete segments provide reinforcing throughout the cross-​section preventing
damage to the edges during handling and installation.

• High-​quality concrete limits fiber-​corrosion to the outer 2 to 5 mm (0.08 to 0.20”) on

uncracked concrete;
• Critical chloride threshold of steel fibers is significantly higher than conventional
carbon steel rebar;
• Corrosion of steel fibers in uncracked concrete does not lead to a long-​term reduction
of the flexural strength and toughness.

Fibers, like conventional rebar, is passivated by the high alkalinity of the concrete, which
protects against corrosion. Rebar requires the bars to be embedded within the concrete,
whereas steel fibers require little to no cover to become passivated. Fibers that are located
within the 2 to 5 mm (0.08 to 0.20”) of the surface are susceptible to corrode with time.
Corrosion initiation is due to ingress of chlorides or other deleterious substances as well as
higher levels of oxygen availability. Fibers will remain passivated in good-​quality uncracked
concrete, but if there is carbonation of the concrete, which lowers the pH in an area fiber
corrosion can take place. Regardless of this, because of the discrete nature and very small
diameter [in a segment typically less than 1mm (0.04”)] of the fiber any corrosion initiation
is isolated (no electrical continuity of the network) and corrosion by product volume is so
little, even when occurring at or near the surface no cracking, spalling or structural degrad-
ation that happens with conventional bar will result.
Durability of a tunnel liner is controlled by limiting the crack width to specific levels to
prevent increased permeability, excessive water leaks, and reinforcement corrosion. Poor
durability of conventional only reinforced segments by the bar corroding is a consequence
of wide crack widths inherent with the bar requiring cover. Steel fibers, on the other hand,
have a very positive effect on cracking and crack propagation. The distance between steel
fibers is much smaller than typical spacing for reinforcing bars.
Unlike reinforced concrete, fibers are homogenously distributed throughout the entire
segment cross-​section providing reinforcement all the way to the concrete surface and edges.
Performance of SFRC in cracked concrete is based on the width of the cracks and if a
fiber near the surface crosses a crack or not. The use of deflection hardening SFRC promotes
multiple-​crack development, thereby significantly reducing the permeability of concrete in
service. Unlike reinforced concrete, fibers are distributed throughout the whole
330 Handbook of Precast Segmental Tunnel Lining Systems

Figure 5.25 Crack propagation in steel fiber reinforced concrete.

Note: Typical arrangement of crack formation in SFRC under bending loads.

section. Hence, there is no area throughout the entire cross-​section without reinforcement.
Furthermore, stresses in the root of a crack can be picked-​up quicker. This is why crack
propagation and crack patterns change when compared to plain or even reinforced con-
crete. Please refer to Figure 5.25.
Steel fibers typically bridge cracks at a non-​perpendicular angle so they will already be
deforming and picking-​up load at small crack widths. Local friction is increased and thus
compressive stresses parallel to the crack surface are induced. As a consequence, the associ-
ated tensile stresses perpendicular to the crack can lead to secondary cracking. Those cracks
may be compared to cracks in conventional reinforced concrete, which can be found in the
zone directly around rebar. With steel fiber concrete, secondary cracking can be observed
over the whole cracked section. Subsequently cracks become more curved. Fragmentation,
translation and multiplication can be identified. Resistance to intruding substances, espe-
cially liquids, is substantially increased with aggregate interlock and friction also enhanced.
With high performance steel fibers, engineers can design the SFRC for a maximum allow-
able crack width. Small crack widths will assure little risk of corrosion of the fibers near
the surface as well as a very low permeability to reduce the ingress of potentially dele-
terious substances into the segment and control leaks. Control of micro-​cracking is also an
important parameter. Steel fibers have been used successfully in tunnel projects to reduce
cracking widths to 0.20 mm (0.80”). When a crack in SFRC does not exceed 0.20mm
(0.80”) width, it presents a very tortuous and discontinuous path. This makes the diffusion
of aggressive agents or water far more difficult. Further research has shown that when a
crack in SFRC does not exceed 0.30 mm (1.20”), self-​healing mechanisms occur, and the
corrosion products fill in the interior of the cracks. These two physical mechanisms conse-
quently obstruct the cracks and, therefore, prevent further deterioration.
In cases where the design requires a combined solution of steel fibers and conventional
rebar, the steel fibers enhance the durability compared to a rebar only reinforced segment.
If there are extremely severe exposure conditions that are present or no surface corrosion
of fibers is desired for aesthetics, galvanized or solid stainless steel fibers are available with
some limitations. If the project requires a combined solution, any of these types of steel fib-
ers can be combined with corrosion resistant rebar (CRR) with no concern of any dissimilar
metal issues in the concrete. For a very durable combined solution the use of SFRC with
glass fiber reinforced polymer (GFRP) bar is a consideration.
The use of electric powered trains in railways is increasing being considered to move to
lower carbon footprints and increase the sustainability of the transportation system. A dur-
ability issue that sometimes calls to be addressed are concerns of durability due to stray cur-
rent induced corrosion in reinforced concrete. Stray currents may be DC or AC depending
 Fiber reinforcement in precast concrete segments 331

upon the source. Typically, electrified railways utilize DC power to drive the trains. In the
case of DC current, a cathodic reaction; e.g. oxygen reduction or hydrogen evolution takes
place, where the current enters the structure (or, reinforcement), while an anodic reaction
takes place; e.g. metal dissolution, where the current returns to the original path, through
the soil (or, concrete). The precast segment tunnels that serve DC electrified railway applica-
tions require careful consideration of the potential for stray current-​induced corrosion of
the reinforcement in the lining.
The initiation of stray-​current induced corrosion in a reinforced concrete segment is the
transmission of current by the steel reinforcement though the lining. Kangkang Tang, a
Senior Lecturer at Brunel University London, published a study entitled Corrosion of Steel
Fiber Subjected to Stray Current Interference in the March 2020 ACI Materials Journal that
showed steel fibers are much less prone to picking-​up and transferring direct current com-
pared to traditional bar reinforcement. Initiation of stray current induced corrosion requires
the steel to pick up and then transport of current throughout the structure. In conventional
rebar reinforced concrete this this is an issue. However, in SFRC it was found that due to
the short length of the fibers it is difficult to have an anode and a cathode form on the same
fiber. Without this formation the transfer of current from fiber to another and then through
the structure cannot occur. Tang’s study concluded that since the fibers likely do not transfer
stray-​current there is in practice no risk of stray-​current induced fiber corrosion in SFRC
used in DC electrified railway applications.

5.5.5 Trends in development of low carbon concretes

Concrete is one of the most widely consumed materials on the planet and is said to con-
tribute approximately 8% of global carbon emissions; the main source of these emissions is
the manufacture of ordinary Portland cement (OPC). In a tunneling project, it is generally
considered that 60% to 70% of embodied carbon is contained in the concrete linings of
the shafts and tunnels. The use of steel to replace all or part of conventional reinforcement
has been demonstrated to lower the embodied CO2 of a segmental lining. While it is pos-
sible to significantly reduce the embodied CO2 of a concrete mix for segment production by
replacing a portion of its cement content with alternative cementitious materials, there is
little or no difference between the cementitious blends, and the contents required to produce
fiber-​reinforced or conventionally reinforced concrete segments for tunnel linings.
Low-​carbon classified concretes today generally comprise a range of concretes with OPC
replacements of up to 70% by ground-​granulated blast-​furnace slag (GGBS) and other
cement replacement materials, such as fly ash and limestone. The use of these typical low-​
carbon footprint concrete in a precast segmental lining has its own unique set of challenges.
The specification of the typical low-​carbon concretes for use in the precast segments should
be evaluated carefully. These materials can have longer setting times, slower strength devel-
opment and might even end up providing higher embodied carbon through the need for
increased heat curing. Heat curing is often applied to the freshly cast segments to achieve
the required demolding strength within 6 to 8 hours from casting. On a major project, this
can mean operating a curing chamber temperature of around 50°C (122°F) on a 24/​7 basis
for months or even years. This will also add to the carbon footprint of the segmental lining
and is dependent on many factors, such as type of fuel, heat losses from the chamber and
size of chamber.
Developments in geopolymer concrete and alkali-​ activated cementitious materials
(AACM) in recent years have made it a possibility to completely replace the OPC in the
332 Handbook of Precast Segmental Tunnel Lining Systems

mix. These special concretes replace the OPC with replacement materials such as GGBS, fly
ash, metakaolin and calcined clay; they are activated with the addition of alkali additives
such as hydroxides and silicates. There are new products that are becoming available that
completely replace the OPC in the concrete and have shown to provide improved proper-
ties for use in precast segments. Alkali-​aggregate reaction (AAR) and alkali-​silica reaction
(ASR) and their associated deleterious expansion of concrete or mortar can cause major
long term durability problems in concrete structures. It is important that testing is done to
address this concern.
In a recent paper, Charles Allen (OTB Consulting) refers to the case for using one such
product in segmental linings called “Earth Friendly Concrete” (EFC®) geopolymer con-
crete. EFC has been developed over the past ten years by Wagner Concrete in Queensland,
Australia. The structural performance of EFC has been tested and shown to be equal to that
of OPC-​based concretes while its durability often exceeds that of OPC-​based concretes.
The results from the beam tests show that the SFR EFC has similar magnitude limit of
proportionality (LOP) and residual flexural tensile strengths to those of a typical C50/​60
Portland cement-​based concrete containing high performance steel fibers. See Figure 5.26.
Steel and polypropylene fiber reinforced EFC has been tested for fire resistance to the RABT-​
ZTV (Eureka) fire curve in Australia and shows equivalent performance to similar concretes
based on OPC cements. In the paper Charles states,

The incorporation of EFC geopolymer concrete is an action that could be undertaken in

the manufacture of reinforced concrete tunnel lining segments. EFC can provide signifi-
cant reductions in the embodied carbon footprint of tunnel linings, estimated to be of
the order 210 kg/​m3 (354 lb/​CY) of concrete–​approximately 1.8 tonnes per linear meter
(0.60 tons per linear foot) of embodied CO2 of a typical rail tunnel. Additionally, when
casting EFC in segment molds on a carousel system, the curing chamber temperature can
be reduced to around 30°C (86°F) thereby, reducing energy consumption and associated
carbon emissions. The final costings and environmental benefits of the use of EFC

Figure 5.26 Earth friendly concrete (EFC) geopolymer concrete days’ age.

Notes: EFC geopolymer concrete with 35 kg/​m 3 (59 lbs/​C Y) steel fibers BS EN 14651 beam test
at 145 days’ age.
 Fiber reinforcement in precast concrete segments 333

in tunnel linings could be quantified by conducting full-​scale production trials with the
segment manufacturer.

A combination of appropriate low-​carbon concrete mix design combined with the efficient
use of steel fibers in the design of the segment can give dramatic reductions in the embodied
carbon of tunnel lining segments when compared with traditional segments used in major
projects.

5.5.6 Project examples
5.5.6.1 Project ­example –​Paris Metro Line 16-​1
The Grand Paris Express project is an investment of $38.2 (€35.6) billion in 2012 num-
bers and will add more than 68 stations and 200 km (125 miles) of trackway to double
the existing mass transit rail network for Paris. The 100 km (62 miles) currently in con-
struction includes Line 16, which is scheduled to be in operation for the Paris Summer
Olympic Games in 2024. The $1.98 (€1.84) billion Grand Paris Express Line 16-​1 is a
19.3 km (12 mile) long tunnel contract that was awarded in February 2018 to a con-
sortium led by Eiffage Génie Civil. The project employs the use of six tunnel boring
machines of three different diameters; 9.0, 7.7 and 9.5 m (29.2, 25.3, and 31.2 feet)
outer diameter, that will install approximately (200,000 m3 (261,600 CY) of precast
concrete segments in the lining.
It was soon after award of initial contracts for the Grand Paris Express that Société
du Grand Paris, the owner of the mega-​project, began to explore the use of steel fiber
reinforcement for the segmental linings to replace the rebar reinforcement. One of their
primary objectives was to enhance the sustainability credentials of the project, to pro-
mote respect for the environment and reduce as much as possible the carbon footprint.
The specifications for the 16-​1 work package provide for conventional rebar only –​no
fibers. Three months after awarding of the contract for Line 16-​1 in 2018, Eiffage Génie
Civil proposed to Société du Grand Paris to implement a 100% SFRC segments. The
objectives by using steel fiber as an alternative to rebar reinforcement was to reduce the
amount of steel required per cubic meter of precast lining and capture the benefits that
steel fiber creates less carbon pollution in its manufacture and transportation. The use of
SFRC plus formulating a low carbon concrete mix would optimize the carbon footprint
of the project.
In France, prior to when Line 16-​1 was launched, steel fiber reinforced concrete (SFRC)
segments had only been used in temporary (sacrificial) or experimental lining segments
(in siding areas, for example). In the rest of Europe, excluding Spain (due to the use of
hybrid segments; i.e. fiber-​reinforced concrete segments combined with a perimeter cage
of reinforcing bar) and the United Kingdom (where smaller diameters are the norm, with
numerous segments per ring; typically 9+​1, to distribute loads as for CTRL), there are no
references for all-​SFRC segments being used in large diameter metro tunnels in Europe
such as Line 16-​1. At a global level, particularly since the publication of the fib Model
Code 2010, the Anglo-​American nations have pioneered the use of 100% SFR segments.
Although the United States, Canada and Australia provided the momentum, the Doha
metro project in Qatar marked a milestone in 2014, by proving that, in a particularly
aggressive environment, an all-​SFRC solution combined with innovative low-​carbon
334 Handbook of Precast Segmental Tunnel Lining Systems

concrete formulations met the specified durability requirements and provided a much
superior sustainable solution.
When the contract for Line 16-​1 was awarded, Eiffage Génie Civil immediately nego-
tiated a subcontract with segment producer Bonna Sabla. Then in the summer of 2018,
Eiffage initiated a $1.3 (€1.2) million research program to validate the performance of the
SFRC segments. This 18-​month campaign included:

• Design tests Formally approve the characterization studies

• Full-​s cale testing Tor Vergata University of Rome was selected as the
laboratory for the full-​s cale tests
• Fire tests Conducted at the CERIB research facility in Epernon
• Suitability tests Validate the concrete manufacturing process with a view to
obtaining approval for the SFRC mix design
• QC process established Both with fiber manufacturing/​s upply and in precast plant
manufacturing

Following this most extensive study in December 2019, the new SFRC segments were
approved for installation for the 2 km × 9.5 m (7.5 mile × 31.2 foot) outside diameter
drives for Lines 16 and 15. Manufacturing was at the same Bonna Sabla factory casting
the rebar reinforced segments. The low carbon mix design contains 40 kg/​m3 (67 lbs/​
CY) of high-​performance steel fibers glued in clips for homogenous distribution in the
mix. This quantity of fibers completely replaces the original 85 kg/​m3 (143 lbs/​CY) of
steel rebar cages for the same 2m (6.6 foot) wide × 9.5m (31.2 foot) outside diameter
segments.
The use of SFRC in a portion of the segments in the Paris Metro Line 16-​1 has pro-
vided substantial environmental benefits and lowered carbon footprint derived from the
following:

• The reduction of steel quantity required provided a corresponding reduction in

transport-​related CO2 emissions. Comparing the 40 kg/​m3 (67 lbs/​CY) of steel fibers
against the 85 kg/​m3 (143 lbs/​CY) of rebar yielded a potential material saving of more
than 50%.
• One truckload delivery of fibers to the manufacturing plant allows the production of
nearly 185 segments, compare with 60 equivalent segments per truckload of conven-
tional reinforcement.
• The smaller diameter of fibers enables toxic emissions from the primary steelmak-
ing industry to be reduced, as the resultant transformation provides wire no thicker
than 0.04 in (1 mm) in diameter. Wire drawing technology is not a major source of
emissions.
• The life cycle of SFRC is up to 300% more beneficial to the environment overall com-
pared with rebar reinforced concrete. See Table 5.7 below.

The production of structural precast concrete segments 100% reinforced with SFRC for
the permanent lining of Paris Metro Line 16-​1 large diameter tunnel is the first of such pas-
senger rail structure in France and Europe. Using SFRC for this type of structure is emerging
as the obvious choice from a technical, economic, and more so now sustainability perspec-
tive in conjunction with low-​carbon concretes. This groundbreaking project opens the door
to more SFRC segment projects in France and Europe and confirms that the use of SFRC
 Fiber reinforcement in precast concrete segments 335

Table 5.7 Environmental sustainability benefits of SFRC.

Environmental sustainability benefits of steel fiber reinforcing concrete (SFRC)
compared to traditional steel reinforcing bar

Equivalent CO 2
Material quantities quantities
Carbon
Item Operation Metric Imperial Metric Imperial footprint
A Data comparisons
1 Steel fiber production 1 kg 2.2 lb 0.88 kg 1.94 lb Carbon dioxide
(CO 2)
2 Rebar production 1 kg 2.2 lb 1.85 kg 4.08 lb Carbon dioxide
(CO 2)
B Comparison summary
1 SFRC segments have 50% less steel than rebar reinforced segments saving more than
5,000 tonnes (5,512 tons) of steel production carbon for 10 km (6.2 miles) of tunnel.
2 One truck can transport 24.2 tonnes (26.7 tons) of fiber per load compared with
17.85 tonnes (19.7 tons) per truck load of rebar.
3 The concrete chosen for the Paris Metro Line 16, Lot 1, fiber reinforced segments had
a low carbon footprint of 170 kg (375 lbs) CO 2 equivalent/​m 3 and reduces the carbon
weight of the steel in the segments by 90 kg (198 lbs) CO 2 equivalent/​m 3 or nearly
11,000 tonnes (12,125 tons) equivalent CO 2.

Figure 5.27 Carbon emission indicative calculation.

Note: Carbon emission indicative calculation as measured in weight of CO 2 equivalent per linear
meter of tunnel for steel fiber vs. rebar design for metro tunnel in Grand Paris 16.1.

for precast segments is the sustainable state of the art solution in France and around the
world. See Figure 5.27.

5.5.6.2 Montreal Metro Blue Line Extension

The Montreal Metro Blue Line Extension Project consists of construction of 6 km (3.7
miles) of tunnel, as well as five new underground stations. The owners’ design engineer,
336 Handbook of Precast Segmental Tunnel Lining Systems

AECOM, as part of a commitment to integrating sustainability best practices, performed a

study utilizing the Envision framework to evaluate alternatives to achieve a most sustain-
able infrastructure project. Based on the results of this study the TBM bored tunnel sections
will be lined with steel fiber reinforced precast concrete segments using low-​carbon supple-
mentary cementitious materials (SCM) concrete. In sections where the sequential excava-
tion method (SEM) will be utilized, a permanent sprayed concrete lining (PSCL) was chosen
as the optimum sustainable solution.
In the TBM tunnel sections lined with precast concrete segments, high-​performance steel
fiber with a dosage of 40 kg/​m3 (67 lbs/​CY) is designed as standalone reinforcement. The
reduction in segment thickness achieved with fibers can be primarily attributed to the con-
crete cover requirements of 60 to 75 mm (2.4 to 3.0 inches) on both intrados and extrados
rebar to ensure the durability against corrosion when designing according to Canadian code
CSA A23.1:19 (2019).
In contrast, when subjected to chloride exposure, corrosion in steel fiber reinforced con-
crete is limited to just a few millimeters from the surface, and nonetheless, does not lead
to spalling cracks and is not regarded as a durability issue. CO2 savings in the segments
is realized by replacing rebar with steel fibers as the quantity of steel required is 50% less
per cubic meter of concrete with fibers [40 kg/​m3 (67 lbs/​CY) vs. 80 kg/​m3 (134 lbs/​CY)].
Additionally, the CO2 equivalent factor for rebar is reported to be 1.85 vs. 0.88 for fibers.
The fiber reinforced segments can be reduced in thickness due to no requirement for cover
like rebar. This quantity of concrete savings also lowers the carbon footprint. Table 5.8
summarizes the results of the evaluation showing a reduction in total CO2 equivalent by
nearly 50% using SFRC with an optimized SCM concrete mix design.
In the sections of the tunnel using SEM, a baseline design, which consisted of an arc shape
section with cast-​in-​place (CIP) concrete lining using ordinary Portland cement (OPC) con-
crete mixture was proposed. The baseline design consists of three components. A 75 mm
(3 inch) temporary shotcrete, 400 mm (16 inch) CIP lining, and 320 mm (12.6 inch) thick
invert. To reduce carbon emissions, an alternative case was evaluated where the base design
temporary shotcrete and CIP arc were replaced with a low-​carbon steel fiber reinforced
shotcrete lining (PSCL). The total embodied carbon footprint per 1 m (3.3 foot) length of
the tunnel-​baseline design was calculated and shown in Table 5.9
The low carbon alternative design was then developed and evaluated. To achieve lower
CO2 equivalent the CIP arc is completely replaced with a reinforced shotcrete lining.
The components consist of 50 mm (2 inch) temporary shotcrete, 100 mm (4 inch) final

Table 5.8 Comparison of total embodied carbon of the tunnel –​ segmental lining alternatives.
Embodied CO 2 comparison of segments between rebar with OPC mix and SFRC with
SCM mix

Ring Ring Total CO 2

Ring Tunnel diameter-​ Segment diameter-​ Ring concrete CO 2 eq/​1 m Total
width length (OD) thickness (ID) volume volume eq/​m 3 tunnel CO 2 eq,
m m m cm m m3 m3 Kg ton ton
1.8 4,086 9.3 40–​b ase 8.5 20.1 45,698 601 6.7 27,470
W/​ OPC
35–​ 8.6 17.1 40,211 363 3.6 14,593
SFRC
W/​ SCM
 Fiber reinforcement in precast concrete segments 337

Table 5.9 Total embodied CO 2 per 1 meter length of tunnel –​ baseline design.

Embodied CO 2 calculation for baseline design in SEM tunnel

Volume, m 3 /​ CO 2 eq 1m
Thickness 1 m tunnel CO 2 eq/​m 3 length
Item Description mm m3 Kg tons
1 Temporary shotcrete 325 5.90 470.3 2.8
2 CIP arc concrete 400 7.03 513.9 3.6
3 Invert concrete 320 2.85 539.7 1.5
Total 7.9

shotcrete lining, and 320 mm (12.6 inch) thick invert. The shotcrete and invert concrete are
reinforced with the same hooked end, collated, high-​performance steel fibers. The concrete
mix design was modified to reduce the carbon emission while having the same strength
properties. Portland cement was replaced with limestone Portland cement and 27% of
the cementitious material was replaced with slag and silica fume [22% slag and 5% silica
fume which is considered moderate supplementary cementitious materials (SCM)]. This
mixture was proposed for both shotcrete and invert concrete. The total embodied carbon
footprint per 1 m (3.3 foot) length of the tunnel alternate design was calculated and shown
in Table 5.10.
In comparing the low-​carbon PSCL design utilizing low-​carbon concrete mix with the
CIP Baseline design the total CO2 emission saving is over 79%!

5.5.6.3 Project ­example –​Forrestfield Airport Link Project (Perth, Western

Australia)
The project comprises the design, construction and maintenance for 10 years of the
Forrestfield Airport Link, which will connect the eastern suburbs of Perth with the existing
suburban rail network as well as the airport. By 2022 the line is expected to generate 20,000
passenger trips on the network every day. It will also reduce road traffic and traveling times.
The precast concrete segmental lining was designed to meet the project’s concrete require-
ments of 120 year service life with a minimum water/​binder ratio of 0.35. Webuild and its
joint-​venture partner NRW Pty Ltd (80% of the joint-​venture is held by Webuild) were the
design build contractors for the project. The tunnel excavation diameter is 7.1 m (23.3 feet);
the segmental lining has an inside diameter of 6.1 m (20 feet), and thickness of segment is
300 mm (12 inch) with an average length of 1,600 mm (5.25 feet); the ring is made by uni-
versal type of five segments plus a key. The segments have a double gasket, which makes
them state-​of-​the-​art and still quite unique. The configuration of rings provides a minimum
tunnel radius of 300 m (984 feet).
The solution utilized for the manufacture of segmental linings for the project is based on a
hybrid system using steel fiber-​reinforced concrete and light steel reinforcement rebar cage.
In addition, micro-​polypropylene fibers are also incorporated within the concrete mix to
comply with the fire resistance requirements of the tunnel lining. Mix-​design was developed
based on triple blend cementitious materials, an accurately graded fine and coarse aggre-
gates with 20 mm (0.80 inches) maximum size as specified. This is to ensure that the seg-
mental linings are of highest quality with very low permeability and porosity, can withstand
any deterioration caused by external conditions and meet its design durability life span.
338 Handbook of Precast Segmental Tunnel Lining Systems

Table 5.10 Total embodied CO 2 per 1 m length of tunnel –​ low carbon design.

Embodied CO 2 calculation for optimized single shell shotcrete design

Volume, CO 2 eq 1 m
Thickness m 3 /​1 m tunnel CO 2 eq/​m 3 length
Item Description mm m3 Kg tons
1 Temporary shotcrete +​ 160 1.80 361 0.7
Shotcrete final lining
2 Invert concrete 320 2.85 361 1.0
Total 1.7

Design of segmental lining considered special type of high-​performance steel fibers

with a tensile strength of 1,800 N/​mm2 (260,000 psi). These fibers were designed for
durable and liquid-​tight structures and can perfectly combine with traditional steel
reinforcement. The production of concrete segments started with the mix design that
had steel fiber content of about 40 kg/​m3 (67.4 lbs/​yd3); however, during the early stages
of production, it was observed that the residual strength of concrete with 40 kg/​m3
(67.4 lbs/​yd3) of steel fiber provided much higher residual strength than that required
by design and specification. On this basis, a reduction of 5 kg/​m3 (8.4 lbs/​yd3) was
implemented during the succeeding production. Several trials were performed with cast
concrete beams according to standard BS EN 14651 (single trial was made with nine
concrete beams) with 35 kg/​m3 (60 lbs/​yd3). The characteristic values of these results
were more than the required serviceability state 5.08 MPa (737 psi) at CMOD1 and
5.28 MPa (765 psi) at CMOD3. The results have confirmed the assumptions for the
quantity reduction and the mix with 35 kg/​m3 (60 lbs/​yd3) of steel fibers content were
adopted in project works.
A LCA (life cycle assessment) evaluation per cubic meter of tunnel lining concrete was
performed to evaluate the hybrid reinforced segment versus a conventional reinforcement
only alternative. Figure 5.28 shows the results of this evaluation:
As can clearly be seen the savings realized from the use of high-​performance steel fibers
to replace a portion of the conventional rebar along with optimized low carbon concrete
provides large reductions in carbon footprint (Scaffidi, 2019).

5.5.7 Life cycle assessment and environmental product

declarations (EPDs)
“One Click LCA” is a global industry wide platform to evaluate alternatives and accelerate
decarbonization for construction projects. One Click LCA is a commercial automated life
cycle assessment (LCA) software that helps you calculate and reduce the environmental
impacts of your project. (Also available in a limited free version –​One Click LCA Planetary.)
The LCA is a set of standardized data. It quantifies a material's impacts on the environment
over its entire existence, from extraction of the raw materials required for its production up
to its end of life. This approach, combined with research into a low-​carbon solution, will
give new momentum to increased use of FRC in segmental linings. One Click LCA allows
the user to benchmark cradle-​to-​gate embodied carbon impacts and materials efficiency of
different design options, different materials, manufacturers, and products. One Click LCA
can be used to power embodied carbon and materials efficiency policies as well as individual
 Fiber reinforcement in precast concrete segments 339

Figure 5.28 Carbon emission indicative calculation.

Note: Life cycle assessment as measured in weight of CO 2 equivalent per cubic meter of seg-
mental tunnel lining for composite steel fiber +​ rebar vs. 100% rebar design.

design, construction and procurement decisions using material data provided through the
product’s environmental product declaration (EPD).
An environmental product declaration (EPD) is essentially an externally verified and
standardized description of the environmental profile of any product or material over its
lifetime. If you are a construction engineer or an architect, you can easily compare the
EPD data of the materials you are about to use and choose the most sustainable option.
The manufacturer through an EPD certifies their product on the issues of climate change,
carbon impact and environmental footprint. Having an EPD for a product does not neces-
sarily mean that the product is environmentally better than others, but it is a way to obtain
transparent information on the environmental impact of the product during its whole life-
time. Table 5.11 is an excerpt from an EPD for steel fibers manufactured by a leading global
producer.
The reduction in CO2 emissions on a precast segmental lining tunnel project is made
possible by modification of the concrete and further reduction by being able to replace the
rebar with steel fibers in a dosage that satisfied all the design requirements. On a per kilo-
gram (or pound) basis the embodied CO2 of conventional rebar and steel fibers previously
had been assumed the same. This example EPD shows that this is not a valid assumption
as the steel fibers as fiber production contributes less CO2 than rebar. Using the One Click
LCA and EPD data provided by the producer gives the owner/​engineer tools to optimize the
sustainability of a project.

5.6 COMBINED SOLUTIONS
Precast segments in service are typically mainly in compression and experience low bending
moments. The use of fibers only is the preferred solution, however sometimes there are
loads that are too high for the fibers alone to resist. Fibers are efficient for resisting tensile
stresses that are uniformly distributed. Bar reinforcement is efficient for resisting localized
higher tensile stresses. A combined solution is one where bar reinforcement is added to
provide additional localized capacity where fibers alone are not sufficient. The bars are
considered secondary reinforcement and the fibers are primary. The fibers provide superior
340 Handbook of Precast Segmental Tunnel Lining Systems

Table 5.11 Environmental product declaration type III ITB No.215/​2 021.

Environmental sustainability benefits of steel fiber reinforcing concrete (SFRC)
compared to traditional steel bar reinforcing

A Project application
1 Producers of steel fibers are used for concrete reinforcement and are an alternative
to steel mesh and bars. They are discontinuous, three-​d imensional and isotropic
reinforcement. The steel fibers bridge cracks at their small widths, distribute stresses
and increase the strength of concrete in a cracked state. Adding the adequate number
of fibers to the concrete plasticizes it, increasing its tensile and shear strength, impact
strength and fatigue resistance. Steel fibers for structural use are used for concrete
and mortar reinforcement for below applications: over ground applications (flooring,
building, civil engineering, etc.), underground applications (segmental linings for
tunneling etc.) and precast.
B Results interpretation
1 The environmental impact of the producers’ fibers [cradle to gate (grave) with options]
is largely dependent on the energy-​intensive production of steel (half product) on which
the manufacturer has a limited influence only. The carbon impact of steel production
(wire rods) in the product stage A1 is as high as 85%. The impact of the production line
A3 largely depends on the amount of electricity consumed by the manufacturing plant
(0.34 kWh/​k g of product). There are no significant emissions or environmental impacts
in the A3 production processes alone (partly gas combustion). The production process
itself does not have significant environmental impacts in the life cycle.
2 Interrogation of the LCA results show that the cradle-​t o-​g ate carbon (global warming
potential) impact of 1 kg of fiber production is 0.88 kg CO 2 equivalent. In comparison,
a ton of steel produced worldwide in 2019 emitted on average 1.85 tons of carbon
dioxide. The LCA results show that the cradle-​t o gate (grave) primary energy demand
of fossil fuel is equal to 9.4 MJ. This is due to the production of nuclear energy in
the factory’s region. The transport of raw materials from considerable distances is
optimized and not significant (0.007 kg CO 2/​k g).
3 Due to the high potential for recycling and reuse (95%), the products have noticeable D
module potential (beneficial to other product systems).

benefits of increased toughness, durability and efficient crack control compared to rebar
alone. The rebar is added to provide localized strength to resist a specific load case(s) where
the fiber only design either cannot meet the requirement, or it is determined a combined
solution is more efficient. Typically, combined solutions are applicable for larger-​diameter
tunnels, which have higher in service internal forces, soft ground conditions, high unbal-
anced loads or special loading cases such as at the start and end of tunnels, at locations of
cross passages, seismic fault zones, etc.

5.6.1 Hybrid solutions with reinforcing steel bar and welded wire mesh
The most common type of bar reinforcement utilized with the fiber is conventional reinfor-
cing steel bar or welded wire mesh. In a rebar +​mesh only segment design, the reinforcing
consists of three general groups of bars that provide specific functions:

• Transverse bars
• Top and bottom curved bars considered the main flexural reinforcement placed
perpendicular to the tunnel axis.
 Fiber reinforcement in precast concrete segments 341

• Longitudinal bars
• Top and bottom straight bars welded or tied to transverse bars and typically
designed as minimum amount of steel required for temperature and shrinkage.
• Joint bars
• Additional bars at both circumferential and radial joints to provide reinforcement
to resist bursting/​spalling forces at the joints.
• Circumferential joints typically consist of transverse curved bars with radial stir-
rups or ties (resembling a curved beam cage).
• In the radial direction not as robust typically consisting of a top and bottom bar
the same size as the other radial steel bars provided transverse resembling a ladder.
In a combined solution, the bars that are added to the fibers are typically the circumferential
joint type reinforcement and all other bars are replaced with SFRC as shown in Figures 5.29
and 5.30.

5.6.2 Hybrid solutions with GFRP bars

The primary benefit of using glass fiber reinforced polymer (GFRP) in place of conventional
reinforcing in a segment is there no risk of corrosion of the bar in the segment. However,
due to the low elastic modulus of the GFRP bar, its stand-​alone use can produce certain
undesired structural effects. In order to evaluate the potential synergistic effect of the use
of GFRP and with high-​performance steel fibers in tunnel segments, experimental full-​scale
testing was performed at the University of Rome Tor Vergata under Dr. Alberto Meda. In

Figure 5.29 Rebar reinforced concrete segment.

Note: View of a typical rebar reinforced concrete segment in a mold.

Figure 5.30 Hybrid solution for steel fiber reinforced concrete.

Note: Hybrid solution for steel fiber reinforced concrete (SFRC) +​ rebar using a circumferential
rebar cage.
342 Handbook of Precast Segmental Tunnel Lining Systems

Figure 5.31 GFRP rebar reinforced concrete segment.

Note: View of a typical combined solution using GFRP rebar +​ SFRC segment in a mold.

Figure 5.32 GFRP rebar cage.

Note: View of a typical circumferential GFRP rebar reinforcing cage.

order to evaluate the structural performance, both in terms of strength and crack width,
bending and point load tests were carried-​out on typical metro tunnel size segments. The
testing consisted of performing the tests on segments reinforced with high performance
SFRC only and on segments with the same SFRC plus a light GFRP cage (SFRC +​GFRP) as
shown in Figures 5.31 and 5.32.
Results of load deflection full scale tests showed that after an initial almost identical
elastic-​type response, the SFRC +​GFRP segment presented a peak load about 63% higher
 Fiber reinforcement in precast concrete segments 343

than the SFRC, 367 kN (41 tons) against 225 kN (25 tons) of the SFRC segment. Maximum
crack widths were measured at different load steps, and the results showed the SFRC +​
GFRP provided further reduction compared to the SFRC solution, of about 60%.
Point load testing was performed by applying three-​point loads on the segment, using the
same steel plates used by the TBM machine. Two loading cycles were performed, first with a
load level equal to 1,580 kN (178 tons) for each pad which corresponds to the service load-
ing thrust of the TBM rams. The second loading cycle to 2,670 kN (300 tons) for each pad,
which corresponds to the maximum thrust of the TBM. The addition of the GPRP cages at
the circumferential edges led to halve the crack width under the service load, and to reduce
it of about 37.5%, under the maximum thrust force. Furthermore, a reduction of the crack
width of about 33% was measured after the complete unloading.
The results of bending tests clearly show the synergic effects of the two materials (SFRC +​
GFRP reinforcement) by increasing the peak load and reducing the crack width. The results
of the point load test further confirm the effectiveness of the addition of a GFRP cage at
the circumferential edges to provide superior performance in regard to counteracting TBM
thrust forces.
Since tunnels can have different geotechnical conditions along their length, in general it
is best to define a SFRC solution that can cover the greatest part of the tunnel and adopt a
hybrid solution only in the remainder. Therefore, if a base SFRC only solution is viable for
the vast majority of the tunnel, the use of a GFRP or steel bar hybrid solution can be con-
venient, e.g. in cross-​passages, in shallower sections of tunnels and in areas where increased
performance is required.

5.6.3 Hybrid solution by circumferential post-​tensioning of the ring

SFRC segmental linings are extremely structurally efficient in compression. In hydraulic
conveyance tunnels in times of peak flow, sustained high internal pressures may occur. If
the internal pressures are high enough, they can put the lining into tension for which the
lining must be designed. An effluent outfall tunnel in Los Angeles that is being constructed
includes a portion of the alignment where at times the internal pressure will exceed the
external pressure. The solution that is being employed for the first time in North America
is SFRC segments with post tensioning of the ring in this zone. To offset the “hoop stress”
developed from internal pressure, a cable is inserted in a duct that runs the circumference of
the lining. Once tensioned, it creates compression in the concrete, so that the concrete itself
never goes into tension, even under the highest internal pressure.

5.7 NEW TYPES OF FIBERS –​ MACRO SYNTHETIC/​P OLYMERS

As discussed earlier in the chapter, the first use of a macro-​synthetic fiber reinforced con-
crete (MSFRC) to replace SFRC in precast segments in North America recently occurred in
the Blacksnake Creek CSO tunnel (St. Joseph, Missouri, United States). Significant research
and development over the last several years has been undertaken to increase the structural
performance of macro-​synthetic fibers in concrete. This has led to fibers that are capable of
replacing or being used in combination with conventional or GFRP bars in segments.
Not all tunnel applications are appropriate for the use of macro-​synthetic fibers because
of their low elastic modulus, low melting point and potential for creep failure if the FRC is
subjected to sustained load. However, applications, where the lining will be in compression,
not subject to fire, and a strain-​softening structural response is sufficient macro-​synthetic
344 Handbook of Precast Segmental Tunnel Lining Systems

fibers are being considered. Furthermore, current design guidelines such as ACI 544-​7R
or the fib Model Code 2010 are valid for steel fiber only. Synthetic fibers are susceptible
to other types of degradation, as is mentioned in the EN 14889 standard. Also, the long-​
term effects of exposure to ultraviolet light, chemicals, microbes, is not well documented.
The example application mentioned earlier in this chapter is a small-​diameter combined
sewer overflow (CSO) tunnel where the fibers were able to meet the structural performance
requirements with the benefit that the segments were free of any metallic reinforcement
eliminating any risk of corrosion.

5.8 TUNNEL LINING REFERENCES USING FRC PRECAST SEGMENTS

Since 1982, the use of FRC in precast segmental linings of TBM bored tunnel projects
around the world has steadily increased. The following project reference list validates that
FRC, predominantly employing steel fibers, has become the preferred material in the pro-
duction of tunnel precast segments. The project listing in Table 5.12 shows the year the

Table 5.12 Reference projects using FRC in precast tunnel segments.

This table shows that since 1982, the use and acceptance of FRC (primarily SFRC)
in precast concrete tunnel segments around the globe has increased significantly

Fibers
Dia.
Year Tunnel or project name Type Country (m) Type(s) Only
1982 Naples Metro–​M etrosud Subway Italy 5.8 Steel Yes
1989 Fanaco, Sicily Water Italy 3.0 Steel Yes
1990 Munich Water Tunnels Sewage Germany 3.4 Steel Yes
1992 Metro Subway Tunnel Napoli Metro Italy 7.3 Steel Yes
1992 Water Tunnel Hydro South Africa 4.6 Steel No
1993 Heathrow Airport Baggage Tunnel Utility UK 4.5 Steel Yes
1994 Heathrow Express Rail UK 5.7 Steel Yes
1995 Naples Metro Metro Italy 5.8 Steel Yes
1995 Lesotho Highlands Water South Africa 4.5 Steel Yes
1996 Essen Metro Metro Germany 7.3 Steel Yes
1996 Kandertal Exploratory Tunnel, Faido Rail Switzerland 5.0 Steel Yes
1996 Portsmouth Wastewater Transfer Hydro UK 3.0 Steel Yes
1996 Wanley Island Wastewater Tunnel Hydro UK 2.9 Steel Yes
1996 Jubilee Line Extension (cont103) Metro UK 4.5 Steel Yes
1997 High Wycombe Water Transfer Hydro UK 3.4 Steel Yes
1998 Cigar Lake Uranium Mine Mining Canada 4.2 Steel Yes
1998 Hachinger, Munich Water Germany 2.2 Steel Yes
1998 Bromborough Wastewater Tunnel Utility UK 2.9 Steel Yes
1999 Dunfermline Duplicate Sewer Sewage UK 2.9 Steel Yes
1999 2nd Heinenoord, Rotterdam Road Netherlands 7.6 Steel Yes
 Fiber reinforcement in precast concrete segments 345

Table 5.12 (Continued)

Fibers
Dia.
Year Tunnel or project name Type Country (m) Type(s) Only
1999 Torbay Wastewater Tunnel Utility UK 3.0 Steel Yes
2000 Esperanza (Trasvases Manabi) Hydro Ecuador 3.5 Steel Yes
2000 Oenzberg Rail Tunnel Rail Switzerland 12.0 Steel Yes
2000 Sörenberg Gas Tunnel Utility Switzerland 4.3 Steel Yes
2000 CTRL Rail Tunnels–​L ots 220 and 240 Rail UK 6.0 Steel Yes
2001 Manapouri Second Tailrace Tunnel Hydro New Zealand 10.1 Steel Yes
2002 Marina Line C825 Metro Singapore 6.4 Steel Yes
2002 Dartford Cable Tunnels Utility UK 2.8 Steel Yes
2003 Genova Metro–​P rincipe/​C ari/​G razie Metro Italy 4.9 Steel No
2003 Bankside Cable Tunnel Utility UK 4.0 Steel Yes
2003 CTRL Rail Tunnel, Contract Lot 103 Rail UK 7.2 Steel Yes
2003 Margate Sewer Tunnel; Sewage UK 3.0 Steel Yes
2003 Transco Gas Pipeline Utility UK 2.4 Steel Yes
2004 Hofoldinger Stollen Hydro Germany 3.1 Steel Yes
2004 Croydon Cable Tunnel Utility UK 3.0 Steel Yes
2005 Metro Barcelona Line 9 Metro Spain 12.0 Steel No
2005 Metro Madrid, Linea 3 Extension Metro Spain 8.6 Steel No
2005 Metro Madrid, Linea 5 Extension Metro Spain 8.6 Steel No
2005 Metro Norte Madrid Metro Spain 9.4 Steel No
2005 Dockland Light Rail Extension Rail UK 5.3 Steel Yes
2005 Big Walnut Sewer, Columbus, OH Sewage USA 3.7 Steel No
2005 Metro Expressway Shinjuku Route Road Japan 10.9 Steel No
2005 Piccadilly Line and Heathrow Express Metro UK 5.7 Steel Yes
Ext’n
2006 Circle Line C856 Metro Singapore 6.4 Steel Yes
2006 Heathrow Piccadilly Extension Metro UK 4.5 Steel Yes
2006 Thames Water Ring Main Brixton Utility UK 3.0 Steel Yes
2006 Thames Water Ring Main River Utility UK 3.0 Steel Yes
2006 Heathrow–​S WOT Water UK 2.9 Steel Yes
2006 Brightwater Sewer Tunnel–​C entral, Sewage USA 4.7 Steel Yes
Seattle
2006 Brightwater Sewer Tunnel–​E ast, Sewage USA 5.1 Steel Yes
Seattle
(Continued)
346 Handbook of Precast Segmental Tunnel Lining Systems

Table 5.12 (Continued)

Fibers
Dia.
Year Tunnel or project name Type Country (m) Type(s) Only
2006 Brightwater Sewer Tunnel–​West, Sewage USA 4.0 Steel Yes
Seattle
2006 San Vicente Water Tunnel Hydro USA 4.0 Steel Yes
2007 North Sewerage Project Melbourne Hydro Australia 2.5 Steel No
2007 Lötschberg Base Tunnel Temp Switzerland 4.5 Steel Yes
2007 Belfast Sewer Tunnels Hydro N. Ireland 2.9 Steel Yes
2007 Metro Napoli, Linea 1 Metro Italy 6.8 Steel No
2008 Gold Coast Desalination Plant Hydro Australia 2.8 Steel Yes
Tunnels
2008 North-​S outh Bypass Brisbane Road Australia 12.4 Steel Yes
2008 Hobson Bay Sewer Hydro New Zealand 3.7 Steel Yes
2008 AVE Frontera Espanola-​F rancia Rail Spain 9.5 Steel No
2008 Metro Madrid, Linea 11 Metro Spain 6.6 Steel No
2008 Tunel Oliola Irrigation Hydro Spain 3.0 Steel Yes
2008 Heathrow Airport Baggage Tunnel 2 Utility UK 5.3 Steel Yes
2009 Adelaide Desalination Hydro Australia 2.8 Steel Yes
2009 Melbourne Desalination Hydro Australia 4.5 Steel Yes
2009 Melbourne Sewer Replacement Hydro Australia 1.8 Steel No
2009 Sydney City West Cable Tunnel Utility Australia 3.2 Steel Yes
2009 Sydney Desalination Plant Tunnels Hydro Australia 3.2 Steel Yes
2009 Yellow Line, Sao Paolo Metro Brazil 8.4 Steel No
2009 Genoa Subway Metro Italy 5.4 Steel No
2009 Metropolitana di Roma, Linea C Metro Italy 7.0 Steel No
2009 Rosedale Wastewater Sewer Hydro New Zealand 3.2 Steel Yes
2009 AVE Frontera Malaga-​Valencia Rail Spain 8.4 Steel No
2009 Cercanias Madrid–​N avalcarnero Rail Spain 6.5 Steel No
2009 West Ham Sewer Tunnel, London Hydro UK 2.8 Steel Yes
2009 Docklands Light Railway Extension, Rail UK 5.3 Steel Yes
London
2009 Brightwater Sewer, Phase 1, Seattle Sewage USA 4.0 Steel No
2009 Copenhagen Heating Tunnel Water Denmark 4.2 Steel Yes
2009 Harefield Gas Tunnel Gas UK 2.6 Macro-​ Yes
PP
2010 Airport Link /​ North Busway, Road Australia 12.4 Steel Yes
Brisbane
 Fiber reinforcement in precast concrete segments 347

Table 5.12 (Continued)

Fibers
Dia.
Year Tunnel or project name Type Country (m) Type(s) Only
2010 South East Collector Sewer Tunnel Sewage Canada 3.8 Steel Yes
2010 Naples Metro, Linea 6 Metro Italy 6.8 Steel No
2010 Monte Lirio–​Pando–​E l Aldo Tunnel Hydro Panama 3.2 Steel Yes
2010 Sabadell Road Tunnel Road Spain 6.6 Steel No
2010 East Kent Tunnel, Phase 2 Road UK 3.1 Steel No
2010 South Brighton and Hove Water Hydro UK 2.9 Steel Yes
Tunnel
2010 Brightwater Sewer, Phase 2, Seattle Sewage USA 4.0 Steel No
2010 San Francisco Bay Utility Tunnel Hydro USA 3.7 Steel Yes
2010 Fortanta–​Trinitat Interconnection Water Spain 5.2 Steel No
2010 Keio Line Rail Japan 6.7 Steel No
2010 FGC Terrassa Rail Spain 6.0 Steel No
2010 Clem Jones–​C lem 7 Road Australia 11.2 Steel No
2011 Legacy Way (Northern Link) Road Australia 11.3 Steel No
2011 Mill Woods Tunnel, Edmonton Sewage Canada 4.0 Steel Yes
2011 ADIF AVE Serrano, Madrid Rail Spain 11.5 Steel No
2011 Crossrail C300 West Running Rail UK 6.0 Steel Yes
Tunnels
2011 National Grid Cable Tunnels 1 Utility UK 4.4 Steel No
2011 National Grid Cable Tunnels 2 Utility UK 3.6 Steel Yes
2011 National Grid Cable Tunnels 3 Utility UK 3.6 Steel Yes
2011 Preston Storm Water Tunnel Hydro UK 2.9 Steel Yes
2011 Portland Eastside CSO Sewage USA 6.7 Steel Yes
2011 Izumi–​O tsu Water Japan 1.8 Steel No
2011 Metropolitan Expressway Road Japan 13.4 Steel No
2012 Sydney City East Cable Tunnel Utility Australia 3.5 Steel Yes
2012 Metro Linea 3, Tramo 3 Metro Brazil 8.8 Steel +​ Yes
MPP
2012 Metro Linea 3, Tramo 7 Metro Brazil 8.8 Steel +​ Yes
MPP
2012 York–​S padina Subway Extension, Metro Canada 6.1 Steel Yes
Toronto
2012 El Torito Hydro Hydro Costa Rica 7.4 Steel Yes
2012 Pando Hydro Project, Chiriquí Water Panama 3.0 Steel Yes
(Continued)
348 Handbook of Precast Segmental Tunnel Lining Systems

Table 5.12 (Continued)

Fibers
Dia.
Year Tunnel or project name Type Country (m) Type(s) Only
2012 Downtown Line 3, Contract C933 Metro Singapore 6.4 Steel Yes
2012 Metro Barcelona, Line 9 Metro Spain 12.0 Steel No
2012 Well Point Close Cable Tunnel, Utility UK 1.8 Steel Yes
2012 Blue Plains Tunnel, Washington, DC Sewage USA 7.7 Steel Yes
2012 Euclid Creek Sewer, Cleveland, OH Sewage USA 7.9 Steel Yes
2012 Midosuji Utility Utility Japan 5.1 Steel No
2012 Sagami Line Road Japan 11.8 Steel No
2013 Grosvenor Coal Mine Mining Australia 7.7 Steel Yes
2013 Santos GLNG Gas Transfer Tunnel Utility Australia 3.4 Steel Yes
2013 Bibiheybet Sewage Azerbaijan 3.3 Steel Yes
2013 Güneşli-​D ernegül Sewage Azerbaijan 2.9 Steel Yes
2013 Xirdalan Sewage Azerbaijan 2.8 Steel Yes
2013 Xocahesen Sewage Azerbaijan 3.2 Steel Yes
2013 Klang Valley MRT–​S BK Line Metro Malaysia 6.4 Steel +​ Yes
MPP
2013 Klang Valley MRT–​S BK Line Metro Malaysia 6.4 Steel +​ Yes
MPP
2013 Waterview Connection, Auckland Road New Zealand 13.0 Steel No
2013 STEP Strategic Tunnel Program, T01 Sewage UAE 2.8 Steel No
2013 STEP Strategic Tunnel Program, T02 Sewage UAE 5.5 Steel No
2013 STEP Strategic Tunnel Program. T03 Sewage UAE 6.0 Steel No
2013 Bridlington Outfall Tunnel Sewage UK 3.6 Steel Yes
2013 Central Subway, San Francisco CA Metro USA 5.4 Steel No
2013 El Alto Water Panama 5.8 Steel Yes
2013 Asada Trunk Line Sewage Japan 4.6 Steel No
2013 Koishikawa Kasen Rail Japan 6.7 Steel No
2013 Oi-​A riake Cable Tunnel Power Japan 4.0 Steel No
2014 Sydney North West Rail Link Rail Australia 6.5 Steel +​ No
MPP
2014 Sao Paulo Metro, Linea 6 Metro Brazil 8.4 Steel +​ Yes
MPP
2014 Evergreen LRT Line, Vancouver, BC Metro Canada 5.4 Steel +​ Yes
MPP
2014 Abu Hamour, Doha Sewage Qatar 3.7 Steel +​ Yes
MPP
 Fiber reinforcement in precast concrete segments 349

Table 5.12 (Continued)

Fibers
Dia.
Year Tunnel or project name Type Country (m) Type(s) Only
2014 Green Line, Doha Metro Metro Qatar 6.2 Steel +​ Yes
MPP
2014 Thomson Line T206 Metro Singapore 6.4 Steel +​ Yes
MPP
2014 Safety Tunnel Giessbach Road Switzerland 5.1 Steel Yes
2014 Anacostia River CSO, Washington, Sewage USA 7.0 Steel Yes
DC
2014 Lee Tunnel (Tideway) Segments Sewage UK 7.2 Steel Yes
2014 Wehrhahn Tunnel Metro Germany 8.3 Steel Yes
2015 Thomson Line, Contract T206 Metro Singapore 6.4 Steel +​ Yes
MPP
2015 Thomson Line, Contract T207 Metro Singapore 6.4 Steel +​ Yes
MPP
2015 Klang Valley MRT, SBK Line Metro Malaysia 6.0 Steel +​ No
MPP
2015 Sao Paulo Metro, Linea 6 Metro Brazil 10.6 Steel +​ Yes
MPP
2015 Liantang Tunnel Road Hong Kong 14.1 Steel No
2015 Doha Metro, Red Line South Metro Qatar 6.2 Steel Yes
2015 Sanierungstunnel Belchen Road Switzerland 13.5 Steel No
2015 Farnworth Rail Tunnel Rail UK 8.0 Steel No
2015 Dugway Storage Tunnel, NEORSD, Sewage USA 7.9 Steel Yes
Cleveland
2015 First Street CSO, Washington, DC Sewage USA 6.1 Steel Yes
2015 LADWP, RSC 5 and 6 Water USA 3.1 Steel Yes
2016 Doha Metro, Red Line North Metro Qatar 6.2 Steel Yes
2016 Doha Metro Gold Line Metro Qatar 6.2 Steel Yes
2016 Hanlan Water Tunnel, Peel, Ontario Sewage Canada 3.6 Steel Yes
2016 Ohio Canal Interceptor, Akron, OH Sewage USA 9.1 Steel Yes
2016 Blacklick Creek Sanitary Sewer, Sewage USA 3.1 Steel Yes
Columbus, OH
2016 Centennial Parkway Sewer–​S hafts Sewage Canada 3 to Steel Yes
8.0
2016 Quito Metro Metro Ecuador 9.1 Steel No
2016 Nice Tramway, Line 2 Metro France 9.3 Steel No
2016 Paris Metro, Line 14 (prolongation) Metro France 9.3 Steel No

(Continued)
350 Handbook of Precast Segmental Tunnel Lining Systems

Table 5.12 (Continued)

Fibers
Dia.
Year Tunnel or project name Type Country (m) Type(s) Only
2016 Follo Line, Oslo Rail Norway 8.8 Steel No
2016 Northern Line Extension Metro UK 5.3 Steel Yes
2016 Shieldhall Sewer Tunnel Sewage UK 5.2 Steel No
2017 Klang Valley MRT, SSP Line Metro Malaysia 5.8 Steel +​ No
MPP
2017 Kasai Bashi TEPCO Cable Tunnel Metro Japan 6.0 Steel +​ No
MPP
2017 Thames Tideway West Tunnel Sewage UK 6.6 Steel Yes
2017 Ottawa CSST, Ottawa, Ontario Sewage Canada 3.7 Steel Yes
2017 South Hartford CSO, Hartford Sewage USA 6.1 Steel Yes
2017 Expolink 2020, Dubai, UAE Metro Dubai 8.5 Steel Yes
2017 CMC Catania Tunnel Metro Italy 9.6 Steel No
2017 Forrestfield Airport Link Rail Australia 7.0 Steel No
2018 Kemano T2 Tunnel, Kitmat, British Hydro Canada 6.5 Steel Yes
Columbia
2018 3R Port, Fort Wayne, IN Sewage USA 5.8 Steel Yes
2018 CBBT–​Parallel Thimble Shoals Tunnel Road USA 12.8 Steel Yes
2018 Mumbai Metro, Line 3–​H CC Metro India 6.7 Steel No
2018 Thames Tideway East Sewage UK 7.8 Steel +​ Yes
MPP
2018 Circle Line 6, Lot C882 Metro Singapore 6.7 Steel +​ Yes
MPP
2018 Circle Line 6, Lot C885 Metro Singapore 6.7 Steel +​ Yes
MPP
2018 North East Line Extension, Lot C715 Metro Singapore 8.3 Steel +​ Yes
MPP
2018 Deep Tunnel Sewer System 2, Lot Sewage Singapore 6.0 Steel Yes
T07
2019 Roma Metro, Linea C Metro Italy 6.7 Steel Yes
2019 Sydney Metro Tunnel Metro Australia 6.5 Steel No
2019 Doan Valley NEORSD CSO, Sewage USA 7.9 Steel Yes
Cleveland
2019 Tideway West /​ Frogmore Street Sewage UK 4.0 Steel Yes
Tunnel
2019 Blacksnake CSO, St. Joseph, Missouri Sewage USA 2.7 Macro Yes
PP
2019 Silicon Valley Clean Water, Palo Alto, Sewage USA 4.8 Steel Yes
CA
 Fiber reinforcement in precast concrete segments 351

Table 5.12 (Continued)

Fibers
Dia.
Year Tunnel or project name Type Country (m) Type(s) Only
2019 Westerly Storage Tunnel Sewage USA 7.6 Steel Yes
2020 Cowell Bypass Tunnel Sewage Canada 6.3 Steel Yes
2020 Ashbridges Bay Outfall Tunnel, Sewage Canada 7.0 Steel Yes
Toronto
2022 Réseau Express Métro (REM), Metro Canada 7.4 Steel +​ Yes
Montréal MPP
2020 I-​7 5 Modernization Tunnel Sewage USA 4.4 Steel Yes
2020 Bergen Point Outfall Replacement Sewage USA 3.8 Steel Yes
Tunnel
2020 Second Narrows Tunnel, Vancouver Hydro Canada 6.5 Steel Yes
2020 Grande Paris Metro–​L ine 16-​1 Metro France 8.7 Steel Yes
2021 Hampton Roads Tunnel, Norfolk, VA Road USA 14.0 Steel No
2021 JWPCP LA Outfall Tunnel, Los Sewage USA 6.5 Steel +​ No
Angeles, CA PT
2021 Annacis Island Phase 1 Tunnel, Water USA 4.2 Steel Yes
Vancouver, BC
2022 Scarborough Subway Tunnel, Metro Canada 11.9 Steel +​ Yes
Toronto, Ontario MPP
2022 Eglinton Subway Tunnel, Toronto, Metro Canada 6.6 Steel +​ Yes
Ontario MPP
2022 Fairbanks Silverthorn Tunnel, Sewage Canada 4.5 Steel Yes
Toronto, Ontario
2022 Broadway Subway Tunnel, Vancouver Metro Canada 6.0 Steel +​ Yes
BC MPP
2022 Pawtucket CSO, Pawtucket, RI Sewage USA 9.1 Steel Yes
2022 Ship Canal Tunnel, Seattle, WA Sewage USA 5.7 Steel Yes
2022 AlexRenew Tunnel, Alexandria, VA Sewage USA 4.5 Steel Yes
2022 Lower Olentangy Tunnel, Columbus, Sewage USA 3.7 Steel Yes
OH
2022 Shoreline Storage Tunnel, Cleveland, Sewage USA 7.9 Steel Yes
OH
2022 Snowy Hydro, Cooma, NSW Hydro Australia 10.6 Steel No
2022 Metro New Airport Link, Sydney Metro Australia 6.6 Steel Yes
2022 West Gate Tunnel, Melbourne, Road Australia 16.6 Steel Yes
Victoria
2022 Cross River Rail Twin Tunnels, Metro Australia 6.6 Steel Yes
Brisbane

(Continued)
352 Handbook of Precast Segmental Tunnel Lining Systems

Table 5.12 (Continued)

Fibers
Dia.
Year Tunnel or project name Type Country (m) Type(s) Only
2022 Thomson East Coast Line T-​3 16, Metro Singapore 5.8 Steel +​ Yes
Singapore MPP
2022 Cross Island Phase 1, Contract Metro Singapore 6.3 Steel +​ Yes
CR109 MPP
2022 Cross Island Phase 1, Contracts Metro Singapore 6.3 Steel +​ Yes
CR115 & CR116 MPP
2022 Tuas Water Reclamation Plant to Water Singapore 6.0 Steel Yes
Jurong Island
2023 Metro Western Package, Sydney Metro Australia 6.6 Steel Yes
2023 Metro Central Package, Sydney Metro Australia 6.6 Steel Yes
2023 Metro Eastern Package, Sydney Metro Australia 6.6 Steel Yes
2023 North-​E ast Link, Melbourne Road Australia 16.6 Steel Yes
2023 Integrated Pipeline Tunnel Section Sewage USA 3.5 Steel Yes
19, Dallas, TX

tunnel was built, name of the tunnel, type or function of the tunnel and the country where
the tunnel is located. Further, the fiber type(s) used in the segments are shown as fol-
lows: “Steel”, indicating steel fibers were used as primary reinforcement; “Steel +​MPP”,
indicating steel fibers were used as the primary reinforcement and micro-​polypropylene
fibers were added for passive fire protection; and “macro-​PP”, indicating a higher per-
formance macro-​synthetic polypropylene fiber was used as primary reinforcement. The
last column indicates if the segment was reinforced with fibers only (fibers only? =​Yes), or
used with conventional reinforcing steel bars or welded wire mesh as a combined solution
(fibers only? =​No).

BIBLIOGRAPHY
For preparation of this chapter, the following references were used, which are grouped into three cat-
egories of published codes, industry guidelines and technical publications.

PUBLISHED CODES
ACI Committee 533. (2020). ACI 533.5R-​ 20: Guide for Precast Concrete Tunnel Segments.
Farmington Hills, MI: American Concrete Institute.
ACI Committee 544. (2016). ACI 544.7R-​16: Report on Design and Construction of Fiber-​Reinforced
Precast Concrete Tunnel Segments. Farmington Hills, MI: American Concrete Institute.
ACI Committee 544. (2017). ACI 544.2R-​17: Report on the Measurement of Fresh State Properties
and Fiber Dispersion of Fiber-​Reinforced Concrete. Farmington Hills, MI: American Concrete
Institute.
ACI Committee 544. (2017). ACI 544.9R-​17: Report on Measuring Mechanical Properties of Hardened
Fiber-​Reinforced Concrete. Farmington Hills, MI 48331: American Concrete Institute.
 Fiber reinforcement in precast concrete segments 353

ACI Committee 544. (2018). ACI 544.4R-​18: Guide to Design with Fiber-​Reinforced Concrete.
Farmington Hills, MI.: American Concrete Institute.
ASTM C09.42 Subcomittee on Fiber-​Reinforced Concrete. (2020). ASTM C1609 /​C1609M–​19a–​
Standard Test Method for Flexural Performance of Fiber-​Reinforced Concrete (Using Beam
With Third-​Point Loading). West Conshohocken, PA,: ASTM International.
ASTM A820 /​A820M–​16–​Standard Specification for Steel Fibers for Fiber-​Reinforced Concrete.
ASTM International.
ASTM C1609 /​C1609M–​19a–​Standard Test Method for Flexural Performance of Fiber-​Reinforced
Concrete (Using Beam With Third-​Point Loading). ASTM International.
British Standards Institute. (2008). BS EN 14651:2005–​Test Method for Metallic Fibre Concrete–​
Measuring the Flexural Tensile Strength (Limit of Proportionality (LOP), residual) (+​A1:2007).
London: British Standards Institution.
British Standards Institute. (2006). BS EN 14889-​1:2006–​Fibres for Concrete. Steel fibres. Definitions,
Specifications and Conformity. London: British Standards Institution.
ISO/​TC 17/​SC 17 Steel Wire Rod and Wire Products. (2013). ISO 13270:2013–​Steel Fibres for
Concrete –​Definitions and Specifications. Geneva, Switzerland: ISO, the International
Organization for Standardization.

INDUSTRY GUIDELINES
National Precast Concrete Association (NPCA). (20 Jan 20). NPCA Quality Control Manual For
Precast Concrete Plantsm, 14th Edition. Carmel, IN USA 46032: National Precast Concrete
Association (NPCA).
Precast/​Prestressed Concrete Institute. (1999). MNL-​116-​99: Manual for Quality Control for Plants
and Production of Structural Precast Concrete Products, 4th Edition. Chicago, IL, USA: Precast/​
Prestressed Concrete Institute.

TECHNICAL PUBLICATIONS
Allen, C. (October 2009). Fibre Decider; Tunnels and Tunnelling International. pp. 35–​37.
Allen, C.. (November 2021). Low Carbon Concrete for Shafts and Tunnel Linings–​Rule of Thumb.
Tunnels & Tunnelling–​The Official Magazine of the British Tunnelling Society, pp. 28–​33.
Bakhshi, M. (2015). Design of Segmental Tunnel Lining for Serviceability Limit State, AASHTO
SCOBS T-​ 20: Technical Committee for Tunnels April 2015 Meeting. Saratoga Springs,
New York.
Bakhshi, M., and Nasri, V. (2017). Water Infiltration and Crack Control for Tunnel Concrete Lining.
Proceedings of the World Tunnel Congress 2017–​Surface Challenges–​Underground Solutions
(pp. 1–​10). Bergen, Norway: ITA–​AITES.
Bekaert-​Maccaferri Underground Solutions. (2018). Info Sheet: What About Stray Currents With
Steel-​Fibre Reinforced Concrete. Ghent, Belgium: Bekaert-​Maccaferri Underground Solutions.
Bischoff, Jean Luc, Benoit de Rivaz, Dieter Hansel, and Roland Herr. (February 2022). Steel-​Fibre-​
Reinforced Segments Improve Carbon Footprint. Tunnel–​Official Journal of STUVA, No. 02/​
2022.
de Rivaz, Benoit. (2016). Design Approach Assisted by Test. Ghent, Belgium: Bekaert-​Maccaferri
Undergound Solutions.
de Rivaz, Benoit. (2019). Characteristics of Steel and Polymer Based Fibre Concrete. Zwevegem,
Belgium: NV Bekaert SA.
de Rivaz, Benoit. (2020). Tunneling Applications: Concrete Segment Reinforcement. Zwevegem: NV
Bekaert SA.
de Rivaz, Benoit. (2022). Low Carbon Tunnel and Shaft Linings–​White Paper. Zvergem, Belgium:
Bekaert.
354 Handbook of Precast Segmental Tunnel Lining Systems

de Rivaz, Benoit. (March 2022). Moving to Low-​Carbon Linings. Tunnels & Tunnelling International,
the official magazine of the British Tunnelling Society, pp. 40–​44.
di Prisco, M., Colombo, M., and Dozio, D. (2014). Fibre-​Reinforced Concrete in fib Model Code 2010;
Principles, Models and Test Validation. Structural Concrete 14 (2013), No. 4, pp. 342–​361.
Edvardsen, C. (October 2010). The Consultant's View on Service Life Design. 2nd International
Symposium on Service Life Design for Infrastructure (pp. 249–​ 264). Delft, The
Netherlands: RILEM–​ International Union of Laboratories and Experts in Construction
Materials, Systems and Structures.
fib Special Activity Group 5. (2010). Bulletin 55-​ 56: Model Code 2010–​First Complete Draft.
Lausanne: International Federation for Structural Concrete (fib).
fib Working Party 1.4.1 Tunnels in Fibre-​Reinforced Concrete. (2017). fib Bulletin 83 Precast Tunnel
Segments in Fibre-​Reinforced Concrete State-​of-​the-​Art Report. (A. Meda, Editor) Lausanne,
Switzerland: Fédération International du Béton (fib).
Gerhard Vitt. (2011). Understanding Steel Fibre Reinforced Concrete: Dramix (Version 1.0 ed.).
Zwevegem, Belgium: NV Bekaert SA.
ITA Working Group 2–​ Research; Bakhshi, M., Nasri, V., Main Authors. (April 2019). ITA
Report No.22–​Guidelines for the Design of Segmental Tunnel Linings. Longrine, Avignon,
France: AITES/​ITA.
ITA Working Group 2–​Research; Tiberti, G., Chiniottri, E., Pizzari, G., Main Authors. (April 2016).
ITA Report No 16–​ Twenty Years of FRC Tunnel Segments Practice: Lessons Learnt and
Proposed Design Principals, Avignon, France: AITES/​ITA.
ITAtech Activity Group Support. (2016). ITAtech Report No.7; ITAtech Guidance for Precast Fibre
Reinforced Concrete Segments–​Volume 1: Design Aspects. Longrine: International Tunnelling
and Underground Space Association.
ITAtech Activity Group Support. (2018). ITAtech Report No.9: Guideline for Good Practice of
Fibre Reinforced Precast Segment–​Volume 2: Production Aspects. Longrine: The International
Tunnelling and Underground Space Association.
Jarast, P., Bakhshi, M., and Nasri, V. (2023), Carbon Footprint Reduction in Montreal Blue Line
Extension: RETC Proceedings (pp. 580–​588). SME.
Latson, J. (22 June 15). The Burning River That Sparked a Revolution. Retrieved from Time.
Com: https://​time.com/​3921​976/​cuyah​oga-​fire/​
Meda, A., and Rinaldi, Z. (2015). Tests on Precast Tunnel Segment in Concrete Newly High
Tensile Strength Steel Fibers Dramix 4D 80/​60BG. University of Rome "Tor Vergata", Civil
Engineering. Rome: TERC–​Tunnelling Engineering Research Center.
Meda, A., and Rinaldi, Z. (July 2017). Beam Tests on Fiber Reinforced Concrete According to EN
14651 and ASTM C1609. Rome, Italy: University of Rome–​Tor Vergata.
Meda, A., Rinaldi, Z., Spagnuolo, de Rivaz, B., and Giamundo, N. (January 2019). Hybrid Precast
Tunnel Segments in Fiber-​Reinforced Concrete with Glass Fiber Reinforced Bars. Tunnelling
and Underground space Technology 86, pp. 100–​112.
Meson, V. M. (2019). Durability of Steel Fibre Reinforced Concrete in Corrosive Environments–​Ph.D.
Thesis. Kongens Lyngby, Denmark: Department of Civil Engineering, Technical University of
Denmark.
NV Bekaert SA. (June 2021). Environmental Product Declaration Type III ITB No. 215/​2021.
Zwevegem, Belgium: Bekaert.
Pascal Guedon (Arcadis) Working Group Leader. (Juillet/​Août 2013). Recommendation of AFTES
No.GT38R1A1–​Design, Dimensioning and Execution of Precast Steel Fibre Reinforced
Concrete Arch Segments. Tunnels et Espace Souterrain–​No. 238, pp. 312–​324.
Rack, J. (20 August 14). A Brief History of Sustainability. Retrieved from The World Energy
Foundation: https://​thewo​rlde​nerg​yfou​ndat​ion.org/​a-​brief-​hist​ory-​of-​sus​tain​abil​ity/​
Rossi, P. C.-​P. (18 April 16). SFRC Durability in a Chlorinated Environment. Tunnel Talk (Direct by
Design), pp. 2.
 Fiber reinforcement in precast concrete segments 355

Scaffidi, S.C., Salini Impregilo S.p.A., Milano, Italy, Anders, A., Porto, S., and Torres, E., SI-​NRW JV,
Redcliffe, Perth, Western Australia, Forrestfield Airport Link Project, Perth, Western Australia;
Precast Concrete Segmental Lining, 2019. ITA-​AITES World Tunnel Congress, Naples, Italy.
Tang, K. (March 2020). Corrosion of Steel Fiber Subjected to Stray Current Interference. ACI
Materials Journal, pp. 99–​112.
Tunnel Talk Reporting. (25 April 19). TunnelTech: First Use of Macro-​Synthetic Fiber Reinforced
Concrete in Segments in the USA. TunnelTalk.
TunnelTalk Reporting. (March 2015). TunnelTech: Lab Comparison of Steel vs. Synthetic FRC
SegmentsTests and Results; by University of Rome Tor Vergaata. TunnelTalk.
United States Environmental Protection Agency, U. S. (17 September 20). Learn about Sustainability.
Retrieved from EPA.Gov: www.epa.gov/​sus​tain​abil​ity/​learn-​about-​sus​tain​abil​ity
United States National Aeronautics and Space Administration (NASA). (30 September 20). Global
Warming. Retrieved from Earth Observatory NASA: https://​earth​obse​rvat​ory.nasa.gov/​featu​
res/​Global​Warm​ing/​page1.php
Vandewalle, M. (2005). Tunnelling is an Art. Zwevegem, Belgium: NV Bekaert SA.
Wallis, Shani. (April 2022). A Leap Forward for SFRC Segmental Linings in France. Tunnel Talk,
pp. 1–​8.
Chapter 6

Connections and accessories

for segmental tunnel linings
Lead authors: Christophe Delus, Sandrine Ordener, and Ivica Duzic
Contributing author: Mathieu Janes

6.1 INTRODUCTION
In the early stage of tunneling, there were no segmental linings but only masonry and the
lining was made of hand-​placed bricks and mortar. The first segments used were not made
of concrete but cast iron. The ring was divided into several small segments in order that
these small pieces could be handled manually, as at this time, tunnels were not mechanically
excavated.
The first concrete segments were produced for tunnel projects in London, England
as described in Chapter 1. At this early stage of concrete segmental linings, their shape
was similar to the cast iron segments. Connections between rings and segments were
basic: standard straight bolts and nuts were used to connect these segments (similar to cast
iron segments). This type of segment was used up to the end of the 1960s.
The next evolution of the segment design started in the middle of the 1960s. At this time
several major developments occurred for the tunneling industry including the development
of tunnel boring machines (TBMs) in soft ground conditions. New mechanical erectors were
also developed, offering a high accuracy in the ring build. From a precasting point of view,
the first sealing gaskets were used in Hamburg on the third Rohrelbe Tunnel, which started
in 1969.
From this time, there have been two kinds of connections as shown below in Figure 6.1: the
connections in the circumferential joint, to connect one ring to the next. The other connec-
tion within a ring is in the radial joint (also known as longitudinal joint), from one segment
to another segment. Their purpose is not the same and the systems used in both joints can
be different one from each other.

6.2 CONNECTING BOLTS
Straight bolts with highly engineered plastic sockets, usually made of polyamide or acrylo-
nitrile butadiene styrene (ABS), are commonly used throughout the world today. It was not
always the case, and initially some basic systems such as curved bolts (also named “banana
bolts”) were used on projects designed by British engineers and were prevalent in Asia.
This curved bolt system was not highly engineered but accomplished its main purpose,
which was to keep the segments in-​place and the gasket compressed. One of the main weak-
nesses of this system is that it requires more openings in the segment (one in each segment to
be linked) and that the installed bolt are subject to corrosion at two locations (on the head
and in the thread or nut). If this type of connection would be used for a water conveyance
system, it would also require many pockets to be filled as each pocket may create minute
turbulences for the water.

DOI: 10.1201/9781032675541-6
 Connections and accessories for segmental tunnel linings 357

Figure 6.1 Segment joints.

Note: The joint connecting segment of one ring to segment of adjacent ring is called circumfer-
ential joint as it is along the circumference of tunnel; the joint connecting segment of one ring to
an adjacent segment in the same ring is called radial or longitudinal joint as it is along the radius
or longitudinal axis of tunnel.

In the early 1980s, the first straight bolts with plastic sockets were used in Europe. These
sockets were not specifically designed for the tunnel industry, indeed in France or Germany,
steel bolts with plastic sockets from the railway industry, designed for the concrete sleepers,
were used. These sockets were usually produced out of high-​density polyethylene (HDPE)
and the bolts used were spike screws with a sharp thread. Properties of the plastic were also
studied, and it was demonstrated that the HDPE is not the proper material for a tunnel bolt-
ing system. Indeed, the HDPE socket creeps under sustained load and is not capable to hold
the load and thus does not provide sufficient safety factor to keep the gasket compressed.
This creep leads to release or relaxation of the gasket pressure once the TBM rams are
removed and before the installation of the next segment.
In a later stage, at the beginning of the 90s, bolting systems were specifically designed
for the tunnel industry. The thread design was different and developed with roll-​formed
threads (round shape). The sockets were produced out of newer material such as polyamide
or acrylonitrile butadiene styrene (ABS), providing higher resistance than HDPE, which are
also not subject to creep.
Now that the force can be held, the bolting system can also be better designed. The reac-
tion load from the compressed gasket results in tension within the bolting system and the
system should be able to withstand this load (Figure 6.2).
The tightening of the bolt can be controlled with sufficient applied torque. Indeed, the
force exerted by the bolting system can be well established thanks to the relation between the
torque and the tensile strength (according to the AFNOR Norme NF E25-​030, Fasteners–​
Threaded Connections with ISO Metric Thread–​Part 1: Design Rules for Tensile Loaded
Bolted Joints–​Simplified Procedure, Figure 6.3).

T =​ F[0.16P +​ µ(0.583D2 +​ rm)]

where
T =​torques applied
F =​tensile force exerted by the bolting system
P =​pitch of the bolt thread
358 Handbook of Precast Segmental Tunnel Lining Systems

Figure 6.2 Force exerted by the gasket on the bolting system.

Note: Reaction load from the compressed gasket resulting in tension within the bolting system
(ACI 533.5R-​2 0).

Figure 6.3 Various precast segment bolting materials.

Note: Illustration of several bolting materials for the assembly of precast segmental tunnel
linings.
 Connections and accessories for segmental tunnel linings 359

µ =​mean friction coefficient under the bolt head and in the bolt thread
D2 =​diameter on the flank thread
rm =​mean radius of the bearing surface under the bolt head

Bolting systems now exist with various diameters and lengths so it can be used on any pro-
ject and with any requirements:

• With narrow segments, using bolting systems with a small diameter [19 mm (0.75”)]
and short socket [100mm (3.93”)]
• For standard transportation projects, with bolting systems the standard diameter is
25 mm (1.0”) and with sockets from 80 up to 160 mm (1.15 to 6.3”)
• For large diameter projects, with larger diameter bolting systems to provide higher
resistance [up to 400 kN (45 tons)].

Connections with straight bolts and plastic sockets require less effort in the construction of
the formwork than the curved bolts because it is necessary to create only one pocket and
groove into which the bolts are inserted. This type of connection is traditionally associated
with rectangular segments and is generally used both between rings and between segments
within a ring. Figure 6.4 shows the typical housing of a straight bolt. Attention should be
paid to the following geometrical details:

• Pockets should be large enough for head of the bolt and pneumatic wrench to be
easily inserted.
• The slot side of the pocket should have a conicity of at least 1° for ease of mold
release.
• The bolt slot in the segment that houses the plastic socket should also have a compat-
ible conicity.
• The bolt axis should pass through the center of the segment.

Figure 6.4 Bolted connection in a radial (longitudinal) joint.

Note: Typical detail of a bolted connection at a radial (longitudinal) joint of a precast segment
tunnel lining.
360 Handbook of Precast Segmental Tunnel Lining Systems

• The distance between the end part of the plastic socket and the extrados of the seg-
ment should be sufficient to prevent concrete cracking

6.2.1 Bolt materials and coatings for corrosive applications

The steel bolts used in combination with the plastic sockets can be made in different mate-
rials; in standard carbon steel (C45 for example, equivalent to ASTM A307) or with heat
treated steel in order to reach very high mechanical properties (up to steel Grade 10.9,
equivalent to ASTM A490). The selection of the steel to be used should be defined in correl-
ation with the resistance of the associated socket and of course of the design load. Usually,
it is commonly accepted that the bolts should not be subject to a higher load than the yield
resistance of the steel, in order that the bolts only work in their elastic range.
Depending on the project type and its location, bolts may be removed once the ring
annulus has been grouted. In this case, the tunnel designers and contractors consider that
once grouted, the ring is stable and should not be subject to any specific load, therefore the
bolts are not required for structural purposes. This practice, which is common in Continental
Europe, is always subject to designer approval elsewhere.
If the bolts should stay in the tunnel, they should also be coated (i.e. hot-​dip galvanized)
or made of durable material such as stainless steel to make sure that they will not corrode
during the lifetime of the project. Depending on the function of the tunnel and the exposure,
the bolts can be produced with different finishing or material.
See below an example of exposure types that should be considered when specifying the
proper material to be used:

• Sea water (desalination tunnel)

• Chemicals (sewer)
• Exhaust gas/​carbon dioxide (car/​truck tunnel)
• Air (metro/​train tunnel)

The stainless steel is of course an option, but several zinc institutes have made studies of
the rate of corrosion depending on the environment to which the material is exposed.
See Figures 6.5(a) and (b) from Galvazinc (French Zinc Institute), which describes the
rate of corrosion zinc coating applied to steel by hot-​dip galvanizing (HDG) depending
on the environment (according to ISO 14713-​1:2017; “Zinc Coatings –​Guidelines and
Recommendations for the Protection Against Corrosion of Iron and Steel in Structures –​
Part 1: General Principles of Design and Corrosion Resistance”).
The sockets are usually made of plastic but in certain projects, and for specific applica-
tions; stainless steel sockets may be used. For example, temporary bolts are always removed
from projects of Deutsche Bahn but at the same time they always design their projects in
keeping the first and last rings being connected with stainless steel bolting systems (bolt and
socket). This solution ensures that the first (and last) rings remain under compression, mak-
ing like two stoppers and also preventing the other rings to decompress. Permanent bolting
system may also be required next to any niche or opening in the tunnel.

6.3 DOWELS AND BICONES

The dowel, also known as connector, is a type of accessory that can only be used in the
circumferential joints because of the kinematics of the assembly process. In other words,
 Connections and accessories for segmental tunnel linings 361

Figure 6.5(a),(b) Galvazinc data about rate of corrosion depending on the environment.

Note: Thickness of HDG zinc coating as a function of required life to first maintenance and cor-
rosion categories (ISO standard 14713-​1 ).

dowels only connect one ring to the next one. Dowels have been used in many tunnel
projects and now a large variety of dowel systems exists on the market. Historically,
the first dowels were used in Switzerland and at that period of time they were made of
wood. In that time, segments did not include any gaskets and for this reason, the main
purpose of these dowels was the centering (installation alignment) of the precast concrete
elements.
After this initial period, the first dowels were developed for the connection of the rings
with the approach of keeping the gasket under compression. These first dowels were usu-
ally made of steel, like the “Fast Lock Dowel” system developed by Sehulster Tunnels in
the United States as described in Chapter 1. In follow-​up developments, the dowels were
made of plastic such as “Bucklock Sof-​Clip” couplers and the Conex Dowel (developed by
Schulter and Wagner, D2 Consult, Linz, Austria) that was made of wood. The dowels could
either be a friction dowel, or a mechanical dowel.
The friction dowel is a dowel designed with a slightly smaller diameter than the receiving
concrete socket in the precast segment. The dowel is engaged by force in the concrete
element in using the TBM thrust jacks. The Conex Dowel and “Bucklock Sof-​Clip” were
the type of dowel predominantly used with this concept.
The mechanical dowel is slightly different as it is composed of a male and two female
parts. The female parts, also named sockets are cast in the circumferential joint of the seg-
ments. This type of dowel allows having higher performance in pull-​out resistance since
the sockets are designed to perfectly engage the dowel. Please refer to the illustrations in
Figures 6.6 and 6.7 for the friction and mechanical dowels.
The expansion of the usage of the dowels started at the end of the 1980s. Initially
the dowels were mainly used for sewer or water conveyance one-​pass lining projects
under 4.0 m (13.0 feet) in diameter. This was for at least two reasons; the dowels were
not strong enough for larger diameter tunnels including transportation tunnels but also
because the usage of dowels eliminates the pockets in the segments. It has many advan-
tages for this type of project, in providing a better durability for the connection and
keeping a smooth intrados. Figure 6.8 below shows finished tunnel lining using only
dowels and guide rods.
362 Handbook of Precast Segmental Tunnel Lining Systems

Figure 6.6 Sof-​c lip friction dowel.

Note: This dowel system is designed with a slightly smaller diameter than the receiving concrete
socket in the precast segment.

Figure 6.7 Sof-​f ix dowel engaged in a socket.

Note: The mechanical dowel system composed of male (dowel) and female (socket) parts cast in
the circumferential joints which allows for high pull-​o ut resistance due to perfect engagement
between the sockets and the dowel.
 Connections and accessories for segmental tunnel linings 363

Figure 6.8 Paris Line-​1 2 Project segmental tunnel lining.

Note: Paris Line-​1 2 Project used only dowels and guide rods and no bolts and bolt pockets.

Currently, there are mainly two types of dowels in use:

• Mechanical dowels to connect one ring to another one to replace the bolting system,
they are used in every circumferential joint
• Shear dowels (or bicone), which are used to strengthen the ring at special locations
and for special application (cross-​passages, for example)

6.3.1 Mechanical dowels
With this significant evolution, dowels are designed not only for alignment but also for their
mechanical properties in pull-​out and shear resistance. The pull-​out resistance should be
designed to meet the reaction load of the gasket and the shear resistance to withstand the
weight of the segment.
The main advantages of the dowel systems are the following:

• Help during the approach of the segment

The dowels also play a role during the installation of the segment. Thanks to their
shape, they can help the worker to properly install the segment and place it at the
defined and proper location. The material used for dowels should be stiff but flexible
enough to provide this ability.
• Segment alignment
The dowels are frequently called a self-​aligning system. Indeed, once fully engaged
in the sockets, they help to align the ring together. The quality of the ring build is
improved in using a dowel system, it also helps to reduce the installation tolerances
for the overall tunnel lining.
364 Handbook of Precast Segmental Tunnel Lining Systems

• Shear resistance
Dowels are typically made of thermoplastic and can have a steel core center if required
by design. This steel core helps the dowels to have a high shear resistance. Compared
to the design of the bolting system, where bolts are engaged in a tapered channel with
a gap between bolt and concrete, the dowels have a significant advantage, as they
neatly fit 1-​to-​1 in the sockets. It means that the shear resistance is directly mobilized
to the dowel without significant displacement.
• Pull-​out resistance
The dowels are designed to achieve high pull-​out resistance with a minimum of dis-
placement and this to keep the gasket under compression. Their design has evolved
with the time. The first dowel systems had two similar sockets with the dowel pushed
in. Please refer to Figures 6.9 and 6.10 below. The last evolution is the usage of asym-

Figure 6.9 Asymmetrical dowel system.

Note: Fully-​a ssembled asymmetric dowel system used for circumferential joints in segmental
linings.

Figure 6.10 Asymmetrical dowel system parts.

Note: Typical asymmetrical dowel systems components. The black-​c olored parts are cast into
the precast segments.
 Connections and accessories for segmental tunnel linings 365

Figure 6.11 Dowel in a circumferential joint.

Note: Typical dowel connection in a circumferential joint using both embedded and loose
components.

metrical dowel system, with one side screwed-​ in and the other side pushed-​ in.
The screwed-​in side is usually placed on the trailing edge and the pushed-​in side on
the leading edge of the segment. Some designers and contractors still prefer to have a
symmetrical dowel system with both sides being pushed in.
• Speed of installation and safety
Joints with dowels require less work for the construction of the formwork and less
manpower in the tunnel for the insertion into segments. Indeed, compared to bolting
system, as it is a self-​locking system, it does not require any human intervention. It
also makes the dowel system a safer system, as it does not require any worker to be
under the segment (for the bolt insertion).
Figure 6.11 shows the typical housing of a dowel, which is placed along the cen­
terline of the segment. This is a further advantage as the reaction load of the gasket
is parallel to the pull-​out force of the dowel. In most cases, the dowel connections
are used with rhomboidal and trapezoidal segments (but it can also be used with
rectangular segments) to avoid early crawling of the gaskets during the segments-​
approach phase of the ring assembly.
  Figure 6.12 shows a segment formwork circumferential face with a socket for dowel
insertion. As there is no pocket, like for the bolting system, the reinforcement has con-
tinuity and is also easier to design.

6.3.2 Bicones
Bicones as illustrated in Figure 6.13 are also named shear dowels as their primary function
is to take high shear loads, but this type of dowel system can be used for two purposes:
366 Handbook of Precast Segmental Tunnel Lining Systems

Figure 6.12 Dowel socket before concrete placement.

Note: Typical dowel connection arrangement in a circumferential joint as seen in the mold before
casting.

Figure 6.13 Range of bicone shapes and sizes.

Note: Different bicone shapes and sizes with capacities ranging from 150 to 500 kN.

• As an alignment dowel when it is used in conjunction with a bolting system, or

• As a shear dowel to strengthen the ring at specific locations

Compared to the mechanical dowel system, this system, which is designed only to take shear
load, is a single dowel without sockets embedded in the concrete. The bicones are engaged
in a designated concrete socket and depending on the primary usage, they may be glued
or not.
The combination of an alignment dowel together with a bolting system has been mainly
used on challenging projects with high water pressure or large diameter tunnels. The
Hallandsås Tunnel in Sweden, for example is one of them. This rail tunnel was very challen-
ging due to the high pressure [up to 15 bar (218 psi)] and high amount of water present in
the ground. In order to build this tunnel, segments with a very wide gasket [44mm (1.75”)],
inducing a high reaction load, were used. Because of this gasket and the resulting reaction
 Connections and accessories for segmental tunnel linings 367

load, it was decided to use bolting systems in the radial (longitudinal) joint and the circum-
ferential joint. In order to minimize the steps and lips, and consequently improve the per-
formance of the gasket, it was decided to use bicone as alignment dowels between each ring.
The performance of this type of shear dowel was satisfactory to the joint venture and
some contractors started to look at this type of dowel for a different application. As this
type of bicone can produce significant shear resistance (up to 500 kN or 56 tons), their
usage was developed not only for alignment but also for specific areas like cross-​passages
(see Section 6.8).

6.4 GUIDE RODS
The guide rods can only be used in the radial (longitudinal) joint. The guide rods are made
of plastic; their diameter may vary from 30 mm for the thin segments up to 80 mm for the
very thick segments. Lengths can vary but typically covers approximately two thirds of the
ring width. They can be used as a single connection in the radial (longitudinal) joint or in
conjunction with one or two bolts. They are fixed in a groove on one side of the segment
and the other side of the segment should be designed with a continuous groove to engage
and receive this rod. An example is shown in Figure 6.14.
As stated by the name, the main function of the guide rod is to provide guidance to the
segment during segment installation, but they also have a locking functionality. Guide rods
absorb shear forces in the longitudinal joints (ÖVBB 2011). An advantage of using guiding
rods is that, in the longitudinal joints, inserted longitudinal guiding rods can prevent the
segments slipping away from each other during ring building [DAUB Recommendations for
the Design, Production and Installation of Segmental Rings (2013)]. Guiding rods can be
used in conjunction with any connection system in the circumferential joint but when they
are adopted in conjunction with dowel connection systems, they result in a smooth ring-​to-​
ring contact.

6.4.1 New developments in anchored (non-​glued) guide rods

Initially guide rods were glued into the recess at the precast segment fabrication facility.
The usage of the anchored gasket has increased in the past 10 years and with this evolution,

Figure 6.14 Segment guide rod and guide rod groove.

Note: General arrangement of a typical guide rod and groove per ÖVBB 2011.
368 Handbook of Precast Segmental Tunnel Lining Systems

Figure 6.15 Guide rod studs before casting.

Note: Guide rod and anchor studs in a segment mold before casting.

Figure 6.16 Guide rod groove studs after stripping.

Note: Guide rod groove and anchor studs on a radial (longitudinal) segment joint.

the gasket glue is not required anymore at the segment precast factory. The precasters were
looking for a solution to anchor the guiding rod without glue.
A new fixation system with studs has been developed as shown in Figures 6.15, 6.16 and
6.17. In order to use the studs, the guiding rods are drilled. The studs are anchored at the
precast factory. Once the segment produced, only two small portions of the studs remain
 Connections and accessories for segmental tunnel linings 369

Figure 6.17 Guide rod fixed with anchor studs.

Note: Guide rod mounted on a radial (longitudinal) segment joint.

exposed outside of the concrete. The guide rods are easily engaged on the studs to hold it.
This fixation is mechanical, durable, and not sensitive to weather alteration and with health
and safety benefits for the workers (e.g. not exposed to solvent).

6.5 GROUTING AND LIFTING INSERTS

6.5.1 Mechanical erector
Historically, first segments used to be installed by a mechanical erector. A socket installed
in the center of the segment together with a screwed in-​place lifting device were used for
this operation. The sockets were first made of steel and over the time the usage of plastic
grout-​lift socket has been developed. The plastic material made these sockets more durable
but also lighter and easier to install in the formwork.
The connection between the erector arm and the segment is made by the means of a lift-
ing pin. This lifting pin is designed by the TBM and accessories manufacturers. Indeed, its
head should be made to suit the design of the erector, while its thread should perfectly match
the thread of the socket. There are different types of sockets used by the industry, including
single or triple thread style. The triple style thread has an advantage that the lifting pin can
be quickly engaged in the socket (one turn is three threads down). But on the other hand,
when you unscrew it by 1/​3 of a turn, the lifting pin loses one full thread in the socket.
Because it is used for lifting purpose and is also a safety tool, some tunnel designers prefer
to use a single thread style.

6.5.2 Combined application –​ socket for vacuum lift guide pins and

grouting
The development of the vacuum erector was a big change in our industry. The first major
project, on which a vacuum erector was used, was the Channel Tunnel Project, which was
built in the 1987. It was delivered by the Kawasaki-​Robbins JV.
370 Handbook of Precast Segmental Tunnel Lining Systems

Using the vacuum lifter, the central socket now has only one single use: grouting. The
style of the sockets also evolved with this evolution in making the sockets lighter, designed
only for this purpose and sometime shorter. Indeed, they could be shorter as they are from
time-​to-​time combined with designated vacuum erector concrete pockets. With this config-
uration, the grouting socket is placed on the vacuum erector pocket, making the installation
of the screw cap more difficult. But at the same time, this configuration reduces the number
of openings in the segment and facilitates the design of the reinforcement.
Another major change has been the development of backfill grouting through the tail
shield of the TBM. After this evolution, the grout socket (if any are used) is only needed for
secondary and/​or proof grouting.

6.6 JOINT PACKING MATERIALS

Packing material is typically used in the circumferential joint to prevent concrete spalling
during ring installation by the TBM. The packers are used to distribute the stresses gen-
erated by the TBM rams during the assembly but also help in reducing friction impact
between the rings.
The packers could be supplied in different shapes and dimensions to match the ring
design. The dimensions of the joint packers depend on the stresses to be distributed (TBM
thrust jack pressure) the thickness of the material and the behavior of the material itself.
Depending on the culture and practice several types of materials can be specified:

• Plywood (typically, waterproof marine grade), see Figure 6.18

• Bituminous packers, see Figure 6.19
• Bituminous/​elastomeric packers
• Elastomeric packers
• Plastic packers (low density polyethylene or LDPE)

Figure 6.18 Segments with plywood packers.

Note: View of plywood packer material glued onto circumferential joints.
 Connections and accessories for segmental tunnel linings 371

Figure 6.19 Segments with bituminous packers.

Note: View of bituminous packer material glued onto circumferential joints.

The packers can be self-​adhesive or glued in-​place with similar glue as used for gaskets.
They are always installed just after stripping and before storage of the segments. The com-
pressed thickness of the packers should always be considered at the design stage. Indeed,
if dowels are used in the circumferential joints, this additional thickness or space between
the rings may have an influence on how the dowels engage and will have negative impact
on gasket performance if not considered during the design. To ensure that the dowel system
always performs properly, the distance between the two dowel sockets should never vary
even if the distance between the concrete faces evolves. The position of the socket in the
formwork should take into consideration this parameter in reducing the installation depth
of the socket by half of the compressed thickness of the packers.
It is of note that on some projects, packers may also be used in the radial (longitu-
dinal) joint.

6.7 TESTING OF CONNECTION SYSTEMS AND ACCESSORIES

One of the current issues in our industry is that there is no international standard or guide-
line on how to test the segment connection systems and accessories. Each manufacturer has
its own procedure of testing, which could lead to different results because the testing param-
eters are not the same. For the gasket, the STUVA developed a guideline; “Recommendation
for Gasket Frames in Segmental Tunnel Linings” to test the gasket and it has been approved
and adopted internationally. The same should occur for the accessories in the next sev-
eral years.
It is always very important to test the full system and not only one part of it. For example,
a bolting system is made of one bolt and one socket, and it is this full combination that
should be tested. For the dowel system, it is also very important to test a full set, including
one dowel with two sockets. Indeed, if the dowel is tested when engaged only in one socket,
the displacement can be different than for the full system.
372 Handbook of Precast Segmental Tunnel Lining Systems

Please add sample test results for bolts and dowels. For example, for dowels show sample
pullout and shear test.

6.8 DESIGN OF CONNECTION SYSTEMS AND ACCESSORIES FOR

SPECIFIC LOADS

6.8.1 Design of connection systems for gasket pressure

The maximum groundwater pressure defines the design of the sealing gasket. Considering
the relaxation effects of rubber, a safety factor of two is applied to the maximum work-
ing water pressure to ensure that the gasket can withstand the design pressure in the long
term. Depending on the type of connection system used, the maximum offset should also be
considered. For example, it is commonly accepted that a dowel system would have better
ring build tolerances and for this reason, the gasket selection can be made in considering a
smaller offset. This leads to the selection of a proper gasket profile.
The maximum instantaneous reaction force in the gaskets from the gasket load-​deflection
curve or, in a less conservative approach, the maximum reaction force after short-​term
relaxation (within 5 minutes) should be considered during the design of segment connec-
tion systems. Figure 6.2 shows forces that are applied on dowels and bolts depend on the
reaction force of specific type of gasket profile used in the project after accounting for
short-​term relaxation. Gasket reaction force in load-​deflection curves is always presented as
force per unit length of each specific gasket, and in units of kN/​m (or kips/​ft). Therefore, it
is important to determine the corresponding length of the gasket that is acting against each
single dowel or bolts (ACI 533.5R-​20).
For the design of dowels, the total length of the gasket used in one segmental ring depends
on the size and outer perimeter of the tunnel lining. Also, depending on the total number
of dowels placed along the whole ring perimeter and considering a uniform gasket pressure
distribution, a gasket length that pushes against one single dowel is determined. A factor
of safety of 1.25 should be considered for the dowel connection system, and the required
pull-​out force should be obtained and used for the selection of dowel types. Bolts often used
in radial (longitudinal) joints are designed in a similar way, considering their inclined angle
with centerline of the radial (longitudinal) joints (for example, 24° in Figures 6.2 and 6.4).
The length of the gasket in this joint is approximately equal to the ring length. Considering
that often two bolts are designed in each radial (longitudinal) joint for the connection and
that gasket pressure is uniformly distributed over the length of the joint, the force in each
bolt due to gasket compression is calculated. A factor of safety of 1.25 is advisable and bolt
connection systems should be selected or verified based on the system pull-​out force and
tensile yield strength.
The previous approach is the most conservative and there is another one, which is to
combine the curve of the reaction load of the gasket together with the connector. Like for
the previous calculation the length of gasket per connector should be considered. The short-​
term relaxation can also be considered, and a factor of safety should be applied on the
load of the gasket. This factor of safety can be of 1.25 up to 1.5 depending on the project
requirements.
The two curves intersect at a theoretical displacement and this is the theoretical equilib-
rium between gasket and dowel. It is then to check what level of watertightness results from
the gasket with this opening (gap). See Figure 6.20, which is an example of this conjunc­
tion curve.
 Connections and accessories for segmental tunnel linings 373

Figure 6.20 State balance curve between dowel system and gasket.

Note: State balance curve between dowel system Anix 60 and CTS 33-​1 6 gasket.

6.8.2 Design of circumferential connectors for accidental removal of

TBM thrust jacks
This approach is not always considered by the designers. Indeed, the latest TBMs are
equipped with safety systems preventing the retraction of the rams. If the connectors should
be designed with this parameter, then the worst-​case scenario should be considered with a
single segment installed on the top of the ring. This configuration, which is the worst-​case
scenario induces a bending moment on the segment and on the ring-​to-​ring connectors.
Figure 6.21 shows a full-​scale test to consider this approach.

6.8.3 Design of grouting and lifting inserts for segment lifting and

secondary grouting
The grout-​lifting socket can have a dual purpose. They can be used for lifting of the segment
and for secondary grouting as described in Section 6.4. The design of the grout lift socket
should consider the application for which they are used.
If the socket is used for lifting, each tunnel designer may specify his minimum factor of
safety to be applied on the weight of the heaviest segment. For example, in Singapore, a
factor of safety of 5.0 on the heaviest segment should be considered. This means the socket
should be able to withstand a load equal to five times the heaviest segment. The TBM manu-
facturer may suggest considering a higher factor of safety depending on the force applied
on the segment by the lifting arm to stabilize it during its erection. A pull-​out test in the
374 Handbook of Precast Segmental Tunnel Lining Systems

Figure 6.21 Full-​s cale test on partial segmental ring.

Note: Full-​s cale test on a partial segmental ring to analyze cantilever forces with only one seg-
ment installed.

concrete with the proper reinforcement should be organized by the contractor, as each con-
crete is different and can lead to different results. Figures 6.22, 6.23 and 6.24 show a pull-​
out test on grout-​lifting socket embedded in a concrete segment.
If the socket is used for grouting purpose, the main concern is the watertightness of the
screw cap after its installation in the socket. Like the gasket, the screw cap will be exposed
to the water pressure during the lifetime of the tunnel. For this reason, a durable material
for the screw cap and its O-​ring should be used: polyethylene or polypropylene for example
for the screw cap and an EPDM base material for the O-​ring. In the same manner for the
gasket, and due to the aging of the material, it is suggested to test the screw cap for 20 hours
and with a factor of safety of 2.0 against the maximum water pressure as per, “STUVA
Praxis 54 Recommendations for Gasket Frames in Segmental Tunnel Linings”.

6.9 DESIGN OF CONNECTORS AT CROSS-​PASSAGE BREAK-​O UT AREAS

There are several types of secondary structures, which can be built during the construc-
tion of a tunnel. For example, ventilation shafts can be connected to a main tunnel, cross-​
passages are built between twin tunnels for rail or metro tunnels for safety of the future
passengers. In the event of large diameter road tunnels, cross-​passages can be built to create
rescue chambers or emergency exits.
During the excavation of these cross-​passages, the main structure of the tunnel should be
cut and removed to create the opening. During this operation, the tunnel lining must remain
stable. For this reason, temporary supports are used. These supports should be able to transfer
the load generated by the opening and be stiff enough to avoid deformation of the rings.
Different types of support, depending on the opening size and the ground conditions,
have been studied and used. For example, full steel ring break-​out, “hamster cages”, rock
bolts and ground freezing are technologies, which have been used.
 Connections and accessories for segmental tunnel linings 375

Figure 6.22 Full-​s cale pull-​o ut test on segment.

Note: Full-​s cale pull-​o ut test on segment lifting insert.

Figure 6.23 Full-​s cale pull-​o ut test apparatus.

Note: Full-​s cale test apparatus for measuring pull-​o ut strength of lifting sockets on segment.

More recently, the usage of high-​strength shear cones has been first introduced success-
fully to the A86 Duplex Project in Paris, tunnel built by Vinci Grands Projets; they have
been quickly adopted on other projects around the world. The concept is to develop and
use a shear cone having similar resistance than the concrete lining itself, if the bicone would
376 Handbook of Precast Segmental Tunnel Lining Systems

Figure 6.24 Grout-​lifting socket pull-​o ut test results.

Note: Results of destructive test on precast segment from grout-​lifting socket pull-​o ut test.

Figure 6.25 Bicones installed in segment circumferential joint.

Note: Two different bicones installed in concrete reservations in the circumferential joint

be too stiff, it may induce failure of the concrete at an early stage, and if the bicone is too
weak, it cannot provide the adequate stiffness to the lining. For this reason, several bicones
with various resistances, stiffness and designs have been developed to cover different project
requirements (see Figure 6.25).
 Connections and accessories for segmental tunnel linings 377

6.10 THROUGH SEGMENT BOLTING/​D OWEL SYSTEMS FOR SEISMIC

APPLICATIONS
Additionally, to the standard shear cones, shear cones with tie rods can also be used for
start and end rings but also next to any cross-​passage (see Figure 6.26). This solution in
addition to the standard bicones provides more rigidity to the ring and helps to prevent
from any opening on the long term.
Special segments should be produced with pipes joining the bicone reservations at each
end of the segment. These pipes will allow the installation of the tie-​rods on-​site (see
Figure 6.27).
As rings are most of the time tapered, several lengths of tie-​rods should be produced. Each
tie-​rod is color coded, to differentiate the correct location the tie-​rod should be installed.
The tie-​rods should be already installed in the segment before they are brought to the tunnel
(see Figure 6.28).

Figure 6.26 Rings connected by bicones and tie-​r ods.

Note: Sequence of segmental rings connected with bicones and tie-​rods.
378 Handbook of Precast Segmental Tunnel Lining Systems

Figure 6.27 PVC duct in precast segment formwork and bicone sockets.

Note: Installation of the PVC duct D40 between bicone sockets in precast formwork for hous-
ing tie rod.

Figure 6.28 Socket with color coded tie-​r od.

Note: Exposed bicone socket at segment circumferential joint with projected color coded
tie-​rod.

The sequence for the ring installation would be the following (see Figure 6.29):

• Installation of the first segment with tie-​rod engaged and tighten in the end bicone or
end cone (half bicone)
• After excavation of the next drift (one ring), retraction of the TBM thrust jacks
• Tightening of the bicone on the ring already installed with tie-​rod
• Tie-​rod of the segment to be erected should be engaged in the bicone manually and
then the segment installed
• The tie-​rod is then tightened with the special tightening tool
 Connections and accessories for segmental tunnel linings 379

Figure 6.29 Tunnel segmental ring installation sequence using tie-​rods.

Note: Schematics of sequences of the installation of segments equipped with tie-​rods.

6.11 RECENT DEVELOPMENTS IN SEGMENT JOINT CONNECTION

SYSTEMS

6.11.1 Boltless radial (longitudinal) joint connection systems

As described in Section 6.4 of this chapter, the usage of guide rods in the radial (longitudinal)
joint has significantly increased over the past years and several projects around the world
already used a combination of dowels with guide rods to connect the segments. This con-
cept offers a smooth intrados to the ring without any opening except those for the vacuum
erector or the grout-​lift socket. This has been mainly used on water conveyance projects but
also on some transportation projects, such as recently on the Grand Paris Project.
The limit to this boltless radial (longitudinal) joint is that the guide rod is only working in
shear. If the tunnel designer and contractor require a system able to act in tension, then the
boltless radial (longitudinal) joint connection with only guide rod would not suit.
A Japanese radial (longitudinal) coupling system made of steel has been in use for several
years. To date, however, they have not been incorporated in the design of any tunnel project
in Europe or North America. These systems are usually welded to the segment reinforce-
ment and are very heavy to install. An example is presented in Figure 6.30.
Recently, new radial (longitudinal) connection systems have been developed (US Patent
No. D930,464 S dated 14 Sep 21) as illustrated in Figures 6.31 through 6.34. They are
380 Handbook of Precast Segmental Tunnel Lining Systems

Figure 6.30 Japanese steel radial (longitudinal) joint connector system.

Note: Japanese radial (longitudinal) coupling system made of steel which are usually welded to
the segment reinforcement.

made of new material such as composite plastic, providing high resistance and durability.
The aim of this new radial (longitudinal) system is to secure the two segments and prevent
them from their opening. Two Omegas are cast in the segments while a “dowel” is making
the connection between the segments. The shape of this pin can be adjusted to the gasket
to have a full gasket compression only when the system is fully closed. This kind of new
technology allows for the use of boltless segments for tunnel linings subject to high internal
pressure.

6.11.2 Connector systems for segment removal in top-​down shaft

construction
The use of concrete segments for construction of shafts has been primarily developed in
the UK. Usually, the shaft segments are of larger dimensions than the one used for tun-
nels. The ground is usually excavated by means of excavator and the segments can be
either installed by jacking (from the top) or underpinned when the ground conditions allow
it. Nowadays, mechanized excavation of shafts has also been developed by Herrenknecht
VSM (see Figure 6.35).
 Connections and accessories for segmental tunnel linings 381

Figure 6.31 Non-​c orrosive radial (longitudinal) joint connector system.

Note: New non-​c orrosive radial (longitudinal) joint connector system allowing use of boltless
segments for tunnel linings subject to high internal pressure.

Figure 6.32 Radial (longitudinal) joint connector.

Note: New non-​corrosive radial (longitudinal) joint connector system installed on segment
joints.
382 Handbook of Precast Segmental Tunnel Lining Systems

Figure 6.33 Patented radial (longitudinal) joint connector (piece 1).

Note: Piece 1 of US Patent No. D930,464 S, 14 Sep 21 for a radial (longitudinal) joint connector
used between adjacent precast segments.

Figure 6.34 Patented radial (longitudinal) joint connector (piece 2).

Note: Piece 2 of US Patent No. D930,464 S, 14 Sep 21 for a radial (longitudinal) joint connector
used between adjacent precast segments.
 Connections and accessories for segmental tunnel linings 383

Figure 6.35 Shaft construction using Herrenknecht VSM Technology.

Note: Circular shaft construction using precast segments for ground support and Herrenknecht
VSM Technology for excavation.

Segments are usually connected radially (longitudinally) with bolting systems, and guide
rods may also be used to ensure a good alignment of the segment. For the circumferential
joint, the rings are connected by means of tie rods that can apply a post-​tensioning on the
ring to ensure the gasket compression. These tie-​rods are usually combined with dowels that
ensure the alignment of the segments.
Depending on the type of shafts, the bolts to connect the segments in the radial (longitu-
dinal) joint can be tightened from the extrados (jacking or Herrenknecht VSM technology)
or from the intrados (underpinning technology).

6.11.3 Systems for project with PVC/​HDPE membrane for corrosion

resistance
For sewer projects, to increase the durability of the concrete, several solutions are avail-
able: Combi-​segments (reference to Chapter 8), polymer concrete but also cast-​in-​place lin­
ing. The thickness of the lining can be adjusted depending on projects requirements.
One of the first projects to select this kind of solution was the Sacramento Upper
Northwest Interceptor Project, in Sacramento, California, USA, which was built in 2008
to 2009 by Traylor-​Shea JV and designed by URS (now AECOM). On this project, a PVC
membrane with “T-​lock” (Ameron) system was placed on the formwork prior to casting.
As this membrane covers completely the intrados, this kind of solution is only possible with
boltless segments. Dowels and guide rods were used in this instance. Note that the mem-
brane is slightly longer and larger than the concrete segment itself. Indeed, this overlapping
allows the welding of the segment membranes in using electric heat guns, this operation is
made on site once the ring has been erected and the annulus grout has set. For this applica-
tion, lifting of segments was performed using a mechanical erector with a casted-​in-​place
lifting socket. The use of vacuum lifting erector would not be possible as it could potentially
384 Handbook of Precast Segmental Tunnel Lining Systems

Figure 6.36 Segment with cast in PVC lining.

Note: Segment with cast in-​p lace PVC lining with opening around the grout/​lifting socket.

tear-​off the membrane from the concrete. The membrane had an opening for the grout lift
socket and this opening was patched over on-​site, at the same time the circumferential and
radial (longitudinal) joints were welded with electric heat guns. Figure 6.36 shows a seg­
ment with a cast-​in PVC lining.
For the IDRIS Project (Inner Doha Resewage Implementation Strategy), Bouygues imple-
mented the same concept with an additional development. To prevent the patching of the
grout-​lift socket openings on site, the grout-​lift socket was fixed to the membrane by the
mean of an adaptor welded to the membrane. For this reason, the socket was made of two
elements:

• The adaptor with an inner thread, which was welded to the membrane
• The socket with an outer thread, which is screwed-​on the adaptor after welding and
prior to concrete placing

In order to ensure a perfect repetitiveness and a sealing, a special welding machine was
developed to weld this adaptor to the membrane. This operation was performed at the pre-
cast fabrication plant.
The implementation of this solution was not only possible because the adaptor and mem-
brane are made of a similar material (for welding purposes), but also because the screw caps
are made of similar material as the membrane and provide the same chemical resistance to
hydrogen sulfide gas (H2S). Figures 6.37 to 6.39 show examples of the welding station, as
well as a segment with HDPE segment.

6.12 FASTENING SYSTEMS FOR TUNNEL SEGMENTS

Fastening systems to the precast segments are a key aspect of the design that should be
considered for every bored tunnel and shaft lining system. These anchoring systems are
 Connections and accessories for segmental tunnel linings 385

Figure 6.37 Welding station for watertight attachments.

Note: Welding station for watertight grout /​ lifting socket attachment.

Figure 6.38 Segment formwork with membrane.

Note: Segment formwork with membrane and grout/​lifting adaptor welded prior to casting.

often used to support components of railway overhead catenary systems as well as both
temporary and permanent mechanical and electrical equipment of road and utility tunnels.
Additional applications of fastening systems include supporting intermediate slabs, corbels,
cross-​passage doors and platform screen doors.
386 Handbook of Precast Segmental Tunnel Lining Systems

Figure 6.39 Finished segment with HDPE lining.

Note: Finished segment with HDPE lining cast onto the segment intrados.

Looking at the overall investment of underground infrastructure projects including the

design, construction and long-​term service life, the costs of fasteners are proportionally less
but have much higher value in safety relevance. Anchorage failures in tunnels are rare, but
when they do occur, consequences can be very costly and tragic. In general, these fasteners
can be divided into two main categories.

• Traditional anchor systems with drilling

• Advanced systems for use without drilling

6.12.1 Traditional anchor systems with drilling

Traditional anchor systems with drilling have some disadvantages, mainly the issues of
drilling and installation quality and workmanship issues, which have made this system less
favorable.
Additional efforts and specialized power tools are needed when installing anchors in fully
cured high-​strength concrete. Most importantly, drilling may damage concrete, reinforce-
ment or segment gaskets resulting in negative impacts on structural behavior, sealing per-
formance, corrosion protection and long-​term durability. Some of the major issues with
traditional post-​installed drilled anchor systems are shown in Figures 6.40 to 6.42.
Tunnel workers are exposed to many health hazards during construction. One of the
most critical work-​related illnesses is silicosis. Heavy machinery traffic, excavation process
and soil transport are some of the primary sources of dust. Common methods of dust con-
trol are the use of water, ventilation and dust collectors among other methods. In relation
to fastening systems in tunnel construction post-​installed anchorage with drilling (shown in
Figure 6.43) results in dust creation, hence dust control measures are needed.
By choosing embedded fastening systems (e.g. inserts, cast-​ in channels) shown in
Figure 6.44 the dust exposure can be significantly reduced in the tunnel. Other common
 Connections and accessories for segmental tunnel linings 387

Figure 6.40 Trial anchor drilling locations.

Note: Typical trial drilling locations for anchor bolts in a precast concrete segmental lining.

Figure 6.41 Anchor placement alignment quality.

Note: Typical drilled-​in anchor alignments showing both perpendicular and misaligned anchor
placements.
388 Handbook of Precast Segmental Tunnel Lining Systems

Figure 6.42 Fixing anchor failure in tunnel crown.

Note: A major issue with post-​installed drilled anchors includes fixing failure connected to sus-
pension rods.

Figure 6.43 Post-​i nstalled anchors.

Note: Issues with post-​installed anchors include dust creation, hence dust control measures are
needed.
 Connections and accessories for segmental tunnel linings 389

Figure 6.44 Embedded fastening systems including threaded inserts and cast-​in channels.
Note: No dust creation conditions and, therefore, no dust control measures needed.

Figure 6.45 Different types of threaded inserts (typical).

Note: Various types of typical threaded inserts used with mechanically fastened channels con-
nection systems.

health hazards are related to noise and vibration. These hazardous effects can be reduced
with embedded fastening systems also.

6.12.2 Advanced systems without drilling

Embedded fastening systems eliminate the need for anchorage zones thus allowing for a fast
and high-​quality installation in a precast facility as well as health, safety and environmental
advantages. There are also no time-​dependent behavior or temperature-​related dependen-
cies related to the performance of cast-​in-​place solutions, which make these systems more
favorable.
Threaded inserts as shown in Figure 6.45 allow reliable fixing points combining the
advantages from embedded fasteners with a screw connection. For tunnel installations,
these inserts can provide a continuous grid of durable anchoring points once considered in
the initial segments design.
390 Handbook of Precast Segmental Tunnel Lining Systems

6.12.3 Cast-​in channel systems with mechanical interlocks

In the wake of providing a universal solution, anchoring technology shifted towards the
use of cast-​in channel systems based on the mechanical interlock principle. Cast-​in systems,
as shown in Figures 6.46 through 6.48, are basically adjustable channels with anchors

Figure 6.46 Anchor installation.

Note: Principle of cast-​in fastening system for segments including channels with welded and hex-
agonal head anchors.

Figure 6.47 Embedded fastening systems.

Note: Hot-​rolled cast-​in channels and bolt with serrations for a three-​a xial load transfer.
 Connections and accessories for segmental tunnel linings 391

Figure 6.48 Cast-​in channel load directions.

Note: Tension load is applied in z-​d irection (in direction of channel bolt), while shear load is
applied in y-​d irection (perpendicular to the channel axis) and x-​d irection (longitudinal to chan-
nel axis).

(preferably I-​anchors) that are placed in segment forms before casting concrete. Such sys-
tems provide efficient, flexible and safe methods of fastening without drilling while signifi-
cantly improves the installation quality. Some key advantages of this system are the increase
in connection adjustability, allowance for more flexibility in the construction schedule,
increased service life, can be used for temporary installations during construction, allow-
ance for accurate planning due to reliable assembly times and reduction of overall construc-
tion schedule due to reduced installation time. These installations can also be structurally
integrated with segment reinforcing elements.
Cast-​in-​place systems also have lower maintenance cost, lead to much easier quality con-
trol, both in precast plants and before ring assembly and provide more favorable point load
properties. Note that definitive preliminary planning efforts are needed for all cast-​in chan-
nel systems, namely preparation of installation drawings for segmental lining (Figure 6.50),
as well as effects on ring construction logistics that need to be closely coordinated with the
tunnel contractor and TBM operation team.
One of the most suitable applications of cast-​in channels is in railway tunnels and espe-
cially high-​speed rail systems. This is due to requirements for anchoring electrical compo-
nents to the tunnel lining, such as catenary wire anchoring, wheel tensioners, anchor points
and line feeders (Figure 6.49).
In addition, in these tunnels, resistance to fatigue as a dynamic type of load is a key
requirement. Finnetunnel, 7.0 km (4.3 miles) long, twin tunnel tubes, 9.6 m (31.5 feet)
diameter high-​speed railway line from Erfurt to Leipzig (Germany), is one such example.
Figures 6.50 through 6.55 show all stages of building a cast-​in fastening system for this tun­­
nel from design to final service.
Loads can result both vertically and horizontally at the point of anchoring as Figures 6.48
illustrate. Serrated cast-​in channels are a new development for applications with high longi-
tudinal loads, where the channel is cast-​in vertically. The special toothed channels guarantee
a positive mechanical connection as illustrated in Figure 6.46. Furthermore, serrated chan­
nels can safely absorb high fatigue-​related stress ranges with a high number of cycles. They,
therefore, meet all requirements for secure anchoring of vertical loads with an increased
392 Handbook of Precast Segmental Tunnel Lining Systems

Figure 6.49 High-​s peed railway tunnel cross-​s ection.

Note: Schematic of electrification components to be anchored to tunnel lining in a typical high-​
speed railway tunnel.

Figure 6.50 Curved cast-​in channel fastening system.

Note: Installation drawings for segmental lining layout in high-​s peed railway tunnels.

requirement on load cycles for example catenary systems, jet fans, passage doors and other
utilities in tunnels with train passages. Hot-​rolled cast-​in channels with serrations are also
suitable for applications in safety relevant areas of nuclear power plants and seismic areas.
To prevent concrete from filling the channel when pouring the concrete, a filling strip is
installed inside the channel that can be readily removed once the concrete has cured. This
allows T-​bolts to be used for rapid installation of equipment in the tunnel. Such integrated
fastening systems can be employed for the temporary assembly of supply and transport
lines, walkways or working platforms while the tunnel is being driven.
Similar systems have been used for subway tunnels such as 25 km (15.5 mile) long, 5+​
1 segmental ring configuration at Shenzhen Subway Line 9 in China. In this tunnel, due to
requirements for anchoring in eight different locations all around the tunnel perimeter, a
full-​ring cast-​in channel integrated system was adopted for power rail fixing, safety walkway
 Connections and accessories for segmental tunnel linings 393

Figure 6.51 Curved cast-​i n channels.

Note: Cast-​in fastening system channels for a high-​speed railway tunnel delivered in pairs to
precast segment plant.

Figure 6.52 Curved cast-​in channels in a segment mold.

Note: Cast-​in fastening system channels laid and assembled in segment formwork for a high-​
speed railway tunnel.

fixation and support of cables, mechanical and electrical equipment. Also, utility tunnels
such as the 8.5 km (5.3 mile) long 3.0 m (10 foot) diameter tunnel in Berlin, Germany with
5+​1 ring configuration have taken advantage of the cast-​in channel systems to house under-
ground high-​voltage cable networks that is future-​proof for anticipated upgrades.

6.12.4 Post-​installed assembly channels through connection systems

Post-​installed assembly channels provide both adjustability and flexibility. This is important
for repair and rehabilitation projects where expansion or upgrades of utility equipment
and production facilities are required. However, the main issue of drilling and installation
quality remains. To overcome this problem, a solution has been developed by mounting
394 Handbook of Precast Segmental Tunnel Lining Systems

Figure 6.53 Segments with curved cast-​in channels.

Note: Handling and storage of segments with integrated cast-​in fasteners in precast yard using
vacuum erector.

Figure 6.54 Curved cast-​in channel system.

Note: Handling and storage of segments with integrated fasteners in precast yard using vacuum
erector.

modified assembly channels at segments utilizing segment connection bolts that do not
require drilling. Figure 6.56 shows this solution for 1.0 km (3,200 ft) long Diabolo railway
line twin tunnels with a 7.3 m (24 ft) internal diameter and 7+​1 segmental ring configur-
ation. As shown in Figures 6.57 through 6.60, installation of curved assembly channels was
accomplished with mounting plates connected to the back of post-​installed channels and
through the bolted connection system in the circumferential joints. This fastening system
has been utilized for installing numerous utilities as well as mechanical and electrical equip-
ment in this tunnel.
Segmental lining systems take advantage of dowel connections in circumferential joints
and, therefore, circumferential pockets designed in segments for fastening bolts are often
not always available. Solutions such as in Figure 6.59 cannot be always guaranteed with
 Connections and accessories for segmental tunnel linings 395

Figure 6.55 Curved cast-​in channel fastening system in a high-​s peed railway tunnel.
Note: Finished segmentally lined high-​
s peed railway tunnel after fit-​
o ut and shortly before
trial run.

Figure 6.56 Curved post-​installed channel fastening system in high-​s peed railway tunnels.
Note: Finished segmentally lined high-​
s peed railway tunnel after fit-​
o ut and shortly before
trial run.
396 Handbook of Precast Segmental Tunnel Lining Systems

Figure 6.57 Bolted in-​p lace channel fastening system.

Note: General arrangement of the bolted in-​p lace channel fastening system that relies on seg-
ment assembly bolts.

Figure 6.58 Bolted in-​p lace channel system.

Note: General arrangement of the bolted in-​p lace channel fastening system that relies on seg-
ment assembly bolts.

Figure 6.59 Bolted in-​p lace channel.

Note: Typical arrangement of the bolted in-​p lace channel that relies on segment assembly bolts.
 Connections and accessories for segmental tunnel linings 397

Figure 6.60 Bolted in-​p lace channel system and cabinet.

Note: General arrangement of the bolted in-​p lace channel fastening system with a permanent
cabinet in place.

Figure 6.61 Tension rod suspension system.

Note: Detail of a tension rod support system.

assembly channel systems and, despite all of its advantages, the anchorage issue and sec-
ondary drilling remain a primary drawback.
Cast-​in fastening systems provide similar opportunities for road tunnels with fixing light-
ing, signal facilities, ventilation and exhaust air ducts. Figures 6.61 to 6.62 show another
398 Handbook of Precast Segmental Tunnel Lining Systems

Figure 6.62 Tension rod suspension system installation.

Note: Modern fastening systems with tension rods for supporting intermediate slab in Dobrovsky
tunnel, Czech Republic.

major opportunity for modern fastening systems in road and rail tunnels for supporting
plenum slabs. Tension-​rod systems are quick to install and reliable mounting systems as
favorable solutions when there is a high demand for load capacity in addition to high
demands for corrosion and fire protection.

6.13 CASE HISTORIES OF INTERESTING PROJECTS

6.13.1 Grand Paris Subway Program, Paris, France

After London and New York, Paris was one of the first cities to develop a metropolitan
network at the very end of the nineteenth century. The first line was opened for the Paris
Universal Exposition in 1900. The lines were regularly deployed in the course of the twen-
tieth century. In the early stage, the lines are mainly central to connect the different areas
within Paris. The lines are then extended to connect to the close-​in suburban cities. The
network is operated by the Régie Autonome des Transports Parisiens (RATP).
At the beginning of the twenty-​first century a large project was initiated to develop fur-
ther the network in extending two existing lines (Lines 11 and 14) and creating four add-
itional lines (Lines 15 to 18):

• Two lines creating a large circular ring around Paris, connecting suburbs cities to each
other. Please refer to Figure 6.63.
• Two lines to connect Paris to Charles de Gaulle Airport, and Orly Airport to Versailles

The Société du Grand Paris is the owner and in charge of the development of this new net-
work. In total this new extension represents 200 km (122 miles) of metro lines and 68 new
stations. With this project, the network of the Paris Metro will almost be doubled. The first
lines should come in operation for the Paris Olympics in 2024.
 Connections and accessories for segmental tunnel linings 399

Figure 6.63 Grand Paris Subway Program, Paris, France.

Note: General arrangement plan of the four new interconnecting suburban metro lines; totaling
200 km (122 miles) of new transit system and 68 stations.
Source: https://​e n.m.wikipe​d ia.org/​w iki/​F ile:Gran​d _​p a​r is_​e xpr​e ss.svg (Creative Commons
License). 

For the new lines, like the most recent extensions built, it was decided that two metro
trains operate in the same tunnel, without any separation wall. For this reason, the internal
diameter for the lines 15, 16, 17 and 18 is about 8.70m (28.5 feet) with segments having
400 mm (16 inch) thickness.
On a design point of view, most of these projects have been built with a 7+​0 ring config-
uration (key segment having the same size than a regular segment). Depending on the choice
of the contractors, the segments can be rectangular or “parallelogramic-​trapezoidal” (i.e.
rhomboidal). A combined (EPDM +​hydro-​swelling cord in the center) anchored gasket has
been implemented on all the projects. For the first time, on Line 16, steel fiber reinforced
concrete has been used in Paris Metro.
On all the projects, segments in the circumferential joint are connected by the mean of
plastic dowels. Symmetrical dowels (Sof-​Fix) were already used on previous Paris Metro
Project, but for the Line 15, some projects decided to use asymmetrical dowel (one side
screwed in, one side pushed in) testing the new features of this system (easier and more
ergonomic installation, for instance).
400 Handbook of Precast Segmental Tunnel Lining Systems

In the radial (longitudinal) joint, guiding rods fixed by the means of studs (and not
glue) are implemented on all projects. Indeed, as the gasket is anchored on all projects, the
usage of studs to fix the guide rods offered the precast fabricators the opportunity to work
“glue free”.
Additionally, the radial (longitudinal) joints were not necessarily bolted; the decision was
left to the contractors. Some decided to use one bolting system in each segment, one only in
the key or no bolting system at all. But it is to note that on all the projects, the bolts had to
be considered for temporary use only. The bolts must be removed to prevent them loosening
and falling on the train tracks due to vibrations.
The grouting is made essentially through the tail shield, most of the contractors did not
install any socket for secondary grouting through the segments. If a socket is used, this
socket is then installed on the top of the erector cone.
For the construction of openings, almost all the projects used shear bicones. They could
be standard shear bicone or bicones combined with tie-​rods. On a specific area, where Lines
15 and 12 crossed themselves, it was decided to use shear bicone with tie rods in a very high
steel class (Grade 10.9) on a long section. Indeed, one portion of the tunnel was almost con-
sidered “a bridge” over the other line. The usage of the bicones helped to stiffen the tunnel
for this specific application (also subject to the vibration of the trains).
Several other innovations have been implemented on the Grand Paris Project. For
example, shafts have been built with segments and using the Herrenknecht VSM tech-
nology. This was one of the first times in France and it was done successfully; on the Line
15-​T3C and L17-​1.
At the time of this writing, the Lines 15 and 16 are under construction with some sec-
tions already completed and Lines 17 and 18 started the excavation phase at the very end
of 2021. The first lines should be opened for the Olympics of 2024, while the total project
completion should not occur before 2030. More information can be found on www.soci​
eted​ugra​ndpa​ris.fr/​

6.13.2 Port of Miami Highway Tunnel, Miami, Florida, USA

Highway tunnels are frequently constructed to solve traffic congestion problems in major
urban areas, the recently completed Port of Miami Tunnel is an excellent example of tun-
neling technology being implemented to improve the quality of life in the City of Miami, FL.
The city is dependent upon international container trade and the tourist industry with the
associated need for people to visit and board cruise ships all of which are based on Dodge
Island near the downtown area. The Dodge Island Cruise Ship Terminal and the Port of
Miami Container Port was connected to downtown Miami with a single causeway that
was connected to the central business district of Miami thus causing traffic congestion and
delays for people and goods trying to go to and leave the Port of Miami.
The solution to the problem was to build twin two-​lane highway tunnels connecting
Route I-​395 north of downtown Miami to the Port of Miami facilities on Dodge Island thus
removing container truck traffic and cruise ship boarding traffic from downtown Miami.
The Florida Department of Transportation began a study in 1987 and evaluated various
options inclusive of bridge widening, alternative routes through the city but in the end
found that the twin highway tunnel option connecting to Route I-​395 in the north was the
best option as it created a safer and less congested downtown Miami while better servicing
the current and future needs of the various business operating in the Port of Miami.
 Connections and accessories for segmental tunnel linings 401

After various studies and evaluations, the Florida Department of Transportation (FDOT)
selected the public-​private-​partnership (PPP) construction model for building the project.
FDOT contracted (through MAT Concessionaire) with the international contractor, Civil
& Building North America Inc. (formerly Bouygues Civil Works Florida). The project con-
sisted of twin-​tube tunnels 1,260 m (4,134 feet) long each with a precast segmental tunnel
lining 11.27 m (37.0 foot) internal diameter × 12.47 m (41.0 foot) external diameter to be
constructed under Biscayne Bay in a soft, porous limestone/​coral formation. There were
many geological challenges that had to be addressed before the tunnel could be constructed
with a Herrenknecht hybrid TBM. Refer to Figure 6.64. Many papers have been written
on the geotechnical challenges encountered and addressed by the tunnel contractor on this
project.
Civil & Building North America Inc. (CBNA) had considerable international experience
in building large diameter highway tunnels in challenging geological conditions and worked
closely with their design engineer of record in developing the optimum final tunnel lin-
ing design for the project. Since the tunnel is located in a subtropical climate with high
humidity, corrosion of any steel segment connecting componentry had to be addressed as
the tunnel has a design life of 100+​years plus the owner complicated the segmental lining
design by adding a “security design requirement” since the lining is under water and had a
relatively shallow cover, the ring-​to-​ring connectors must support a defined force exerted if
an explosion occurred in the tunnel.
CBNA and their designer developed a very innovative segment connection system in con-
junction with the Tunnel Engineering Department of Anixter (now Optimas Solutions). In
lieu of corrosion resistant bolts being installed as permanent connectors in the circumferen-
tial joint faces, a special plastic over steel bicone was developed to connect segment ring to
segment ring. As shown in Figure 6.65 the bicones were designed with a hard plastic outer
cone that was injection molded over a high strength steel center steel axis. This innovative
design gave the bicone a high shear rating, 375 kN (42 tons) but flexibility in movement to
minimize concrete damage during load distribution.
For the radial (longitudinal) joint connections, the lining designers specified guide rods
only as this is a common practice in Europe where no permanent steel bolts are used to
connect segment to segment in the longitudinal joints of the lining. Guide rods are cir-
cular plastic rods that are glued into recesses cast into each radial (longitudinal) joint face.
Because of the large diameter of this tunnel, the diameter of the plastic guide rods was
60 mm (1.35”). Refer to Figure 6.66. The all-​plastic guide rods provided shear resistance
against segment to segment movement and being 100% plastic eliminated any concerns
about long-​term corrosion.
To facilitate assembly of the (8) eight-​piece ring with a segment width of 1.67 m (5.5
feet) and with each segment weighing many tons, the segments needed to be temporarily
connected ring to ring in the TBM so that when the rams were retracted, the gaskets
remain compressed, and each segment remain in its installed position. A temporary con-
nection system was implemented from the leading segment ring using 28 mm (1.10 inch)
diameter steel bolts. Refer to Figures 6.67 and 6.68. After the segmental lining was back­
fill grouted with a cementitious backfill grout and the grout hardened, the temporary
steel bolts were removed and “leap-​frogged” down the tunnel for reuse. This is a very
common practice in other parts of the world, one reason the bolts are also removed is
that they can back out over time and fall to the roadway thus being a potential long-​term
safety hazard.
 newgenrtpdf
402
Handbook of Precast Segmental Tunnel Lining Systems
Figure 6.64 Port of Miami Highway Tunnel, Miami, Florida, United States.
Note: TBM breaking into the shaft of twin-​t ube Port of Miami highway tunnels.
 Connections and accessories for segmental tunnel linings 403

Figure 6.65 Sofrasar Sof-​S hear 375 Miami Bicone.

Note: Bicones were designed with a hard plastic outer cone that was injection molded over a
high strength steel center steel axis.

Construction began on the project by CBNA on 24 May 2010, tunneling began on 11

Nov 2011 and was completed on 06 May 2013. The project received many awards for
engineering and construction excellence and exceeded its planned benefits for the City of
Miami, FL.
404 Handbook of Precast Segmental Tunnel Lining Systems

Figure 6.66 Radial (longitudinal) joint detail with guide rods.

Note: Drawing from Port of Miami Highway Tunnel design package presenting radial (longitu-
dinal) joint connection via the guide rod.

BIBLIOGRAPHY AND READING LISTS

For preparation of this chapter, the following references were used, which are grouped into three cat-
egories of codes and standards, tunnel industry research documents and patents.

CODES AND STANDARDS

Guide for Precast Concrete Tunnel Segments. American Concrete Institute, Committee 533, Report
533.5R-​20, USA, April 2020, 84 pages.
ISO 14713-​1:2017: Zinc Coatings –​Guidelines and Recommendations for the Protection Against
Corrosion of Iron and Steel in Structures –​Part 1: General Principles of Design and Corrosion
Resistance. International Organization for Standardization (ISO).
 Connections and accessories for segmental tunnel linings 405

Figure 6.67 Circumferential joint with bolts.

Note: Typical detail on circumferential joint with bolting materials and key dimensions.

Figure 6.68 Circumferential joint with bolts installed.

Note: Circumferential joint with bolt installed and key dimensions.

Norm NF E25-​030: Fasteners–​Threaded Connections with ISO Metric Thread–​Part 1: Design Rules
for Tensile Loaded Bolted Joints–​Simplified Procedure. Association Française de Normalisation
(AFNOR).

TUNNEL INDUSTRY RESEARCH DOCUMENTS

Guideline for Concrete Segmental Lining Systems, 2011. Austrian Society for Concrete and
Construction Technology (ÖVBB).
Lining Segment Design: Recommendations for the Design, Production, and Installation of Segmental
Rings (October 2013). Deutsche Ausschuss für Unterirdisches Bauen e.V. (DAUB), German
Tunnelling Committee.
Recommendation for the Design, Sizing and Construction of Precast Concrete Segments Installed
at the Rear of a Tunnel Boring Machine (TBM)s, 2005. French Tunneling and Underground
Engineering Association (AFTES).
Recommendations for Gasket Frames in Segmental Tunnel Linings, Research +​Practice, Volume 54,
Empfehlung für Dichtungsrahmen in Tübbingauskleidungen. Studiengesellschaft für Tunnel
und Verkehrsanlagen e. V. (STUVA).
Tunnel Lining Design Guide, 2004. British Tunneling Society (BTS) and The Institution of Civil
Engineers (ICE).
406 Handbook of Precast Segmental Tunnel Lining Systems

PATENT DOCUMENTS
United States Design Patent

COUPLING
• Patent No: US D930,464 S
• Inventer: Sandrine Ordener
• Date of Patent: Sep. 14, 2021
Chapter 7

Gasket systems for sealing segmental

tunnel linings
Lead author: Andreas Diener
Contributing author: Peter Tiedermann

7.1 HISTORICAL BACKGROUND
The history of gasketed tunnel linings goes back to the year 1874 when cast iron segments
were used to construct underground metro tunnels in London, England. Ordinary lead was
used as a sealing material, which was mortised into the joints between the segments. Cast
iron segments were commonly used in shield-​driven tunnel construction until the 1970s.
The first rubber-​made compression sealing gaskets were used on cast iron segments
designed for the construction of the New Elbe River Crossing Highway Tunnels in Hamburg,
Germany. See Figure 7.1 below. Construction of the ramp sections of these three shield-​
driven tunnel tubes started in 1968 and were completed in 1974. Ready corner-​vulcanized
chloroprene rubber gasket frames made of a comb-​shaped sealing profile were used at that
time. This project is known as a milestone for the TBM-​driven tunnel method using seg-
mental tunnel linings equipped with rubber sealing gaskets. It opened-​up a new area in
modern mechanized tunnel construction.
In the late 1970s and early 1980s cast iron segments were increasingly replaced by rein-
forced concrete segments and comb-​shaped profile geometries on the segment gaskets were
replaced by hollow chamber profile geometries that could cope with larger building toler-
ances regarding gaps and offsets in the segment joints. Chloroprene elastomer was replaced
by ethylene-​propylene-​diene-​monomer (EPDM) elastomer, a material which offered a better
durability and long-​time performance at lower cost. Refer to Figures 7.2 and 7.3.

7.2 DESIGN AND CONSTRUCTION CONSIDERATIONS FOR SEGMENT

GASKETS
Segment gaskets are made of extruded and vulcanized rubber compression sealing pro-
files. They are positioned around the individual tunnel lining segment in the form of a
corner-​vulcanized frame and primarily near the lining extrados to provide the joint tightness
against water seepage from the surrounding ground.
By this positioning and sealing arrangement every segment joint is equipped with two
opposite gaskets having contact to each other, filling-​up and closing the joints between the
segments. Refer to Figures 7.4 and 7.5.
The watertightness is generated by the restoring (reaction) forces developing from the con-
tact pressure and deformation of the gaskets when being compressed and embedded into
the gasket grooves around the segments. The higher the rate of gasket deformation and the

DOI: 10.1201/9781032675541-7
408 Handbook of Precast Segmental Tunnel Lining Systems

Figure 7.1 Cast iron segments used on the new Elbe River Tunnel in Hamburg, 1968 to 1974.
Note: Construction of New Elbe River Crossing Tunnel using cast iron segments.

Figure 7.2 Comb-​s haped gasket design.

Note: Comb-​shaped gasket design developed in 1969 and predominantly used on cast iron
segments.
 Gasket systems for sealing segmental tunnel linings 409

Figure 7.3 Hollow-​c hamber gasket design.

Note: Hollow-​c hamber gasket design developed in 1985 and used in concrete segmental tunnel
linings covering larger build tolerances and avoiding high gasket restoring (reaction) force.

Figure 7.4 Typical concrete tunnel segment with gasket. Figure 7.5 Gasket groove with
gasket profile.
Note: Typical concrete tunnel segment with gasket groove
positioned near the extrados.
410 Handbook of Precast Segmental Tunnel Lining Systems

higher the utilization of the gasket groove volume by the compressed gasket rubber volume,
the higher is the restoring force and respectively the watertightness. Thus, the design of the
corresponding gasket groove geometry and joint configuration on the segments play a major
role in addition to the design of the gasket profile geometry.
The design and dimensions of both the gasket groove and gasket profile geometry are
mainly dependent on the waterproofing requirements of the tunnel project and must always
be considered together. The main criteria are the maximum allowable ring building toler-
ances in terms of gaps and offsets in the circumferential and longitudinal segment joints in
conjunction with the applicable water pressure.
In this regard, the tunnel designer’s desire for small gasket grooves often conflicts with
the required performance capabilities of the gasket. In general terms, the higher the water-
proofing requirements, the larger must be the dimensions of the gasket groove and profile.
In principle, wider gasket grooves and profile geometries can be found on larger tunnels as
the segment width is usually increasing together with the increasing tunnel diameter. More
space is available on the segment face to design a wider gasket groove without taking too
much risk that the distance between the gasket groove and the edge on the lining extrados
is becoming too small, which can result in potential concrete edge spalling.
On large diameter tunnels with an exceptionally large lining thickness, it can be possible
and feasible to place a second gasket near to the intrados lining face. Such double-​gasket
designs have been used on a small number of outstanding traffic tunnel projects with special
waterproofing requirements.

7.3 GASKET PROFILE DESIGN

Today the sealing profile geometries are designed with comb-​shaped ribs on the bottom
sections as being a relic of the first profile generation from the 1960s and 1970s. These ribs
provide support against the gasket groove bottom and perform as sealing barriers when
glued onto the segment. In addition to the lower rib sections all profile geometries found
today usually have rows of hollow chambers on the upper profile section. Hereby a dis-
tinction is made between single-​rows of hollow chambers and multiple-​rows of hollow
chambers. The latter is found on higher-​profile shapes designed for deeper gasket groove
geometries and these can usually cope with higher water pressures at larger gaps between
the segments. The larger the specified maximum gap the larger the height of the sealing pro-
file must be. Refer to Figure 7.6.
In general, it is also important that the volume of the compressed gasket profile is smaller
than the minimum volume of the corresponding gasket groove geometry in the event of the
smallest possible gap, between the segments (defined as minimum possible GBD (groove
bottom distance).
A massive cross-​section of the two deformed gasket profiles exceeding the volume cross-​
section of the gasket geometries at the moment of reaching complete closure of segment joint
during segment installation will cause a major increase of the gasket’s restoring force. Refer
to Figures 7.7 and 7.8. In extreme cases, this can cause immediate concrete spalling, pre­­
dominantly occurring at the gasket groove shoulder on the lining extrados. As an example,
the examination of the groove volume utilization is affected on the following basis:

Design criteria

• Groove bottom width 44 mm (1.73 inches)

• Groove depth 13 mm (0.51 inches) (effective)
• Flank angle 15°
 Gasket systems for sealing segmental tunnel linings 411

Figure 7.6 Typical state-​o f-​t he-​a rt gasket profile.

Note: Gasket profile geometry with hollow chambers and specific structure of ligaments and fil-
lets for glued-​o n application method.

Figure 7.7 Gasket profile in contact.

Note: Groove bottom distance (GBD) =​ 40 mm.

Groove area

Agroove =​tan (15 +​13 +​44) =​617.3 mm2 (0.957 in2)

Profile area

AProfile =​578.6 mm2 (0.896 in2)

Groove filling (area utilization in groove)

Aprofile/​Agroove =​93.7% (<100%)

On complete compression [gap =​0 mm (0 inches), groove bottom distance =​26 mm

(1.0 inches)], the area utilization in the groove is as shown below in Figure 7.9.
The gasket groove volume utilization should be preferably between 90% and 95% taking
account of the corresponding tolerances in the concrete gasket groove and on the gasket
profile. Further to that, reserves are desirable for the event of ring ovalization effects (bird-
smouthing) and under consideration of the critical T-​joints where gasket frame corners with
a higher volume of cross-​sectional gasket profile meet.
412 Handbook of Precast Segmental Tunnel Lining Systems

Figure 7.8 Gasket profile in complete compression.

Note: Groove bottom distance (GBD) =​ 26 mm.

7.4 GASKET GROOVE DESIGN

The gasket groove width is the decisive parameter for the maximum allowable offset (dis-
placement) between the segments. The greater the groove and profile width, the larger the
maximum allowable offset can be. The maximum allowable gap can be influenced by the
groove depth. Here again, the greater the groove depth, the larger the maximum allowable
gap can be. The groove depth is closely related to the required watertightness, which is gen-
erally specified by the surroundings of the underground structure. Once the requirements
concerning gap, offset and water pressure have been set, it would, in principle, be possible
to design the optimal gasket profile geometry for this purpose. However, over the course of
years, the following standard groove dimensions have become established for certain cat-
egories of tunneling projects. Refer to the data listed in Table 7.1.
Naturally different groove and profile geometries may be used for special project
requirements.
An increasingly discussed concern is the event of a segmental ring ovalization that result
in sloping longitudinal (radial) segment joint faces. Some designers call that event “bird-
smouthing”. Besides the risk of a gasket over-​compression, there can be an additional risk
of concrete spalling when the segment edges on the lining extrados make contact with each
other. Refer to Figure 7.10.
For that reason, the gasket groove shoulders are usually designed with a continuous,
sometimes sloping, recess on the lining extrados and a confined and much smaller recess on
the inside gasket groove shoulder. That is why experienced and well-​recognized gasket man-
ufacturers usually communicate the most favorable lay-​out-​design groove depth for their
proposed gasket profile and recommend that as “best-​case-​scenario” with regard to the
maximum possible waterproofing performance. Changes on that groove shoulder heights
made by the segment designers for reasons as mentioned before mostly have an influence on
the waterproofing performance, especially when tested at an offset scenario.
In general, the higher the gasket groove shoulders are, the better the waterproofing per-
formance of the gasket will be especially at large offset scenarios. However, the final gasket
 Gasket systems for sealing segmental tunnel linings 413

Figure 7.9 Groove filling or area utilization in groove.

Note: Typical graphics of groove filling by compressed gasket profile as a function of groove bot-
tom distance.

Table 7.1 Typical gasket groove dimensions.

Typical gasket groove dimensions with corresponding ranges of gaps and offsets and
maximum water test pressure established as industry standard over recent decades

Groove Average gap Average allowed Maximum water Tunnel categories

dimensions range offsets test pressure in diameter
Item Metric Imp. Metric Imp. Metric Imp. Metric Imp. Metric Imp.
1 26 × 1.0 × 3 to 0.125 10 mm 0.40“ 12 bar 175 3 to 9.8‘ to
10 mm 0.4“ 5 mm to psi 6m 19.7‘
0.20“
2 33 × 1.3 × 3 to 0.125 15 mm 0.60“ 12 bar 175 5 to 16.4‘ to
10 mm 0.4“ 5 mm to psi 10 m 32.8‘
0.20“
3 36 × 1.4 × 3 to 0.125 15 mm 0.60“ 20 bar 290 8 to 26.2‘ to
12.5 mm 0.5“ 5 mm to psi 12 m 39.4‘
0.20“
4 44 × 12/​ 1.7 × 4 to 6/​ 0.16 20 mm 0.78“ 33 bar 480 9 to 29.5‘ to
13 mm 0.5“ 8 mm to psi 17m 55.8‘
0.25“
414 Handbook of Precast Segmental Tunnel Lining Systems

Figure 7.10 Typical gasket groove design with sloped edges.

Note: Typical gasket groove design with sloped edges on extrados side avoiding contact and con-
crete flaking caused by ring ovalization (birdsmouthing) effect.

groove design is mostly a compromise solution under consideration of all these factors in
which case the leading designer needs to consult with the gasket manufacturer of record.

7.5 MANUFACTURING CONSIDERATIONS FOR GASKETS

Segment gaskets are made of elastomer and are manufactured by an extrusion process. Refer
to Figures 7.11 and 7.12. The material is irreversibly interlinked by means of vulcanization,
creating permanent elastic properties. In a second work step, the extruded gasket profiles
are cut in lengths while in a third work step they are spliced to ready corner-​vulcanized
gasket frames. See Figure 7.13. That is usually done by nozzle-​injection-​molding procedure,
whereby each of the gasket frames is tailor made to dimensionally fit on the individual tun-
nel segment.

7.6 GASKET MATERIALS
The elastomer material mainly used for gaskets is EPDM (ethylene-​ propylene-​diene-​
monomer). It is a synthetic rubber that has outstanding characteristics in terms of aging,
stress-​relaxation and chemical durability.
Economically and technically, EPDM elastomer is the most suitable material and best
possible compromise to meet the demand for a design life of up to 125 years. EPDM has
excellent resistance to the key drivers provoking the aging process which are ozone, oxygen
 Gasket systems for sealing segmental tunnel linings 415

Figure 7.11 Gasket vulcanization press.

Note: Vulcanization press for gasket frame corner-​injection-​m olding procedure.

Figure 7.12 Gasket extrusion machine.

Note: Extrusion machine for the manufacture of tunnel gasket profile geometry.

and ultraviolet (UV) radiation. EPDM also has a good resistance to high temperatures [up
to 150°C (302°F) short-​term and up to 50°C (122°F) long-​term]. Therefore, EPDM gaskets
installed onto the segments can withstand atmospheric exposure during long periods of
unprotected outdoor storage in the segment stockyard without showing any deterioration
and change in physical properties.
416 Handbook of Precast Segmental Tunnel Lining Systems

Figure 7.13 Typical gasket frame layout.

Note: Gasket frame consisting of extruded vulcanized sealing profile sections and injection-​
molded corner sections.

EPDM elastomer also offers a low degree of stress-​relaxation when constantly being
deformed and compressed in the gasket grooves alongside the joints of the segmental lining.
Further to that, EPDM elastomer has a very good permanent resistance to various aggres-
sive and highly concentrated chemical substances. It can resist substances with an acidity of
down to pH =​3. Refer to Figure 7.14.
EPDM elastomer also has a low gas flow rate (permeability), which can be an important
aspect for tunnel construction in strata containing methane gas. The resistance of EPDM
elastomer to hydrocarbons and mineral oil is limited, especially when tested in high concen-
trations and under permanent contact, respectively, under conditions of accelerated aging.
However, regarding the application situation of tunnel segment gaskets, EPDM elastomer
was approved and used on several tunnel projects in which soil and ground water contam-
ination with hydrocarbons, mineral oils and also tar sands have been serious concerns. An
installed tunnel segment gasket is embedded within the gasket grooves located in the closed
joints between the tunnel segments and behind a 100 to 200 mm (4 to 8 inches) thick layer
of grouting mortar. In this specific installation-​situation the gaskets normally do not come
in any direct contact with the contaminated soil.
Even in the case of a permanent infiltration and flow of groundwater seeping-​out of the
surrounding soil, percolating through small cracks in the grouting mortar and reaching the
gasket in the segment joints, the chance of contact with highly concentrated residues of such
substances over a wide contact surface is low.
Further to that, the concentrations of hydrocarbons, mineral oils and tar sands found on
recent projects were proportionally low, causing no concern for the resistance of the EPDM
gaskets.
 Gasket systems for sealing segmental tunnel linings 417

Figure 7.14 Typical gasket installation.

Note: Typical installation location of gasket near the extrados of the segmental lining surrounded
by layer of grout and soil.

An alternative elastomer material, which in theory offers a better resistance to hydrocar-

bons, oils, and tar, and from which tunnel gaskets could be manufactured, is a chloroprene/​
styrene butadiene rubber (CR/​SBR). However, CR/​SBR elastomers offer a much lower aging
resistance and a weaker stress-​relaxation performance. Further to that, CR/​SBR compounds
are currently much more expensive. For these reasons the use of CR/​SBR elastomer can only
be recommended in the proven event of direct and permanent contact of the gaskets with
high concentrated hydrocarbons, mineral oils and tar. Project-​specific site investigations
and analyses of the soil contamination are mandatory to judge if the use of this alternative
elastomer is necessary. Tunnel designers should obtain a comprehensive soil contamination
report prior to start of the gasket sealing system evaluation. Gasket manufacturers will
advise the designer if the evaluated product is applicable to the soil contaminations noted
in the project specific report.
 n

418 Handbook of Precast Segmental Tunnel Lining Systems

Physical property material tests on EPDM tunnel gaskets are recommended to ensure the
gasket was made of a high-​quality EPDM compound and the extruded sealing profiles have
reached sufficient grade of vulcanization. Such tests can also serve to check and monitor the
gasket production processes.
As an example, the German STUVA (Studiengesellschaft für Tunnel und Verkehrsanlagen
e.V., Cologne) specified the following physical property requirements within their
Recommendation for Gasket Frames in Segmental Tunnel Linings (published in ‛Forschung
+​Praxis, Volume 54, December 2019’) as summarized in Table 7.2.

7.7 WATERTIGHTNESS TESTING
The watertightness test is one of the elementary tests to check the suitability of the gasket
profile geometry and corner design. A project-​ specific approval test in the laboratory
should hereby simulate the segment design and installation conditions as closely as pos-
sible. Technically and economically, it makes sense to perform waterproofing tests using test
devices made of steel that simulate the T-​joint section of a tunnel lining, which is the most
critical area. Such T-​joint test devices have been used for tunnel gasket waterproofing tests
for more than 50 years. See Figures 7.15 and 7.16. They are common international industry
practice and are being recommended by authorities such as the following:

• STUVA Studiengesellschaft für Tunnel und Verkehrsanlagen e. V. (Germany)

• AFTES Association Française des Tunnels et de L’espace Souterrain (France)
• BTS British Tunnel Society (United Kingdom)
• ITA International Tunneling and Underground Space Association (Global).

The T-​joint test device comprises of a flat steel panel and two panels at an angle of 90°.
The T-​joint test device comprises of a flat steel panel and two panels at an angle of 90°.
The respective gasket groove design is milled into all three panels. Specially manufactured
testing frames are used for testing, also including “flat corners”. These flat corners do not
occur in tunnel design and must be suitably prepared for testing, in order to rule-​out pre-
mature failure in this irrelevant area. The relevant area of testing are the 90° shaped gasket
frame corners with the abutting profile sections. The T-​joint test device is designed to simu-
late different joint gap and offset scenarios by enlarging the distance between the panels
(gap) and shifting one of two angle panels out of the building line (offset).
A series of minimum two single tests performed per combination of gap and offset is
usually required (one test +​one repeat test). German STUVA recommends conducting an
additional third single test in the case of a large deviation in the achieved results to judge if
the lower result from the first two single tests has been an outlier.
The results of tightness testing are usually shown in a diagram such as that shown in
Figure 7.17 from which the groove bottom distance, the corresponding gap and offset of
each individual test set-​up is shown. It is also mandatory to show the gasket groove geom-
etry of the test device taking account of the initial gap. The initial gap may vary on the air
side and on the pressure side, depending on the groove geometry and recesses next to the
gasket groove. For project-​specific approval testing the gasket groove geometry and joint
configuration design on the T-​joint test device must be precisely similar to the one used on
the segments of the tunnel for which the gasket is tested for.
 newgenrtpdf
Table 7.2 Physical material properties of EPDM tunnel segment gaskets.
Requirements acc. to STUVA recommendation for gasket frames in segmental tunnel linings

Test method STUVA value Dimensions

Item Description DIN ISO ASTM Metric Imperial Metric Imperial
1 Shore A hardness 53505 7619 D 2240 66–​7 5 IRHD IRHD
(1 sec)
2 Tensile strength 53504   37 D 412 ≥9 MPa ≥1,300 psi N /​ mm 2 psi
3 Elongation at break 53505 37 D 412 ˃200 %

Gasket systems for sealing segmental tunnel linings

4 Accelerated aging 7 days at 70°C 53508 188 D 573
5 Artificial aging –​ change of shore A hardness 53505 7619 D 2240 –​5 up to 8 IRHD
(1 sec)
6 Artificial aging –​ change of tensile strength 53504   37 D 412 –​2 0 up to 10 %
7 Artificial aging –​ change of elongation at break 53504   37 D 412 –​3 0 up to 10 %
8 Compression set 22 hours at 70°C 53517 815 D 395 B ≤ 20 %
9 Relaxation test 3 months at 70°C 53509 1431A Less than 45% loss of stress %
10 Restorative capacity at 20°C 53509 1431A ˃ 90% %
11 Ozone resistance 53509 1431A No crack %
Notes:
ASTM American Society for Testing Materials
DIN Deutsches Institut für Normung E.V. (German National Standard)
ISO International Organization for Standardization, Geneva 20, Switzerland
IRHD International Rubber Hardness Degree
STUVA Studiengesellschaft für Tunnel und Verkehrsanlagen e. V.

419
420 Handbook of Precast Segmental Tunnel Lining Systems

Figure 7.15 T-​j oint test device.

Note: T-​joint test device for gasket waterproof testing.

Figure 7.16 T-​j oint waterproofing test.

Note: General arrangement for waterproofing test.
 Gasket systems for sealing segmental tunnel linings 421

Figure 7.17 Watertightness performance diagram of tunnel segment gasket.

Note: Typical watertightness diagrams provided for different gasket gaps and offsets.
422 Handbook of Precast Segmental Tunnel Lining Systems

7.8 GASKET RELAXATION AND FACTOR OF SAFETY

When constantly deformed, the restoring force and, therefore, the sealing abilities of a rub-
ber compression profile will decrease due to changes of functional locations of the molecular
structures. This physical process is defined as stress-​relaxation. Apart from possible degrad-
ation caused by chemical influences, the stress-​relaxation behavior of a compression profile
installed and embedded in the closed joints between the tunnel segment linings is the most
important factor to be considered with regard to the sealing performance over a life span of
100 years or even longer.
The loss of restoring force caused by the stress relaxation must be considered by deter-
mining a factor of safety for the T-​joint-​waterproofing test. According to the STUVA rec-
ommendations, the maximum “working” water pressure expected for the segmentally lined
tunnel structure has to be increased by a factor of “2” to cover the loss of the sealing
profile’s restoring force. According to the STUVA recommendations, the stress-​relaxation
simulation test must prove that the residual stress of the sealing profile will be ≥55% after
accelerated stress-​relaxation simulation over a period of 90 days.
The test device for the stress-​relaxation simulation test consists of two grooved steel bars
with a length of 100 mm. The ends of the bars are closed with end plates. See Figures 7.18
and 7.19.
Both steel bars are equipped with profile specimen of the same length. They are posi-
tioned opposite to each other without offset and at a defined gap. Depending on the profile
shape this gap corresponds to 30% of the difference between the sum of the nominal
heights of the two compressed profiles and the two related groove depths in the steel bars.
For the simulation of the long-​term behavior, heat is applied. During a test period of 90 days
under accelerated heat aging at 70°C (158°F) the test specimens are taken out of the heating
oven several times for measuring the force needed to compress the profiles by another
1.0 mm (0.03937 inches). For an extrapolation of the measured values for a period of
usually 100 years the WLF-​equation (developed by Williams, Landel, Ferry) is used. See
Figure 7.20 below.

Figure 7.18 Tests rails for gaskets.

Note: Test rail apparatus for stress relaxation tests on gasket profiles.
 Gasket systems for sealing segmental tunnel linings 423

Figure 7.19 Test rail assembly.

Note: Test rail assembly for stress relaxation tests on gasket profiles.

For specific projects, this extrapolation can be extended to a longer design life as it is
becoming increasingly common for owners and designers to look at extended design life of
up to up to 125 years.

7.9 GASKET LOAD-​DEFLECTION BEHAVIOR

The determination of the gasket’s restoring force is mainly important for the tunnel
designer for dimensioning the fasteners to be used on the circumferential and longitu-
dinal joints of the segmental lining. The fastener system must be laid-​out to countervail
the gasket’s restoring forces, in particular, in the circumferential segment joints and
from the point of time when the thrust jacks of the TBM will be removed from the
lastly built segmental ring. In addition to that, the segment erector must provide suffi-
cient forces to precompress the gaskets in the longitudinal joints during segment erec-
tion. In accordance with the STUVA recommendations, two each, 200 mm (8 inches)
long specimen of gasket profile are to be tested. Each of them is placed in a test rail
with the corresponding groove geometry. The edges of the test rails are to be closed
with end plates, in order to prevent axial gap expansion of the gaskets. The gaskets are
compressed without displacement, up to the steel contact of these rails. The result is
normally converted to a length of 1,000 mm (39.37 inches) and shown in a pressure-​
path diagram. See Figures 7.21 and 7.22. Usually, the restoring forces are determined
at ambient temperature.
Due to the fact that a substantial loss of stress is observed within the first minutes after
compression of the gaskets, it has become a common industry practice to keep the plungers
of the test device in the closed mode after having reached contact of the two test rails (pos-
ition at minimum GBD) and also measure the decline in restoring force over a period of five
continuous minutes. This decrease in restoring force is called “short-​term relaxation”. See
Figure 7.23.
424 Handbook of Precast Segmental Tunnel Lining Systems

Figure 7.20 Stress relaxation diagram.

Note: Typical stress relaxation diagram obtained from accelerated heat aging test and extrapo-
lation of measured values for a period equal to service life of tunnel using the WLF-​e quation.

Depending on the gasket profile geometry and the material shore hardness, the decline
in the restoring force within these five minutes holding time can be 25 to 40% compared
to the initial force measured during gasket compression. Erecting the tunnel segments
and building one complete lining ring usually takes more than five minutes. That gives
the designer some confidence that the fastening system has some substantial reserves
with regard to the required holding forces that are needed when removing the TBM
thrust jacks.

7.10 GASKET CORNER DESIGN

The injection-​molded gasket corner sections and the weld-​connections to the extruded pro-
file lengths on tunnel segment gaskets must fulfill the following important criteria:

• Guarantee a maximum possible waterproofing performance at extreme gap and offset

scenarios in the critical T-​joints
• Provide sufficient tensile strength and stiffness to withstand all kinds of mechan-
ical impacts and excessive stress scenarios during gasket and segment installation
procedures
 Gasket systems for sealing segmental tunnel linings 425

Figure 7.21 Gasket load-​d eflection diagram.

Note: Typical gasket load-​d efection or load-​groove bottom distance diagram obtained from
gasket compression tests when they are placed in test rails such as the one shown in Figure 7.22.

Figure 7.22 Testing machine for load-​d eflection analysis.

Note: Testing machine for load-​d eflection tests recommended by STUVA.
426 Handbook of Precast Segmental Tunnel Lining Systems

Figure 7.23 Short-​t erm relaxation diagram.

Note: Typical gasket load-​t ime diagram also known as short-​t erm stress relaxation diagram.

• Limit the punctual loads in the T-​joints developing from the increased restoring force
of the injection molded corner section to an extent that concrete spalling on the tunnel
segments will be avoided

The latter criterion contradicts the aforementioned and, therefore, requires a reasonable
compromise for the gasket corner design. See Figure 7.24 below. Hereby the interaction
with the adjoining gasket lengths as far as profile geometry and material composition are
concerned must be taken under consideration when designing the gasket corner detail.
In recent years, well-​established and experienced tunnel gasket manufacturers have
developed various design concepts and production methods seeking to minimize the cross-​
sectional volume of the injection-​molded corner sections as much as possible. In this regard
it is also important to minimize and control the material flow into the hollow chambers
of the abutting extruded profile sections during the injection molding procedure. See
Figures 7.25 and 7.26.
Gasket frame corners are usually designed with angles of 90°. Due to the flexible gasket
material, such 90° corners can usually absorb deviations in the segment edge angles (</​>
90°) to quite a large extent. However, depending on the profile size, the geometry and the
general gasket corner design, adjustments of the frame corner angles might become neces-
sary in the case of extreme degrees of acute and obtuse angles or/​and extreme rotations on
the segment corner. Such adjustments require additional tooling equipment and effort for
the gasket manufacturer.
 Gasket systems for sealing segmental tunnel linings 427

Figure 7.24 Gasket frame corner with an angle of 90°.

Note: Gasket frame corner designed to minimize the cross-​s ectional volume of the injection-​
molded corner sections.

Figure 7.25 Solid gasket frame corner.

Note: Solid gasket frame corner with reduced material infiltration into the hollow chambers of
the abutting profile sections.

Adjustment of gasket frame corners to acute and obtuse angles on the concrete segment
corners are shown in Figures 7.27 to 7.29.
Installed on segment corners with angles </​> 90°, elastomer sealing gaskets with standard
90° frame corners usually show slight deformations on the gasket corners in the form of a
contraction on an acute segment corner and a hump on the obtuse corner.
Within a tolerable range such deformations are usually just a visual phenomenon without
creating any technical problems, neither with regard to the segment installation procedure
nor to the sealing performance. Deviations within a range up to +​/​–​5 mm (0.20 inches) out
of alignment are to be tolerated and will be fully compensated within the gasket grooves
after segment installation with the gaskets being compressed. On segment corners with
angles ranging between 80° and 100°, gasket frame corners designed with standard 90°
corner angles usually show deviations, which are much smaller. See Figure 7.30. For seg­
ment corners with angles smaller than 70° and larger than 110°, the gasket corner angles
428 Handbook of Precast Segmental Tunnel Lining Systems

Figure 7.26 Gasket frame corner.

Note: Gasket frame corner with removed profile legs showing extent of material infiltration from
injection molding procedure.

Figure 7.27 Developed plan view of segmental ring –​ rectangular segmental ring arrangement.
Note: Specific angles of radial (longitudinal) side flanks on the key segment in a rectangular seg-
mental ring system that result in acute and obtuse corners on the key and counter key segments.

Figure 7.28 Developed plan view of segmental ring –​ rhomboidal segmental ring arrangement.
Note: Angle of radial (longitudinal) side flanks on trapezoidal/​p arallelogram shaped segments
in a universal rhomboidal segmental ring system that result in acute and obtuse corners on all
segments.
 Gasket systems for sealing segmental tunnel linings 429

Figure 7.29 Universal rhomboidal segmental ring system.

Note: Isometric rendering of a universal rhomboidal segmental ring design with trapezoidal and
parallelogram shaped segments.

Figure 7.30 Deviations drom 90° gasket frame corners.

Note: Deviations from 90° gasket frame corners that frequently occur on acute and obtuse seg-
ment corners.
430 Handbook of Precast Segmental Tunnel Lining Systems

Figure 7.31 Distorted angle on key segment.

Note: Distorted angle on key segment occurring when side flanks are not aligned to the circle
center.

must definitely be adjusted accordingly since the degree of deformation would become
intolerable and a proper bond to the gasket groove could be compromised (especially for
glued-​on gaskets).
Experience has shown that the shorter the molded gasket corner sections are, the less
deformation on the installed gasket will occur. In general, anchored (cast-​in) gasket frames
show less deformations on the gasket frame corners as well.
Depending on the key segment design, torsion on the segment corners can appear on the
key and the counter key segments. This torsion result in a distorted angle (angle ß) in the
gasket groove. See Figure 7.31. As shown in Figure 7.32, the gasket frame corner should be
adjusted to such distorted angles in case of being larger than 10°.

7.11 FIRE RESISTANCE
According to the German Standard DIN 4102-​1 B2 –​Fire Behaviour of Building Materials
and Components, tunnel segment gaskets made of extruded and vulcanized EPDM elastomer
are usually classified as being “conditionally permanent” to fire. However, in the event of
direct and unprotected contact to fire with the intensity, temperature and duration as we
have experienced in some disastrous tunnel accidents, the EPDM gasket would start burn-
ing quickly. The fire resistance can be substantially improved by the addition of flame pro-
tection additives. But this in general could have a negative impact on the physical material
properties and the durability of gaskets.
Therefore, fire rating requirements for tunnel segment gaskets are very rarely seen in tun-
nel project specifications. The common approach that has been accepted by project owners
and their designers is to look at the specific installation situation of a tunnel segment gasket.
Since the gaskets are mostly installed at the lining extrados and embedded within the
gasket grooves alongside the closed segment joints the chance of direct contact with fire
occurring from inside the tunnel is very low. This installation situation has been repeatedly
 Gasket systems for sealing segmental tunnel linings 431

Figure 7.32 Gasket frame corner with adjustment to distorted angle.

Note: Gasket frame corner with adjustment to the distorted angle ẞ because of distorted angle
on key segments.

Table 7.3 Result of segment fire tests at/​n ear gasket locations.

Results of precast segment fire test showing the developed
temperatures at specific measuring points

Measuring point 1 Measuring point 2

Item Time Metric Imperial Metric Imperial
1 After 5 minutes 30°C 86°F 50°C 122°F
2 After 10 minutes 30°C 86°F 55°C 131°F
3 After 15 minutes 30°C 86°F 60°C 140°F
4 After 20 minutes 30°C 86°F 60°C 140°F
5 After 25 minutes 30°C 86°F 60°C 140°F
6 After 30 minutes 30°C 86°F 60°C 140°F

investigated in simulation testing with different scenarios regarding lining thickness, dis-
tance of gasket groove and installed gaskets from lining intrados, fire temperature and dur-
ation. All these tests have shown that the temperatures developing at the area of the gasket
groove are absolutely non-​critical for the gasket and that no negative impact and damage
must be expected, provided that the fire can be extinguished within a period of approxi-
mately 30 minutes.
An example of the temperatures that have been measured in the gasket groove area and
published on one of such fire simulation tests are listed below in Table 7.3.
Results are related to a test performed on a 250 mm (10 inches) thick lining with a distance
between the intrados lining edge to the gasket groove of 190 mm (7.50 inches). Measuring
point 1 was located directly at the installed gasket in the gasket groove and measuring point
2 was located inside the joint gap close to the gasket (temperature of trapped air).
432 Handbook of Precast Segmental Tunnel Lining Systems

Figure 7.33 Burn test general arrangement at iBMB.

Note: Set-​
u p for a tunnel segment burn test at the Institut für Baustoffe, Massivbau und
Brandschutz (iBMB) in Braunschweig, Germany.

In this test, a burner was directly located a few centimeters below the segment simulating
a fire source from the intrados side developing a measured heat of 800°C (1,470°F). The
temperatures at the measuring points were detected after 30 minutes burning time.
Under very similar test conditions several project-​specific fire rating tests have been con-
ducted by specialized institutions such as the iBMB (Institut für Baustoffe, Massivbau und
Brandschutz) in Braunschweig, Germany. One of these tests has been performed in conjunc-
tion with the road tunnel project, “Herrentunnel Lübeck” in Germany in 2003 in which
450 mm (17.7 inch) thick segments and 44mm (1.72 inch) wide EPDM gaskets were used.
See Figure 7.33.
The main purpose of this test was to generally investigate the effect of plastic fibers to
improve the fire and heat protection as this was the time when companies such as Hochief
AG from Germany began to make very successful research and development implementing
this technology on concrete mixture used for tunnel lining segments.
For this test, the segment has been put directly over a burning chamber made of concrete
in which a fire with a temperature of 1,200°C (2,190°F) was produced. Within a burning
time of 60 minutes the highest temperatures reached around the gasket groove area were
below 100°C (212°F), which is absolutely uncritical for such a short-​term exposure on an
EPDM gasket.
No damage was observed on the compressed EPDM gaskets after 60 minutes burning
time and no changes in the physical material properties were found in subsequent labora-
tory testing on these gasket test specimens. EPDM gaskets can temporarily withstand tem-
peratures of up to 200°C (392°F) without any problems. With the latest fire life safety
standards and technologies already implemented in most developed countries and regions
today, the 30 minutes period as the most frequent requirement is technically a sufficient
time to put out a fire in a traffic tunnel.
Enormous efforts and improvements have been made after the world has seen several
disastrous accidents followed by deadly fires, such as in the Mont Blanc Tunnel, France in
1999 or in the Gotthard Tunnel, Switzerland in 2001.
 Gasket systems for sealing segmental tunnel linings 433

After the severe accident in the Eurotunnel in September 2008 in which a fire with tem-
peratures of up to 1,300°C (2,370°F) had occurred for a duration of several hours, massive
damage on the concrete segment linings was observed. On some of these damaged concrete
segments, some sections of gaskets were laid bare from the inside. Surprisingly, the gaskets
were still faultless and in good order and no leakage has occurred.
Whenever fire protection and safety are an absolute must inside a tunnel (which more
and more is the fact for nearly all traffic tunnels in the world), the market offers some very
effective and simple options to protect the tunnel structure against fire from the inside (fire
protection boards).
In 2005, the STUVAtec in Germany, for example, carried-​out fire tests on concrete
segments equipped with EPDM gaskets and fire protection panels, whereby a fire with
1,200°C (2,190°F) was produced in accordance with German ZTV-​ING-​Teil 5 Tunnelbau
(Zusätzliche Technische Vertragsbedingungen und Richtlinien für Ingenieurbauten) pub-
lished by German Authority BAST (Bundesanstalt für Straßenbau). In this test, neither the
concrete segments nor the gaskets have been in any way affected or damaged by the fire or
by high temperatures caused. In this case, the distance between the fire and the groove in
which the gaskets were installed was only 100 mm (4 inches), which is much less than in any
real tunnel. The time period of this test was 200 minutes, which was 85 minutes longer than
what is required by German authority for railway construction EBA (Eisenbahn-​Bundesamt)
and their published Guidelines for Fire Protection and Safety in Railway Tunnels.

7.12 GLUED-​O N GASKETS
From early on, the installation of tunnel segment gaskets onto the tunnel lining segments
has been carried-​out by different kinds of bonding procedures using adhesives with solvents
as well as special equipment. Both the “brush-​on” and the “spray-​on” adhesive methods are
still well-​established and approved procedures among the industry. See Figure 7.34 below.
Rubber-​based adhesives specifically developed for that purpose and offered by the leading
gasket manufacturers usually provide high and durable bonding strengths even in the event
of unprotected outdoor storage of the segments over a period of up to one year or longer.
However, these bonding procedures implicate the following risks and disadvantages:

• Cost involved with purchase and safe stocking of adhesives (potentially dan-
gerous goods)
• Cost involved with purchase and installation of spray guns and pressing frames
• Cost involved with protection measures for the workers and the environment
• Costs involved with recycling of empty adhesive tins (statutory in many countries)
• Segment groove repair measures required before gasket installation
• Process times for gasket installation and bonding of 15 to 20 minutes per segment
• High risk of leakage in the tunnel caused by insufficient gasket bonding due to
improper workmanship/​failures during gasket installation

7.13 ANCHORED GASKETS
In the early 1990’s, the first so-​called “anchored” tunnel segment gasket, which is directly
cast-​in the reinforced concrete segment was developed in Germany. For that purpose, the
gaskets were designed to be installed into a special gasket groove on the side parts inside the
segment formwork before casting. See Figures 7.35 and 7.36.
434 Handbook of Precast Segmental Tunnel Lining Systems

Figure 7.34 Glued gasket installation procedure.

Note: Glued gasket installation procedure at a typical gluing station in a segment production
facility with adhesive spray gun and pneumatic pressing frame.

Figure 7.35 Anchored gaskets –​ section.

Note: Section through anchored gaskets for cast-​in application in tunnel segments.

Successful field trials with this new gasket technology were carried-​out at that time in
Germany and Japan. In 2008, a 500 m (1,640 foot) long sewer tunnel section in Moscow,
Russia was successfully equipped with anchored gaskets. However, the first reference that
has received high recognition among the international tunnel industry and that is known as
 Gasket systems for sealing segmental tunnel linings 435

Figure 7.36 Anchored gaskets –​ arranged in segment formwork.

Note: Anchored gasket installed on the side parts of the segment formwork before casting
process.

the pilot project for this new gasket application technology has been the Lee Tunnel Project
completed in the year 2014 in London, Great Britain. This sewage and rainwater overflow
collector tunnel with an outer diameter of 8.5 m (27.6 feet) laid the foundation for anchored
tunnel segment gaskets becoming the new state-​of-​the-​art technology. From that time on,
anchored gaskets have increasingly been used in projects all around the world, predomin-
antly in regions with high requirements for environment protection and high labor costs.
In comparison to the traditional “glued-​on” gasket design and installation procedure the
anchored gasket technology provides the following advantages:

• Better waterproofing performance due to less failures on concrete groove side

• Better waterproofing performance due to additional barriers against water penetra-
tion underneath the sealing profile
• Saving of time and costs for gasket groove repair measures on concrete segments
before gasket installation
• Much better and safer bond to segments avoiding gaskets coming lose, especially
during key stone insertion (e.g. up to 5 times higher bonding force compared with
glued-​on gaskets)
• No issues with gaskets falling-​off the segments during outside storage due to insuffi-
cient gluing and influences of weathering (i.e. rain, frost, etc.)
• Better working conditions and higher safety for segment production staff due to abo-
lition of solvents (which are used in all standard brands of tunnel gasket installation
adhesives)
• Cost savings related to installation equipment (spray guns and pressing frames), pur-
chase of adhesives and cleaner, stock for hazardous goods, disposal of empty contain-
ers, protection measures and safety gear for workers, etc.
436 Handbook of Precast Segmental Tunnel Lining Systems

• Environment-​friendly workplace
• Decisive reduction of gasket installation process time (i.e. in average from 15 to 3
minutes, depending on segment/​gasket frame size)

Despite these advantages it must be emphasized that the involved formwork maker should
have sufficient experience and expertise on the special design required on the formwork’s
side parts. In this regard they must closely cooperate with the gasket manufacturer and tun-
nel designer and Engineer-​of-​Record. The best possible gasket insertion groove design on
the formwork is required to reduce the risk of damaging the segments during the formwork
stripping process and to prevent miscalculation of the effective gasket groove filling rate. In
this regard, the issue of concrete spalling during segmental ring assembly that might result
from gasket over-​compression, insufficient groove to edge distance or disadvantageous
gasket corner design must be looked at even more critically. In comparison to the “glued-​
on” gasket technology, the shape of anchored gasket profile defines the gasket groove on
the concrete side.
Further to that, a high degree of diligence and discipline is required to properly clean the
segment formwork after each casting cycle. Any smallest residue of concrete in the gasket
insertion grooves on the segment formwork will hinder a proper gasket installation and may
cause failures on the gasket’s embedment to the concrete during casting.
Limited repair procedures for such installation failures on anchored gaskets do exist, but
these methods require much more effort than repairing a defective mounting on a “glued-​
on” gasket and possibly require special approval by the project owner.

7.14 INSTALLATION OF GASKETED SEGMENTS INTO THE TUNNEL

Apart from the established procedures to install “glued-​on” and “anchored” gaskets onto
the tunnel segments, respectively, onto the segment casting formwork, there is another
important aspect to be considered in terms of gasket installation. At the point of time when
the tunnel segments, either equipped with “glued-​on” or with “cast-​in” gaskets, are being
erected and installed into the tunnel lining it is absolutely mandatory to lubricate the con-
tact surface of the gaskets. See Figures 7.37 and 7.38.
Surface friction between the gaskets must be reduced, preferably on all longitudinal
(radial) joint sides, indispensably on the longitudinal (radial) joint faces between the key
and counter key gaskets. Without proper lubrication, severe damage to the gaskets can
occur. Particularly during key segment insertion, high surface friction together with an
increasing degree of gasket compression can cause the gaskets to become detached and
dislocated within the gasket groove. That can result in concrete spalling and/​or leakages in
the critical T-​joint sections.
Without lubrication, the mechanical impact developing in terms of stress and tearing
force on the gaskets is extreme and can be multiple times higher than the bonding force of
the adhesion or anchoring system that shall keep the gaskets in place. Also, the degree of
key segment angle is playing an important role. The larger the key segment angles are being
designed, the sooner the contact between the gaskets will be made and the larger the degree
of contact compression and friction will become during insertion.
Several sorts of approved lubricants are available for that purpose and gasket manufac-
turers can provide recommendations or even offer their own products. An environmentally
friendly and highly efficient lubricant that is used on many tunnel construction sites is a
product called “green soap”.
 Gasket systems for sealing segmental tunnel linings 437

Figure 7.37 Key segment installation.

Note: Key segment being pushed-​in and brought into final position by the TBM thrust jacks.

Figure 7.38 Principle of key segment insertion.

Note: Principle of key segment insertion with relevant longitudinal (radial) joint sides where
contact friction between segment gaskets develops. Lubrication on these sections is mandatory
to avoid installation problems.

Some gasket manufacturers provide prelubricated gaskets as a ready factory provided

solution. But experience has shown that these special gliding surfaces mostly lose efficiency
over time, especially under the usual weathering effects during outdoor storage when
installed onto the segments. Lubrication freshly applied onto the gaskets shortly before the
segment installation is always the safest and most reliable option.
438 Handbook of Precast Segmental Tunnel Lining Systems

This subject is of major importance and one of the most under-​estimated issues in a seg-
mentally lined tunnel construction. Even after 50 years of using gasketed segment tunnel lin-
ing as a proven and well-​established construction method, severe failures repeatedly occur,
simply because of improper or missing lubrication on the segment gaskets.
Insufficient or missing gasket lubrication can easily be identified by the phenomenon of
gasket corners coming loose and being squeezed-​out of the radial joints or even stretched
around the segment edges when looking from behind the last built segment ring onto the
long circumferential side of the installed key segment. In worst cases, the gasket frame is
completely detached and pulled away from the segment groove and hanging loose from the
longer circumferential key segment side. Proper and reliable repair of such failures are not
possible. The only way to keep moving forward with TBM advance and segment ring build
is cutting the stretched and detached gasket section off, trying to bring both cut gasket ends
together again by a cold joint bonding. Leakage on such a repaired gasket section will occur
most likely. In this regard it should be mentioned that improper bonding of the gaskets onto
the segments caused by prior gasket installation failures in the segment factory can cause
a similar phenomenon as well or can increase the failures caused by insufficient or missing
lubrication on the gasket.

7.15 GASKET REPAIR
Repairs on gaskets installed onto the segments that have been damaged during transporta-
tion or in the segment stockpile yard are generally possible. Both, the “glued-​on” and the
“anchored” type of gaskets can be repaired by cutting and removing the damaged gasket
profile section from the segment and by properly inserting a replacement profile section by
using adhesive on the gasket groove and on the abutting joint sections.
Recognized gasket manufactures usually provide specific repair instructions and assis-
tance in terms of demonstration training on site. Project-​specific waterproofing tests on
such repaired gaskets have shown that the sealing performance can be proven as long as the
waterproofing requirements are in a moderate range. However, it must be emphasized that
a repaired gasket does not have the same properties as a virgin gasket frame being endlessly
vulcanized all around the segment. At the cold-​welded joints of the repaired section, it has
a weaker resistance to mechanical stress that might be caused during segment installation
procedures, which may lead to premature damage resulting in a leakage. In this regard, the
gasket manufacturers usually do not take responsibility and the acceptance of repaired gas-
kets and the final approval allowing segments with repaired gaskets going into the tunnel
must be provided by the owner’s Engineer-​of-​Record. Particularly in the case of high water-
proofing requirements, such an approval must be based on project-​specific waterproofing
tests on repaired gaskets.
Experience in the past has shown that a proper gasket repair is practically feasible and
has not evidently led to failures in the tunnel. However, it is absolutely mandatory to install
segments with repaired gaskets with utmost care and to use an approved lubricant that
prevents surface friction between the gaskets regardless of the segment type (key, counter
key or standard segment).
 Gasket systems for sealing segmental tunnel linings 439

7.16 PROJECT-​S PECIFIC GASKET PERFORMANCE TESTING

As emphasized in Section 7.3. “gasket groove design” the gasket groove geometry has a
major influence on the gasket’s performance. Standardized gasket groove geometries with
regard to groove shoulder heights and recesses next to the gasket grooves do not exist. From
project-​to-​project you mostly find different gasket groove geometries and changing joint
face configuration designs, disregarding a few exceptions.
In addition to that, the specified design water pressures and consequently the required
test pressures that must be proven change from project-​to-​project. For these reasons, it is
recommended to conduct project-​specific gasket performance testing to securely produce
an authentic proof of suitability. In this regard “project-​specific” means that all test equip-
ment used for performance testing must dimensionally be equipped with exactly the same
gasket groove geometry and nearby recesses including its distances from the gasket groove
as designed and found on the segments that will be going into the tunnel.
The German STUVA recommendation conducted the following project-​specific perform-
ance tests to demonstrate the suitability of a gasket design to get it approved for a project.

7.16.1 Short-​term waterproofing tests

By a series of several single test set-​ups, a variation of gap and offset scenarios that can occur in
the ring build has to be performed in a T-​joint test device made of steel. Details of the test device
and set-​up are described in Section 7.6 –​Watertightness testing and shown in Figures 7.15 and
7.16. In each of these single tests, the required project-​specific test pressure must be achieved
and held over a period of 20 hours without leakage occurring. STUVA recommends minimum
one repeat test for each single set-​up whereby the lower test pressure achieved is the relevant
proven test pressure result. In the case of large deviations in the results or a failure in one of
these two tests, STUVA allows to carry-​out a third single test to judge if a possible outlier has
occurred. Two of three single tests must prove the required test pressure.
Depending on the waterproofing requirements, the testing program can be reduced to the
most relevant gap and offset scenarios whereby the minimum requirement should be to per-
form at least two “worst-​case-​scenario-​tests”, which means both at the maximum allowed
gap and offset scenario. To estimate the additional reserves in the gasket’s sealing capacity
after having reached and kept the required test pressure for 20 hours it is recommended to
stepwise increase the water-​pressure until leakage occurs.
In the event of crucifix joints in the tunnel segmental lining it can be advisable to perform
special waterproofing tests in crucifix test devices. German STUVA recommends avoiding
crucifix joints in segmental lining design and construction as they can increase the risk of
leakages due to the fact that a greater variation and summation of possible ring building
tolerances (gaps and offsets) may occur. Past project-​specific gasket testing in crucifix test
devices have shown that under perfect ring building tolerances the waterproofing perform-
ance could have been improved rather than worsened. Basically, the experience from cru-
cifix testing is rare due to the low number of tests performed over the past decades.

7.16.2 Load-​deflection-​behavior test on the gasket profile

As described in Section 7.9 –​Gasket load-​deflection behavior, the restoring force in depend-
ency on the rate of gasket compression is investigated in load-​deflection behavior testing.
Hereby the maximum load required to fully compress two gasket profiles down to the min-
imum possible design groove bottom distance is the most interesting value for the tunnel
440 Handbook of Precast Segmental Tunnel Lining Systems

designer. In addition to that, it makes sense to also investigate the decline in restoring force
within the first 5 minutes after compressing the gaskets, which simulates the installation
situation within the first 5 minutes from erecting and installing the segments. German
STUVA recommends performing such load deflection tests minimum three times as project-​
specific suitability test. See Figures 7.39 and 7.40.

Figure 7.39 Test rails for load-​d eflection tests.

Note: Apparatus for load-​d eflection force testing on gasket corners simulating the application
situation in the critical T-​Joint.

Figure 7.40 Load-​d eflection testing machine.

Note: Testing machine with assembled test device during load-​d eflection test on gasket corners.
 Gasket systems for sealing segmental tunnel linings 441

7.16.3 Load-​deflection behavior test on the gasket frame corners

Due to the fact that the gasket corner sections usually have a higher rate of gasket groove
space filling, designers are always concerned about the increase of restoring force in the
T-​joint where two gasket frame corners are being compressed against each other. Concrete
spalling tests that simulate this specific application situation under laboratory conditions
are involved with high effort and costs. In addition to that, concrete test blocks usually
do not provide the same surface strength and reinforcement compared to the actual cast
segments used in the tunnel. This has created problems in some of the laboratory spalling
tests already carried-​out for tunnel projects in the past. For that reason, German STUVA
recommends conducting load-​deflection tests in T-​joint steel devices, which demonstrate the
increase and distribution of loads occurring in the specific location of a T-​joint.
Based on the measured restoring forces, experienced designers can usually consider the
risk of concrete spalling in these areas and can conclude if any measures are required,
such as increasing the gasket groove to edge distance, increasing the concrete strength, or
improving the reinforcement in this area.

7.16.4 Stress-​relaxation tests on the gasket profile

Based on stress-​relaxation simulation tests as described in Section 7.8 –​Gasket relaxation
and factor of safety, gasket manufacturers can provide expectations for the loss of stress
over the required tunnel design life (mostly 100 or 120 or even 125 years) made by extrapo-
lation in accordance with the WLF equation (Williams–​Landel–​Ferry theory). This is the
most commonly used practice, and this approach is also recommended by German STUVA.
The expected loss of stress should be covered by a factor of safety considered for the test
pressure that has to be defined and achieved in the project specific waterproofing tests.

7.17 GASKET QUALITY CONTROL

Over the past decades, recognized tunnel segment gasket manufacturers have installed self-​
monitoring systems, which secure the product quality throughout the complete project
supply time in terms of dimensional stability, continuity of material properties and gasket
corner stress resistance.
Usually, the following internal testing procedures are included in these self-​monitoring
checks:

• Permanent laser light control of the profile width and height during the complete
extrusion process
• Profile geometry dimensions checks on 10:1 scale overhead projector on samples fre-
quently being taken from the profile extrusion process (tolerances according to ISO
3302-​1, Class E2)
• Dimensional checks during profile cutting process
• Dimensional checks on ready vulcanized gasket frames
• Final visual check of all vulcanized gasket corners
• Destructive tearing force and tensile strength testing of frame corners by spot checks
• Tensile strength checks per DIN 53504 on the profile
• Elongation at break checks per DIN 53504 on the profile
• Compression set testing per DIN 53517 on the profile
• Shore hardness checks per DIN 53505 on the profile
442 Handbook of Precast Segmental Tunnel Lining Systems

The material property checks listed here can also be found in a chart defined as “Works
Production Controls –​Table 5” required within the STUVA “Recommendation for Gasket
Frames in Segmental Tunnel Linings”, published in December 2019 as shown in Figure 7.41.

7.18 NEW DEVELOPMENTS IN GASKET SYSTEMS

In more than 50 years of developing and using elastomer compression gaskets for seal-
ing the joints between tunnel lining segments the potential and technical possibilities for
new ideas, innovations and improvements are exhausted and has almost come to an end.
Fundamental changes with regard to the principle design and elastomer material are rarely
to be seen in the future, especially under consideration of economical and affordable manu-
facturing costs.
Recent new developments the industry has seen in the application of hybrid composite
gaskets combined with hydrophilic materials in specific markets and regions where owners
favorize and specify such alternative gasket designs. In markets where owners still trust and
rely on mono-​compression gaskets, new developments have occurred in the form of slight
improvements on the profile geometries.

7.18.1 Hybrid composite gasket designs

Since the early 1990’s “hybrid composite gaskets” have been developed with an additional
section made of a hydrophilic material that was inserted into a small location groove on the
profile’s top surface.
In the late 1990’s, hybrid profiles with a flat hydrophilic layer made by co-​extrusion pro-
cess have been developed and became very well established as an innovative technology
in Singapore and some years later in Malaysia and India as well. In these specifics regions
they have been successfully used on hundreds of kilometers of new metro and sewer tun-
nels. After more than 20 years in use, they are still being specified as the preferred gasket
technology.
However, co-​extruded composite gaskets never seriously made inroads into other markets
as they feature some disadvantages. Expanded hydrophilic material has a low-​grade density
and develops a noticeable restoring force in a fully embedded situation only. Installed in the
segment joints, thin hydrophilic layers covering the whole top surface of the profile can nei-
ther provide a very efficient restoring force nor sealing function by expanding due to a miss-
ing counter force on the left and right side of the compressed carrier profiles. Particularly
in the case of an offset scenario, a fully confined situation of the hydrophilic material is not
provided.
Furthermore, co-​extruded hybrid profiles come as a finished factory-​provided product.
Therefore, tunnel segments equipped with gaskets made of such co-​extruded profiles must
be protected against rain and humidity during storage in the open stockpile yard. Premature
swelling must be avoided as segments with expanded hybrid gaskets cannot be installed in
the tunnel.
Further to that the durability and long-​term performance of the hydrophilic materials used
on co-​extruded hybrid gaskets can neither be simulated nor proven in laboratory testing in
the same way as it is possible on EPDM compression gaskets. See Figures 7.42 and 7.43.
However, designers and owners in specific regions insist on using such “hybrid systems”
assuming an additional security provided by the hydrophilic material. Following-​up the
design approach from the early 1990’s, the use of hybrid gaskets makes sense in the form of
 Gasket systems for sealing segmental tunnel linings 443

Figure 7.41 Works production controls table recommended by STUVA.

Note: Detailed Works Production Controls as recommended by German STUVA, per
“Recommendation for Gasket Frames in Segmental Tunnel Linings” (Table 5).
444 Handbook of Precast Segmental Tunnel Lining Systems

Figure 7.42 Hybrid segment gasket.

Note: Photo of hybrid gasket consisting of an EPDM compression carrier profile and an inserted
hydrophilic round cord.

Figure 7.43 Hybrid segment gasket –​ section.

Note: Sketch of hybrid gasket cross-​s ection consisting of an EPDM compression carrier profile
and an inserted hydrophilic round cord.

a compression carrier gasket with a separate hydrophilic round cord, which is completely
embedded within the carrier profile, and which can be installed shortly before transporta-
tion of the segments down into the tunnel.
The hydrophilic cord would only be activated in the case of contact with water, for
example, in the case of a leakage through insufficiently compressed gaskets due to bad ring
build (in the event of extremely large gaps in the segment joints). In this case and under
certain advantageous circumstances, that leakage can be stopped within a period of two
to three days due to the additional restoring force developed by the expanding hydrophilic
cord. Such a “self-​healing-​effect” has been observed in laboratory testing. Fully embedded
into the groove on top of the compressed carrier profile, such a restoring force developed by
expanding and pushing against the surrounding barriers can provide an efficient additional
sealing performance over time.
This hybrid gasket design approach has partially become established in metro tunnel
construction in France where the authority AFTES (Association Française des Tunnels et de
 Gasket systems for sealing segmental tunnel linings 445

L’Espace Souterrain) requests a “double-​barrier sealing system” (EPDM compression profile

plus a separate hydrophilic strip) for segmentally lined railway and metro tunnels. However,
proven evidence that hybrid gaskets provide drier tunnels compared to pure “mono-​EPDM-​
compression-​gaskets” does not exist. As long as the gasket profile shape and the corre-
sponding gasket groove geometry can be dimensioned wide and high enough to cope with
all possible “worst-​case scenarios” concerning gaps and offsets in the segmental lining, the
use of a mono-​compression gasket without hydrophilic portion is preferred. In this regard,
material prices for any type of hydrophilic seal compared with the ones used for vulcanized
elastomer compression profiles are many times higher. And hybrid sealing systems require
special handling and care for storage and installation, which involves additional costs.

7.18.2 Gaskets used on corrosion protected segments

Increasing demands regarding durability and longevity of concrete segment linings used
in wastewater tunnels have recently led to various innovative technical developments and
industry solutions. Improved concrete technologies used in the form of “sandwich seg-
ment systems” consisting of two layers of different concrete mixes, whereby the main layer
provides the required strength and the smaller inside layer provides the corrosion protec-
tion by a special poly-​concrete mix are among the possible options that are successfully in
use. A project that has found international recognition is the twin 2 by 10 km (2 by 6.2
miles) long main wastewater collection tunnel forming a major part of the Emscher River
Renaturation Project in Germany. Glued-​on EPDM compression gaskets with a high resist-
ance to aggressive chemicals and in accordance to EN DIN 681-​1 have been used.
Another option are concrete segments with integrated corrosion protection inserts applied
during the segment casting process.
The latest product developments in Germany have shown that it makes sense to design
special segment sealing gaskets to become an integrated component of such corrosion pro-
tection panels. They are directly laminated into HDPE-​linings, which are placed on the bot-
tom of the segment casting formwork. After segment casting, they form a cast-​in gasket that
is either fully integrated to the HDPE-​linings or partially anchored to the concrete segment
as well, securely sealing and protecting the segment joints against aggressive wastewater
from the tunnel intrados side.
The first successful project references with integrated gasket systems used on HDPE-​
corrosion protection linings such as the Mississauga West Trunk Sewer Project in Ontario,
Canada, do exist and further developments on improving the gaskets design in terms of
installation procedures and economic efficiency are underway.

7.18.3 Mono-​c ompression caskets with improved profile geometry

designs
In 2017, a new gasket profile design has been developed for the construction of the DB
Albvorlandtunnel in Germany, which provenly offered a decisively higher water-​proofing
performance without specific offset-​related weaknesses and by simultaneously having a
lower restoring force. This improvement was made by a special archway-​shaped profile
design. That specific product development and related tunnel project are described in the
following chapter.
446 Handbook of Precast Segmental Tunnel Lining Systems

7.19 CASE HISTORIES OF MAJOR TUNNEL PROJECTS WITH

STRINGENT GASKET REQUIREMENTS

7.19.1 Deutsche Bahn Albvorland Tunnel near Stuttgart, Germany

The new railway-​high-​speed-​line between Wendlingen and Ulm, just 60 km (37.3 miles)
in total length, is currently under construction in the southeast of the Stuttgart metropol-
itan region as part of the expansion of the pan-​European high-​speed-​rail-​network. See
Figure 7.44. Half of this line, 30.4 km (18.9 miles) in all, will run underground, in 11 tun­
nels. The complete line that is closely related to the Stuttgart 21 project and that forms a
section of the west-​east-​axis between Paris (France) and Budapest (Hungary) is scheduled
to be put in service in 2022.
One of these major large diameter tunnel sections, the around 8.3 km (5.15 miles) long
Albvorlandtunnel near Kirchheim unter Teck was completed in 2019 and forms the final
tunnel link closing the gap between Stuttgart and the Swabian Alps, a 1,000 m (3,280 feet)
high volcanic mountain range. See Figure 7.45. The alignment of this tunnel passed under a
high-​pressure gas pipeline, a NATO fuel pipeline and, for a distance of 1.3 km (0.81 miles),
also the Stuttgart-​to-​Munich Autobahn. A maximum factor of safety was needed for this
project, with overburden of 65 m (213 feet) to a minimum of only just 9.50 m (31.2 feet)
and groundwater pressure of up to 45 m (148 feet).
This enhanced safety awareness was also reflected in the technical schedule of works for
the segment sealing gaskets in the context of the project tendering stage. The project client,
Deutsche Bahn Netz AG, had applied in advance a factor of safety allowance to the design
groundwater pressure for a planning of 4.0 bar (58 psi) and raised this figure to 5.0 bar
(72.5 psi), resulting in a test pressure of 10 bar (145 psi) in accordance with the Deutsche
Bahn Standard Ril 853/​ZTV.Ing and contractual conditions for the qualification certificate
to be furnished for the segment sealing gaskets.
This was to be demonstrated taking account of the initial design of joint gap and an
additional gap expansion of 6 mm (0.25 inch) and strictly in accordance with the provisions
of the German TL/​ TP DP (Technical Conditions of Supply and Technical Inspection
Regulations for Sealing Profiles) code of practice, i.e., with joint offsets of 0 to 20 mm (0 to
0.80 inches).

Figure 7.44 Deutsche Bahn Albvorland Tunnel near Stuttgart, Germany.

Note: The new east-​w est link of Europe’s high-​s peed-​r ailway network under development and
construction includes the Deutsche Bahn Albvorland Tunnel near Stuttgart, Germany.
 Gasket systems for sealing segmental tunnel linings 447

Figure 7.45 High-​s peed rail line between Stuttgart and Ulm, Germany.
Note: The new east-​w est link of Europe’s high-​s peed-​r ailway network under development and
construction includes the Deutsche Bahn Albvorland Tunnel near Stuttgart, Germany.

The contractor applied even further safety margins to the specification stipulated by
Deutsche Bahn Netz AG for the watertightness test certificate. The manufacturer of the
sealing gaskets was required to perform additional tests for a simulated “worst-​case” scen-
ario involving 8 mm (0.30 inches) joint gap and 20 mm (0.80 inch) joint offset.
Deutsche Bahn Netz AG specified a minimum contact width of 38 mm (1.5 inch) for
the sealing gasket profile geometry to be used, which closely approximated to a “classical
sealing” gasket profile for a groove-​base width of 44 mm (1.75 inch) as previously used in
various large diameter rail-​tunnel projects.
Based on experience and existing test reports in conjunction with project-​related approval
testing with such “classical” 44 mm (1.75 inch) sealing profiles, it had to be assumed that
with the test parameters specified by DB Netz AG, i.e. with a joint expansion of only 6 mm
(0.25 inch), a critical limit range would already be reached, especially when tested at an
offset range from 0 to 20 mm (0 to 0.8 inch) as specifically required for the Albvorlandtunnel
project. See Figure 7.46.
This included CTS Cordes Tubes & Seals GmbH & Co. KG to enter into the development
of a new suitable sealing profile geometry specifically for the test requirements applying to
the Albvorlandtunnel. After intensive research and development work and product tests
performed across a period of several months mainly at the recognized testing institute MPA
in Hannover (Materialprüfanstalt für das Bauwesen und Produktionstechnik Hannover),
in full accordance with the STUVA Recommendations, a totally new profile design concept
was achieved.
The advantages of this newly developed geometric concept can be found, essentially, in
the form of a considerably greater sealing effect, which exhibits no significant, geometrically
influenced fluctuations whatsoever across all joint offset dimensions as illustrated below
in Figure 7.47. It was possible to demonstrate fulfilment of the minimum watertightness
448 Handbook of Precast Segmental Tunnel Lining Systems

Figure 7.46 New archway-​s haped gasket profile design.

Note: New archway-​s haped gasket profile design developed for the Deutsche Bahn Albvorland
Tunnel.

requirements for “joint gap” of 6 mm (0.25 inch) in accordance with the project specifica-
tion across all offsets and with significant additional reserves.
The resultant overall concept for groove design and profile optimization permitted the
above-​described improvement in sealing action with no significant increase in the maximum
restoring force with the smallest joint gap. It was possible to achieve a test pressure of 10
bar (147 psi) even in the “worst-​case” tests performed additionally with a joint gap of 8 mm
(0.30 inch) [i.e. a groove-​bottom-​distance of 34 mm (1.33 inch)] and an offset of 20 mm
(0.80 inch).
In comparison with the former classic 44 mm (1.75 inch) profile shape CTS 44/​20, used
on large diameter railway tunnels, the improvements made with this improved profile geom-
etry design, CTS 44/​20 AVT (Albvorlandtunnel), can be summarized and concluded as listed
in Table 7.4.
This innovative development has proven that even after 50 years of designing segment
gasket profile geometries, accompanied with a large number of international patents granted
to the involved gasket manufacturers, there is still potential for improved and patentable
new concepts. Hereby, the improvements in the profile extrusion technology plays an
important role, which has made it possible for CTS Cordes to design and manufacture a
sophisticated profile geometry with an increased number of hollow chambers and a com-
plex arrangement of ligaments.
Another invention being used on this new geometry concept was the special archway-​
shape arrangement, which has been a key to achieve a high waterproofing performance
without offset related weaknesses and moderate restoring forces at the same time. Stored
segment for the Albvorland Tunnel are shown in Figure 7.48.
 Gasket systems for sealing segmental tunnel linings 449

Figure 7.47 Gasket water tightness diagram.

Note: Gasket water tightness diagram used in the design of gaskets.
450 Handbook of Precast Segmental Tunnel Lining Systems

Table 7.4 Characteristics and performances of standard and innovative gasket profiles.

Summary of criteria, performances and advantages of the newly developed geometric
concept of CTS 44/​2 0 AVT gasket profile in comparison with standard profile
solution

CTA 44/​2 0 (standard) CTS 44 /​ 20 (AVT)

Item Criteria and performance Metric Imperial Metric Imperial
1 Profile shapes

2 Basic groove design 12 × 44 mm 12 × 1.75” 13 × 44 mm 13 × 1.75”

3 Profile cross-​s ection 511.1 mm 2
0.792 in 2
521 mm 2
0.807 in 2
4 Maximum GBD at 6 mm gap 30 mm 1.18” 32 mm 1.26”
opening
5 Maximum GBD at 8 mm gap 32 mm 1.26” 34 mm 1.34”
opening
6 Groove volume usage 90.2% 93.7%
7 Initial load deflection force 76 kN/​m 5,210 lbs/​f t 69 kN/​m 4,730 lbs/​f t
8 Load deflection force after 5 63 kN/​m 4,315 lbs/​f t 48 kN/​m 3,290 lbs/​f t
minutes
9 Watertightness under 10 bar No Yes
pressure over 24 hours at 0 to
20mm offset
10 Pressure increase at 32 mm Up to Up to 218 Up to Up to 725
GBD 15 bar psi max 50 bar psi max
maximum maximum
11 Gasket corner design State of the art Improved

Project fact sheet

• Project Name Albvorlandtunnel/​N BS Wendlingen–​U lm PA 2.1 Lot A

German High-​S peed Rail Link between Stuttgart and Ulm
Part of the new European High-​S peed Rail Network
• Owner/​c lient DB Netz AG (German Railway Authority)
• Contractor Implenia Construction GmbH, Division of Implenia Group
• Length of tunnels 2 × 8.17 km (2 × 10.6 miles)
• Tunnel outer diameter 10.500 m (34.45 feet)
• Tunnel inner diameter 9.60 m (31.50 feet)
• Lining thickness 450 mm (17.7 inches)
• Segmental ring width 2,000 mm (6.6 feet)
• Segmental ring design seven segments per ring; four ordinary segments +​ two counter
keys +​ one large key segment
 Gasket systems for sealing segmental tunnel linings 451

Figure 7.48 Tunnel segment storage site at the Albvorland Tunnel.

Note: Tunnel segment stockyard at the construction site for Deutsche Bahn (DB) Albvorland
Tunnel in Kirchheim, Germany.

7.19.2 WSDOT SR 99 Alaska Way Tunnel in Seattle, Washington

Seattle’s elevated SR-​99 Alaskan Way had to be replaced by four-​lane double-​deck tun-
nel. The State of Washington Department of Transportation (WSDOT) recently com-
pleted construction of the SR 99 Alaska Way Tunnel, the largest highway tunnel built in
North America. The tunnel was 2,774 m (9,100 feet) long with a finished inside diameter
of 15.85 m (52.0 feet) to accommodate two lanes in each direction plus a safety lane.
See Figure 7.49. The tunnel was driven using a 17.52 m (57.5 ft) diameter Hitachi Zosen
EPB tunnel boring machine and supported by a one-​pass ten-​piece precast segmental
tunnel lining with an outside diameter of 17.0 m (56.0 ft). Width and thickness of each
segment are 2.0 m (6.5 feet) and 610 mm (2.0 feet), respectively. See Figures 7.50 and
7.51 below.
The performance requirements for the EBDM gaskets were a 7.5 bar working or 15
bar tested rating with an 8 mm (3/​8 inch) gap and 20 mm (7/​8 inch) offset. Because it
was a highway tunnel the owner required a secondary sealing system of a hydrophilic
system integral to the EPDM gasket. The hydrophilic gaskets had to resist all different
water types (distilled, tap and salt water) and comply with stringent maximum swelling
requirements.
The project requirements were met with the precasting company using Datwyler (now
Sealable Solutions) gasket profile M389 25 “Alaskan Way”, a special EPDM profile
452 Handbook of Precast Segmental Tunnel Lining Systems

Figure 7.49 WSDOT SR 99 Alaska Way Tunnel.

Note: Cross-​s ection of the 17.52 m (57.5 feet) outside diameter SR 99 Alaska Way Tunnel in
Seattle, Washington, United States.

Figure 7.50 Segment transport in the tunnel.

Note: Precast segmental tunnel linings being transported by multi-​f unction vehicles in the tunnel.

developed for the project with a 5 mm (0.25 inch) Black Swell (hydrophilic) round cord
integral to the gasket. See Figure 7.52. The EPDM gaskets were glued to each segment at
the casting plant in Puyallup, WA, and the hydrophilic round cord was installed on each
segment at the project site prior to each segment being transported into the tunnel.
The 44mm (1.75 inch) gasket was chosen based on the “worst-​case scenario” of 8 mm
(0.31 inch) gap and 20mm (0.80 inch) offset and tested with 15 bar (218 psi) pressure and
fulfilled this requirement and more. The actual gap open to water pressure during the test
was 11mm (0.43 inch); groove bottom distance of 32 mm (1.26 inches) with a shoulder
height of 10.5 mm (0.41 inches).
 Gasket systems for sealing segmental tunnel linings 453

Figure 7.51 Completed tunnel lining.

Note: View of the completed tunnel lining as the TBM advances. Rebar is in place for secondary
concrete.

Figure 7.52 WSDOT SR 99 Alaska Way Tunnel gasket profile.

Note: Cross-​s ection of the Datwyler/​s ealable solutions gasket profile M 389 25 with hydrophilic
round cord used on the SR 99 Alaska Way Tunnel in Seattle, Washington, United States.

Excessive testing required for selecting the best gasket for this project. Besides the demand-
ing watertightness test mentioned above, there were requests for swelling tests of the hydro-
philic material in all different water types (distilled, tap and salt water), external spalling
tests, chemical resistance of the gasket and hydrophilic material. These tests were required
in addition to the standard mandatory tests described in the specifications and typical for
North American tunnel projects.
Due to unexpected project delays during tunnel construction, the segments with installed
EPDM gaskets had to be stored outside for more than two years. This required further tests
on the gaskets, confirming that the gasket performance was not affected. The additional
tests showed that the EPDM compound showed no impact from the outside storage. On
another recent project in United States., EPDM gaskets have been stored outside in the sun
for over four years without impacting gasket performance.
Since many factors go into the creation of a dry tunnel, the Washington State Department
of Transportation (WASDOT), the project design engineers and the tunnel contractor evalu-
ated the various project requirements, conducted the required tests and selected a gasket
sealing system that met the needs of a very complex tunnel project.
454 Handbook of Precast Segmental Tunnel Lining Systems

The tunnel opened on 04 Feb 19 creating a bypass for auto and truck traffic through
downtown Seattle thus improving the transit time for the people of Seattle and creating
a waterfront park when the old viaduct was removed. Tunnels improve the quality of life
wherever they are constructed.
Project fact summary

• Project Name:   SR 99 Alaska Way Tunnel

Washington State Highway system in the Seattle city center. Tunnel
replacement for an obsolete two-​level viaduct.
• Owner/​client:   Washington State Department of Transportation (WSDOT)
• Contractor:   Dragados–​Tudor-​Perini
• Length of tunnel:   2,774 m   (9,100 feet)
• Tunnel outer diameter: 17.52 m   (57.5 feet)
• Tunnel inner diameter: 15.85 m   (52.0 feet)
• Lining thickness:   610 mm (24 inches)
• Segmental ring width: 2,000 mm (6.5 feet)
• Segmental ring design: ten segments per ring; seven ordinary segments, two counter
key segments +​one large key segment

BIBLIOGRAPHY AND READING LISTS

For preparation of this chapter, the following references were used, which are divided into two cat-
egories of authority published documents and authored documents.

INTERNATIONAL STANDARDS AND TESTING NORMS, GENEVA,

SWITZERLAND
ISO 37 Rubber, Vulcanized or Thermoplastic –​Determination of Tensile Stress-​Strain Properties
(Caoutchouc vulcanisé ou thermoplastique –​Détermination des caractéristiques de contrainte-​
déformation en traction).
ISO 188 Rubber, Vulcanized or Thermoplastic –​Accelerated Aging and Heat Resistance Tests.
ISO 815 Rubber, Vulcanized or Thermoplastic –​Determination of Compression Set; Part 1: At
Ambient or Elevated Temperatures.
ISO 1431A Rubber, Vulcanized or Thermoplastic –​Resistance to Ozone Cracking –​Part 1: Static and
Dynamic Strain Testing.
ISO 7619 Rubber, Vulcanized or Thermoplastic –​Determination of Indentation Hardness –​
Part 1: Durometer Method (Shore Hardness).
ISO 3302-​1 Class E2 Rubber –​Tolerances for Products –​Part 1: Dimensional Tolerances.

GERMAN STANDARDS AND TESTING NORMS

STUVA –​ STUDIENGESELLSCHAFT FÜR TUNNEL UND VERKEHRSANLAGEN E. V

Recommendations for Gasket Frames in Segmental Tunnel Linings, Research +​Practice,

Volume 54
Empfehlung für Dichtungsrahmen in Tübbingauskleidungen, Studiengesellschaft für Tunnel und
Verkehrsanlagen e. V. (STUVA), Köln.
Mathias-​Brüggen-​Straße 41, 50827 Köln /​Cologne, GERMANY.
 Gasket systems for sealing segmental tunnel linings 455

EISENBAHN-​B UNDESAMT (EBA) GERMAN AUTHORITY FOR RAILWAY CONSTRUCTION

Guidelines for Fire Protection and Safety in Railway Tunnels.

DEUTSCHE BAHN (DB)

Deutsche Bahn Standard Ril 853/​ZTV.Ing.–​

Richtlinie der Deutsche Netz AG, Zentrale I.NVT 4, Theodor-​Heuss-​Allee 7, 60486 Frankfurt.

BUNDESANSTALT FÜR STRASSENWESEN, BRÜDERSTRASSE 53, 51427 BERGISCH

GLADBACH

German ZTV-​ING-​Teil 5 Tunnelbau–​Zusätzliche Technische Vertragsbedingungen und Richtlinien

für Ingenieurbauten ZTV-​ ING Teil 5 Tunnelbau, published by BAST–​ Bundesanstalt für
Straßenwesen, Brüderstraße 53, 51427 Bergisch Gladbach.
German TL/​TP DP (Technical Conditions of Supply and Technical Inspection Regulations for Sealing
Profiles).
Technische Lieferbedingungen und Technische Prüfvorschriften für Dichtungsprofile, published by
BAST–​Bundesanstalt für Straßenwesen, Brüderstraße 53, 51427 Bergisch Gladbach.

DIN –​ DEUTSCHES INSTITUT FÜR NORMUNG E. V. (GERMAN NATIONAL STANDARD)

DIN EN 681-​1Elastomeric Seals–​Material Requirements for Pipe Joint Seals Used in Water and
Drainage Applications–​Part 1: Vulcanized Rubber; German Version EN 681-​1:1996 +​A1:1998
+​A2:2002 +​AC:2002 +​A3:2005.
(Elastomer-​Dichtungen–​Werkstoff-​Anforderungen für Rohrleitungs-​Dichtungen für Anwendungen in
der Wasserversorgung und Entwässerung–​Teil 1).
DIN 4102 B2 Fire Behaviour of Building Materials and Elements; Part 1: Classification of Building
Materials Requirements and Testing.
(Brandverhalten von Baustoffen und Bauteilen–​ Teil 1: Baustoffe; Begriffe, Anforderungen und
Prüfungen).
DIN 53504 Testing of Rubber–​Determination of Tensile Strength at Break, Tensile Stress at Yield,
Elongation at Break and Stress Values in a Tensile Test Standard.
(Prüfung von Kautschuk und Elastomeren–​ Bestimmung von Reißfestigkeit, Zugfestigkeit,
Reißdehnung und Spannungswerten im Zugversuch).
DIN 53505 Shore A and Shore D Hardness Testing of Rubber.
(Prüfung von Kautschuk und Elastomeren–​Härteprüfung nach Shore A und Shore D).
DIN 53508 Accelerated Aging of Rubber.
(Prüfung von Kautschuk und Elastomeren–​Künstliche Alterung).
DIN 53509 Determination of Resistance of Rubber to Ozone Cracking under Static Strain.
(Prüfung von Kautschuk und Elastomeren–​Bestimmung der Beständigkeit gegen Rissbildung unter
Ozoneinwirkung–​Teil 1: Statische Beanspruchung).
DIN 53517 Testing of Rubber and Elastomers; Determination of Compression Set after Constant
Strain.
(Prüfung von Kautschuk und Elastomeren; Bestimmung des Druckverformungsrestes nach konstanter
Verformung).

GERMAN TUNNELLING COMMITTEE (DAUB)

DAUB:2013 Lining Segment Design: Recommendations for the Design, Production, and Installation
of Segmental Rings.
456 Handbook of Precast Segmental Tunnel Lining Systems

BRITISH STANDARDS AND TESTING NORMS

BRITISH STANDARDS INSTITUTION (BSI)

BS PAS 8810:2016 Tunnel Design –​Design of Concrete Segmental Tunnel Linings –​Code of Practice.

BRITISH TUNNELLING SOCIETY (BTS) AND THE INSTITUTION OF CIVIL ENGINEERS (ICE)

BTS ICE 2004 Tunnel Lining Design Guide.

FRENCH STANDARDS AND TESTING NORMS

AFTES –​ ASSOCIATION FRANÇAISE DES TUNNELS ET DE L’ESPACE SOUTERRAIN

Recommendations for the Design, Production, and Installation of Segmental Ring–​Recommendation

for the Design, Sizing and Construction of Precast Concrete Segments Installed at the Rear of a
Tunnel Boring Machine (TBM)s (2005).

INTERNATIONAL TUNNELING AND UNDERGROUND SPACE

ASSOCIATION
ITA –​ INTERNATIONAL TUNNELING AND UNDERGROUND SPACE ASSOCIATION

ITA WG 2 2019 Guidelines for the Design of Segmental Tunnel Linings.

Chapter 8

Formwork systems for precast

segmental tunnel linings
Lead author: Stefan Medel
Contributing author: Florian Werres

8.1 INTRODUCTION
This chapter discusses segment mold design and manufacturing and all the aspects related to
segment production, including the equipment used in segment precasting plant and the typ-
ical configuration of the segment production process. It also explains the tolerances of the
mold and the advances measurement techniques for the mold and the segment. The segment
and mold quality control and documentation as well as innovation in segment production
and relevant case histories are also described in this chapter.

8.2 SEGMENT MOLD DESIGN

In this section, different types of ring geometry and various parameters affecting the mold
design are explained. Then, the configuration of the mold and its different components are
discussed and finally, the critical aspects of vibration and durability are described.

8.2.1 Ring designs
Once having defined the tunnel track, the geometrical dimensions as well as the tunneling
mode, the designer develops a concept for tunnel lining. The lining must meet or rather
guarantee the following requirements:

• Structural stability during tunneling and during operation

• Serviceability during operation
• Tightness against water pressure –​external or internal
• Fire protection

After the selection of the tunneling method and the tunnel boring machine (TBM) type, a
segmental lining concept must be defined, which includes the following:

• Tunnel diameter and segment thickness

• Ring type
• Number of segments per ring
• Requirements to tightness
• Requirements to the connection systems
• Concrete class, reinforcement, etc.
• Tunneling speed

DOI: 10.1201/9781032675541-8
458 Handbook of Precast Segmental Tunnel Lining Systems

Current state-​of-​the-​art ring types:

• Rings with one-​side taper

• These rings are often designed as left and right tapered rings
• Rings with both-​side taper
• This ring is also often called universal ring
• Rings with a small key segment
• In general, the standard segments are rectangular and only the key segments and
counter key segments have an angled joint.
• In most cases, the size of the key segment is only 1/​3rd of the standard seg-
ment length
• Rings with large key segment (or even several key segments)
• In this case the standard segments are generally designed in the shape of a paral-
lelogram and the key segments in trapezoid shape

Both rings with one-​side taper as well as those with both-​side taper can be designed with a
small or a large key segment, as shown in Figure 8.1.
The common designation for rings with a small key segment is X +​1. Universal rings;
i.e. rings with a large key segment are usually described with X +​0. Until a few years ago,
universal rings were used in most cases for smaller tunnel diameters, as the individual seg-
ment length was considerably shorter and, therefore, easier to handle in tight conditions.
However, this ring type is recently applied for almost any diameter and is getting more and
more important.
Principle sketch of ring types:

8.2.2 Mold design
The mold design is based on the ring or segment design, which must be defined by the
designer. However, the design focuses on the implementation of the static requirements,
i.e. bearing the bending moments and normal forces, the size of the contact faces as well
as initiation of point and line loads by bolting and EPDM gaskets. Therefore, the designer
considers all occurring loads and load cases as well as the states of construction.
After having approved the ring design for production of the lining segments, the mold
manufacturer checks the ring design with regard to geometrical feasibility and implementa-
tion for mold manufacture. For this purpose, the mold manufacturer prepares a 3D model
at the beginning of the project and compares it to the existing 2D/​3D model of the approved
ring design. Further, the mold manufacturer checks all drawings regarding:

• Possibility of demolding without clashes

• Reasonable demolding angles

In addition, the mold manufacturer adds adequate radii and transition zones.
Another aim is to make sure that opening of the mold does not cause any restraints that
could possibly damage the early age concrete.
Further tasks of the mold manufacturer:

• Verification of the correct position of the gasket joint and the connections between
rings and segments
 Formwork systems for precast segmental tunnel linings 459

Figure 8.1 Current state-​o f-​t he-​a rt for ring types.

• Matching of the segment angles and the positions of the thrust ram shoes with the
specifications of the TBM manufacturer
• Verification of the ring taper on the basis of the radii of the planned tunnel track

According to the current state of the art, mold design is developed in three-​dimensional
space (3D design), thus, on the one hand, enabling the designer to correctly visualize the
complex geometrical features –​particularly in non-​radial and angled surfaces –​and, on the
other hand, to simulate the movements of the side shutters and covers to detect possible
interfering edges.
In general, the mold manufacturer issues the following drawings:

• Drawings of the steel structure: they include all cutting measures and manufacturing
dimensions as well as positions and dimensions of the weld seams
• Drawings for mechanical processing: they include all information and measures for
milling as well as specifications regarding roughness of the machined surfaces
• Assembly drawings: these drawings include all information for assembly as well as
dimensions for inspection
460 Handbook of Precast Segmental Tunnel Lining Systems

In the first step, the tolerance requirements of the designer are decisive for the dimensions,
angles and measurement specifications implemented by the mold manufacturer. In the
next step, the mold manufacturer defines the tolerance necessary for manufacturing the
molds so that the produced concrete segments, i.e. the inside of the molds, show the correct
dimensions.

8.2.3 Implementation of molds
The molds are generally implemented with the following:

• Movable side shutters

• Movable front shutters
• Movable covers
• Rigid mold bottom

The side and front shutters can be moved by means of a tilting system or a horizontal dis-
placement. Depending on the mold manufacturer, there are different philosophies regarding
execution of the molds. Figure 8.2 shows different closing systems of the molds.

• The front shutters are closed first and then the side shutters are bolted
• This system offers the best adjustment and replicability in regard to the width of
the segment
• Or the side shutters are closed first and afterwards the short front shutters
• This system offers a better adjustment of the very important arc lengths and angles
in the longitudinal joints
• Further, shrinkage and creeping of the concrete in longitudinal direction can better
be controlled and probably re-​adjusted

Figure 8.2 Different closing systems of the molds.

 Formwork systems for precast segmental tunnel linings 461

Production of the lining segments is subject to very strict requirements regarding precision.
For example, a tolerance of ±0.6 to 0.8 mm (0.024” to 0.032”) is usual for the arc length.
The width of the segments is in most cases tolerated with ±0.5 mm (0.02”). For this reason,
high-​precision molds are required, which are able to permanently produce the required
dimensions and tolerances.
Re-​adjustable mold systems that can permanently guarantee the required dimensions, are
the preferred solution.
In general, tilting and bolting systems for opening, closing and locking of the side shutters
and covers are implemented. Spring-​supported lever-​arms are necessary especially for the
covers and front shutters to enable a manual movement of the components.
Design of a mold for stationary production generally does not differ from design of a
mold for carousel production. In the case of carousel molds, the base frame is additionally
equipped with wheelhouses and high-​quality wheels.

8.2.4 Vibration of the molds

External vibrators are installed below the mold bottom for distribution and compacting of
the concrete. Segment producers commonly use compressed air vibrators or electrical vibra-
tors with frequency converter.
There are currently no precise procedures in order to exactly determine the required
vibration energy in advance. Finite element models (FEM) mostly fail with modeling of
the concrete due to its different mix design and behavior during concreting and compact-
ing. Therefore, exact conclusions regarding behavior of the mold bottom during vibration
are also not possible. The number of vibrators is commonly determined using empirical
methods.
The decision between compressed air vibrators or electrical vibrators (as shown in
Figures 8.3 and 8.4) is in most cases made by the precast company itself. In recent years,
however, the following can be observed:

• Compressed air vibrators have been used for segment thickness up to 400 mm and
500 mm (15.75” to 19.60”)
• Electrical vibrators haven been used for segment thickness exceeding 400 mm to
500 mm (15.75” to 19.60”)

Figure 8.3 Electrical vibrator.

462 Handbook of Precast Segmental Tunnel Lining Systems

Figure 8.4 Pneumatic vibrator.

Control of compressed air vibrators is very restricted. Intensity can be varied via air supply
only. In contrast, electrical vibrators offer an exact adjustment of the frequency by means
of a converter.
However, experience has shown that the use of electrical vibrators quite often leads to
damage on the mold bottom. Possible causes include the following, for example,

• Appearance of frequency overlays

• Relatively high eccentricity with very low speed of rotation
• Very large or high frequencies

Due to the low cost of acquisition and operation combined with satisfactory compacting
results, compressed air vibrators are used in most of the cases.
Independent of the type of compacting, the type and location of frequency in the mold
bottom should be provided as directly as possible. Movement of the vibrator on the mold
bottom in almost any case leads to cracks and subsequently deformations. Therefore,
rigidity of the mold bottom must be sufficient, however, at the same time it has to be flexible
enough to enable the vibrations to be transferred to the concrete. Suitable steel plates, gen-
erally with a thickness between 10 mm and 12 mm (0.39” to 0.47”), must be used.

8.2.5 Durability of the molds

The type of design, the quality of manufacture and the intensity of their use have an
important influence on the durability of molds. We make a difference between the following:

• Serviceability
• Dimensional stability

Serviceability focuses on the permanent movability of the side and front shutters as well
as the covers. For this purpose, the mold manufacturer provides appropriate material and
greasing possibilities, respectively lubrication-​free bearings.
Dimensional stability is a product of precision in manufacture and careful use of the
mold. On average, approximately 600 to 1,000 segments are produced with one mold.
During this period, the mold is not allowed to exceed the required tolerances. Should the
 Formwork systems for precast segmental tunnel linings 463

tolerance be exceeded, the mold must be re-​adjusted. Several projects with up to 2,500 pro-
duction cycles per mold have been realized. These are exceptions.

8.3 SEGMENT MOLD MANUFACTURING

The mold is manufactured in a special workshop disposing of the required operating
resources and welding capacities. Mold manufacture is divided in the following steps:

8.3.1 Steel fabrication
After cutting and rolling of the sheets, the mold bottom is composed in a special gauge and
welded. The following has to be considered:

• Use of high-​quality steel –​S235, S355 for welded elements

• Steel sheet suitable for vibration for the mold bottom (tmin =​10 mm) (0.4”)
• The welding seams must be executed as intermittent welds in intervals

After inspection of the weld seam, a ground coat has to be applied on the lower surface of
the mold bottom. Figure 8.5 shows an example of steel fabrication.

8.3.2 Mechanical processing
During steel works and welding, the side and front shutters are exposed to a high-​
temperature impact. Due to thermal expansion this leads to increased tensions within the
material. Therefore, a thermal heat-​treatment; i.e. stress relief heat treatment at 550 to
600°C (1,022° to 1,112°F) is necessary before milling of the side shutters. Subsequently,
the steel parts must be sandblasted or shot-​blasted to be optimally prepared for mechanical
processing.
A common machine for mechanical processing is a five-​axle CNC milling machine with
a sufficiently big space for processing. After processing, the contact surfaces and the gasket
joint have to show a surface finish of Roughness, Ra 2.3. The use of a portal milling machine
could possibly be necessary for large side shutters or complex geometries.

Figure 8.5 Steel construction.

464 Handbook of Precast Segmental Tunnel Lining Systems

Figure 8.6 Milling of a side shutter for a segment mold.

Figure 8.7 Mold assembly.

Milling of the complex geometries in principle requires a lot of experience and care. An
example of the milling of side shutters is shown on Figure 8.6.
Ra is the arithmetic average of surface heights measured across a surface, Ra surface fin-
ish. Simply average the height across the microscopic peaks and valleys. Surface roughness
can be measured by a profilometer, a surface profile measurement tool.

8.3.3 Assembly
The molds must be assembled on an appropriate, even surface. Attention must be paid to
avoiding any torsion of the mold bottom by attaching the side and front shutters. Already
during assembly of the side and front shutters, maximum precision is necessary in order to
enable a subsequent precise adjustment of the surfaces between each other. Figure 8.7 shows
a mold assembly setup.
 Formwork systems for precast segmental tunnel linings 465

To be considered with regard to attachment of the parts:

• C 40/​50 unalloyed steel for pins, bolts

• Screws, nuts according to standard in Class 8.8 /​10.8

Steps of mold assembly

• Torsionless erection and fixing of the mold bottom

• Attachment of the swivel arms
• Assembly of the side shutter
• Assembly of the front shutter
• Assembly of the covers
• Adjustment of the side and front shutters
• Verification of the correct position by measuring
• Welding of the cones to the corner joints
• Inspection by measuring
• Installation of the vibrators
• Painting
• Preparation for transport

8.3.4 Test ring construction

The mold manufacturer usually produces a test ring in the factory (see Figures 8.8 and 8.9).
After concreting, the test ring is erected on a measuring surface. It is recommended to use
adjustable supports. The ring should be erected without EPDM gasket as compression of
the gaskets requires considerable forces.
Vertical erection of a test ring requires a lot of experience and needs quite some time.
The benefit of an erected ring, however, is very limited. The following information can be
obtained

Figure 8.8 Test ring for a metro project ID 9,110 mm.

466 Handbook of Precast Segmental Tunnel Lining Systems

Figure 8.9 Three test rings for a sewage project ID 5,500 mm.

Figure 8.10 3D measuring of the mold.

• Inner diameter (measured at several points regarding position and height)

• The outer circumferential length of the ring (measured in several heights)
• A visual inspection of the accuracy of fit of the joints (contact surfaces) and connec-
tions (bolting)
• Possibility of demolding of the segments without restraints/​damages

Further information cannot be gained. In the recent past, segment producers more and
more renounced the production of test rings and requested a virtual test ring instead.
To get a virtual test ring, the 3D measuring data of a mold are used to virtually build a
ring. Please also refer to Section 8.8 –​Master ring erection and virtual ring build.

8.3.5 Inspection
Before a mold leaves the manufacturer’s factory, a mechanical verification should be made
in any case. Today, a 3D measuring by means of a laser tracker is standard (see Figure 8.10).
The measuring report confirms the correct dimensions before transport.
 Formwork systems for precast segmental tunnel linings 467

Figure 8.11 Molds ready for transport.

8.3.6 Painting
Despite the relatively short service time of a mold, it is provided with a paint coat at the
manufacturer’s premises. It consists of adequate base coats and finishing coats. The painting
supports the necessary cleaning of the mold during operation. Besides the optical effect or
classification, the painting does not offer further functionality.

8.3.7 Transport
The mold manufacturer has to work out a concept for transport of the molds considering
the following aspects:

• How will the molds be shipped (container or truck)?

• Can the molds be stacked?
• What are the available loading widths and clearances?
• What is the loading capacity?

The molds have to be equipped with clearly marked positions for attachment, shackles or
forklift guides. If the molds are attached on other positions, torsions may cause a permanent
damage to the molds; an example of molds transport is shown in Figure 8.11.
If the molds are transported in stacks it must be made sure that the stack cannot move
within itself. For this, a static proof is necessary.
In sea containers the molds or stacks must be protected against slipping and tilting; i.e.
against any kind of movement.
With truck transport this is not possible in most of the cases. Here, fixing of the position
has to be made sure by means of lashing straps.

8.4 DESIGN OF SEGMENT PLANT LAYOUT AND MAIN COMPONENTS

When planning a segment production plant, we generally ask how many segments or
respectively rings will be required per day or week. Planning of the segment production
plant can only start when this question is clearly answered. However, definition of this
468 Handbook of Precast Segmental Tunnel Lining Systems

production curve is anything but easy and can never be considered as an absolute value.
Typical influencing parameters are as follows:

• Length of the tunnel

• Number of TBMs
• Possible concurrence of advances/​TBM running in parallel
• Planned interruptions; e.g. conveyor belt extension
• Cross-​sections/​cross-​passages
• Stations
• Tool replacement, etc.
• Unscheduled interruptions (geological or hydrological problems, machine damages, etc.)
• Available storage capacity
• Planned working time on the tunneling machine

Having defined the production quantities and the required storage capacity, the necessary
components of the segment plant must be planned. In general, these are the following:

• Segment factory/​field factory

• Batching plant with stocking of concrete aggregates
• Production of reinforcement cages
• Concrete laboratory for quality control
• Storage yard with cranes/​gantries (see Figure 8.12)
• Utility systems; i.e. water, sewage, electricity, gas/​oil, Wi-​Fi, phone, etc.
• Infrastructure requirements such as roads, drainage systems, pavement of storage
yard, etc.

When starting a project, the responsible planners should carefully check whether a new
plant is required or if experienced precasters are available in the region who could produce
a sufficient quantity of high-​quality segments.

Figure 8.12 Segment factory with storage yard.

 Formwork systems for precast segmental tunnel linings 469

8.4.1 Segment factory
The segment factory/​field factory (Figure 8.13) is generally installed in a steel building with
a concrete foundation. Depending on the surrounding conditions (temperatures, noise
emission), the factory should be insulated. The following utilities must be installed in the
building:

• Electricity
• Water
• Compressed air
• Optional sprinkler system
• Ventilation
• Wi-​Fi/​VoIP (Voice over Internet Protocol)

The overhead cranes (gantry cranes) should be able to safely pick-​up the loads (in most
cases, the total load of segment, handling device and lifting force must be considered), espe-
cially with regard to their minimum distance between crane rope and wall.
Besides the production area itself, the following additional areas must be provided:

• Curing/​cool down of the concrete

• Accessories (gaskets, plywood boards, etc.)
• Turning and evacuation
• Measuring of concrete segments (single or stacked)
• Stocking of consumables

Sufficient lighting has to be installed in the hall (light bands at the ceiling and on the walls,
ceiling lights) and access gates have to be planned.
A rescue concept defines the position of the emergency exits and helps to make all areas
accessible.
Fire loads have to be planned in advance and agreed with the responsible authorities.

Figure 8.13 View inside a segment factory.

470 Handbook of Precast Segmental Tunnel Lining Systems

8.4.2 Batching plant with stocking of concrete aggregates

The position of the batching plant in most cases results from the production mode, however,
it is best located directly besides the segment factory. A short distance is recommended, as
the concrete for the segments is commonly filled in the mold at a temperature of +​25° to
+​28°C (77° to 82°F) and as the effect of the concrete admixtures (superplasticizer, etc.) is
limited in time and example of a batching plant is shown on Figure 8.14.
As a standard, mobile batching plants with the following characteristics are used:

• 1.5 to 3.0 m³ (2.0 to 4.0 CY) mixing volume

• Pan mixer or twin-​shaft mixer

The number and type of silos depends on the concrete concept. In most cases, the following
is necessary:

• Cement silos with screw conveyor

• Silos for fly ash or slag sand
• Tanks for superplasticizer

The aggregates (sand, fine and coarse gravel) must be stored separately in bunkers/​boxes.
Depending on weather conditions it is recommended to provide a housing for the aggre-
gates. The size of the bunkers/​boxes should be sufficient to bridge a failure of supply or a
series of holidays. The same applies to cement, fly ash, etc.
For the case that steel fibers or fire protection fibers are used, delivery of the fibers into
the mixer must be well planned in advance. The fibers can be dispensed into the mixer dir-
ectly or via the conveyor belt or the inclined concrete transport hopper. As in most cases

Figure 8.14 Batching plant.

 Formwork systems for precast segmental tunnel linings 471

steel fibers are supplied in big packs, sufficient access and lifting space has to be planned.
Further, access areas for cement and aggregates must be considered when planning the
batching plant.
Finally, a concept for the treatment of concrete sewage and concrete residue should be
drawn up already in the beginning. Consumption of water and electricity should be agreed
with the local providers.

8.4.3 Reinforcement production factory

Also, in present projects conventional steel bars and meshes are still used for segment
reinforcement. In some countries, also steel fibers are used quite often. Put in global terms,
however, the use of conventional reinforcement still prevails.
As the production of the reinforcement cages requires a similar amount of spaces and
capacities as the segment production, it is often outsourced to a subcontractor. Should the
production of the reinforcement cages be planned on the jobsite, however, the following
areas should be provided:

• Steel building with crane equipment

• Area for the supply of raw materials
• Area for cutting, bending and rolling
• Optionally an area for pre-​fabrication of individual components
• Intermediate storage area
• Area of cage assembly with material storage
• Welding stations with exhaust system
• Cranes for taking off the cages
• Space and equipment for transport of the finished cages
• Cage storage yard

The rescue and fire protection concepts are similar to the segment factory. Correspondingly,
planning of gates, ventilation and lighting.
Special attention should be paid to power consumption, as a large number of welding
devices with a high input power are used.
The reinforcement cages have to be produced in specially manufactured gauges. Besides
the outer dimensions (length, width, thickness, radii, angles) the gauges must define the
areas without reinforcement and the recesses.

8.4.4 Concrete laboratory for quality control testing

Concrete used on jobsites is subject to in-​house as well as third-​party monitoring. Therefore,
it is necessary to provide a concrete lab with the corresponding equipment. Tasks of the con-
crete lab include the following:

• Tests with the fresh concrete properties

• Tests with the young and matured concrete
• Retained samples
• Measurement of temperatures
• Recording
472 Handbook of Precast Segmental Tunnel Lining Systems

8.4.5 Outside storage with gantry cranes

In the outside storage yard, the segments are piled and mature until having achieved the
required concrete compressive strength, which is defined by the planner/​designer. For many
projects, the common 28-​day compressive strength is obligatory. This rule can be justified
only to an extent, as in the case of a statistically distinct proof, an earlier use of the seg-
ment; i.e. with achievement of the compressive strength, can be permitted. Particularly with
regard to segment concrete, which is processed in a serial production with a constant recipe
throughout the complete production period and an intensive and regular monitoring of the
components, this could be a practicable possibility.
The outside storage yard also serves as a buffer between production and tunneling; i.e.
the required segments. Two factors mainly influence determination of the outside storage
yard’s size:

• The available area that can be used

• A progressive graph visualizing the production (accumulated) and the consumption
(accumulated)

However, the graph cannot consider any kind of interruptions, which can lead to a consid-
erable increase of the required storage area.
Therefore, segment production experts often work with a yard, which is capable to store
an output of three to four months of regular production. This approach proved to be applic-
able for a regular production.

• The output of the first month of production may not be used due to an insufficient
compressive strength
• The output of the second and third month of production can be used
• The output of the fourth month of production serves as a buffer for any kind of
interruptions

Further attention should be paid to special segments and rings for the following
considerations:

• Cross-​sections
• Rings with special inserts
• Rings with increased reinforcement
• At the beginning of the storage planning it is necessary to draw up a concept consider-
ing the complete way of the segments/​rings from production until installation of the
tunnel. This is what finally influences the sequence and arrangement of the piles. The
following should be considered:
• Arrangement of the molds in the segment factory
• Arrangement of the segments in the pre-​storage area
• Piling sequence for storage on the outside yard
• Piling sequence for release from stock
• Transport to the TBM
• Further intermediate storage facilities
• Segment crane on the TBM
• Segment feeder on the TBM
• Possible ring arrangements and the resulting sequence of ring building
 Formwork systems for precast segmental tunnel linings 473

Particularly for transport of the segments, attention has to be paid to avoid a mirror-​inverted
supply of the segments. For this purpose, strict guidelines have to be issued for segment
handling (truck, forklift, etc.) or must be provided in a structural manner.
It is usual to turn the segments in the segment factory already (INTRADOS side showing
towards the top). The following is required for piling:

• An adequate support on the floor

• Intermediate supports

The support on the floor in most of the cases consists of reinforced concrete elements with
a wooden, plastic or rubber support. It is important to have a space between the support
and the bottom side of the segment to enable the segment clamp to grab the segment or the
pile. When it comes to the points of support, experience showed that a division in thirds (in
vertical direction) turned-​out to be adequate. Finally, this has to be defined by the designer.
The intermediate supports generally consist of wood. For heavy segments, hard wood is
used. When defining the vertical support points, it is necessary to check the inner radii of
the segments with regard to an application of the piling wood without slipping or tilting.
The height of the piling woods is calculated on the basis of

• Inner and outer radii

• Safety space between the segments
• Transport clamps intended to be used

Further, position and quantity of the piling woods results from the compressive strength on
the one hand and the decision, whether a line load can be transmitted into the young seg-
ments, on the other hand. It often turns out that a three-​point-​support makes more sense as
the risk of restraints or torsion tends towards zero.
Another important aspect regarding piling is to avoid horizontal offsets. The pile could
possibly collapse as a consequence of unilateral load peaks when the load is transmitted in
the floor. Further, the structural clearance of the clamp is limited. This would quickly lead to
damages on the segment. Experience showed that an eccentricity of ±50 mm (2”) proved to

Figure 8.15 Segment transport with a mechanical clamp on a crane.

474 Handbook of Precast Segmental Tunnel Lining Systems

be an acceptable value for horizontal offset. However, this has to be calculated and defined
by the planner.
Certain minimum spaces between the piles are necessary for storage and release of the
segments from stock. On the one hand, as said before, the dimensions of the clamp as well
as its opening and closing have to be considered. On the other hand, storage yards are often
defined as accessible areas and are therefore subject to applicable health and safety regula-
tions. Local and national regulations have to be considered here.
The number of segments per pile results from the following:

• Achieved compressive strength of the concrete

• Available storage yard
• Tonnage capacity of the ground
• Lifting equipment (forklift or gantry crane)

Usually, at least one complete ring is piled per storage position. This makes identification
of the pile easier. Making use of a warehouse logistics system, higher piling is also possible
when the corresponding conditions are established and maintained. Meanwhile it is usual to
pile 1.5 or even two complete rings on one storage position, thus increasing storage capacity
considerably. Resulting from the overall concept and the crane capacity, in most cases pile
packages of three or four segments each are planned.
Special attention has to be paid to small key segments. They always have to be the top
segment on a pile or separate piles are made using key segments only.
For historical reasons, a storage per ring has been implemented for many projects. This
means that on each storage position one complete ring is piled, following a defined sequence.
However, this leads to a loss of storage capacity in many cases. Further, due to the many
handling activities on the way to the TBM, the sequence often cannot be kept. Therefore,
experience showed that a storage per type could offer advantages. This means that there are
only identical types of segments on one storage position. They are carried to the place of
installation (TBM) in this way and the piling sequence is configured directly there.
Due to the large number of piles it is recommended to use a data management system.
In this case, during production already, the segments are provided with an identification
number (segment internal diameter) and can thus be assigned to a special storage position.
The different options and possibilities for organization of the warehouse are described in
detail in Chapter 10.
On most of the outside storage yards, gantry cranes are used. When it comes to dimen-
sioning of the gantry cranes, it is necessary to consider the following:

• Capacity also includes the segment clamp

• Cranes with two lifting hoists (with individual control) are provided
• Cantilever is planned on one or both sides for loading of the trucks
• Sufficient lifting and travel speeds
• Clearance for the movement of the crane is guaranteed at any position
• Control of the crane is provided in a driver’s cab
• Adequate industry classification of the crane is chosen

The outside storage yard generally must provide a sufficient soil bearing capacity. The
relatively high loads from the piles must be transmitted safely to the ground. If necessary,
 Formwork systems for precast segmental tunnel linings 475

the soil must be improved. In this regard, a possibility for drainage of the surface water
(rain, ice, snow, etc.) is also important. For this reason, watertight surfaces (asphalt, con-
crete, etc.) should be allowed to drain.
In the case of an outside storage yard with gantry crane, the foundations for the
crane runway have to be dimensioned and built in a way that dangerous settlements are
avoided.
For the case only forklifts are intended for storage and release from stock, sufficient sur-
faces should be provided around the piles.
Generally, all roads for loading and unloading as well as all access roads of the storage
yard must be prepared for heavy cargo.

8.4.6 Utility systems (water, sewage, electricity, gas/​oil Wi-​Fi,

phone, etc.)
The segment plant has to be equipped with the standard utilities:

• Water: mainly in the area of the batching plant and cleaning of the molds. In this con-
nection, the treatment of concrete sewage should also be considered.
• Sewage: besides concrete sewage mainly in the area of the sanitary installations
• Electricity: at the beginning of the planning, an electricity concept should be drawn
up, evaluating the main consumers (batching plant, gantry cranes, overhead cranes)
with regard to their power consumption and simultaneous operation. The use of a
transformer station could possibly be necessary.
• Gas/​Oil: for operation of the heating system for the segments, the segment factory
and optionally for the concrete aggregates
• WiFi: for data management and production control in the factory, on the outside
storage yard up to installation in the tunnel
• Telephone: communication between batching plant and concreting operator, concrete
demand, etc.

8.4.7 Infrastructure
When planning the segment plant, an infrastructure concept has to be drawn up considering
any logistics and transport of goods, particularly, the following:

• Supply of aggregates, cement and reinforcement steel

• Transport of the segments to the place of installation
• Supply and erection of large components (gantry girders, etc.)

At the same time, an unlimited access to any utilities systems; e.g. ground wires/​air ducts,
must be provided for from the beginning.

8.4.8 Safety and rescue concept, emissions

A safety and rescue concept for the segment plant should be worked out and implemented
on the basis of legal provisions. Further, all components and areas must be checked for emis-
sions (noise, dust, etc.)
476 Handbook of Precast Segmental Tunnel Lining Systems

8.4.9 Redundancy
One of the most important topics when it comes to planning of a factory is redundancy. It
is necessary to check which components have an important influence on quality and safety
of production. In the case of failure of a compressed air generator, for example, production
can be continued with a mobile power pack in most of the cases. In contrast, however, when
the only overhead crane of a factory installation fails, the whole production must possibly
be stopped for days or weeks.
The following components are considered to be critical for the production:

• Batching plant
• Carousel system
• Cranes
• Energy supply
• Concrete supply systems

Often a second batching plant is not provided, stating that the plant provides a sufficiently
high output. When the plant is working smoothly this may be correct, however, based on
experience, the batching plant turned out to be the bottle neck. Considering a concrete
plant located far away to help out in case of a failure is not an option as the require-
ments to the fresh concrete are very high (temperature and consistency as well as plasticity
over time).
When it comes to the carousel plant, cleaning and maintenance are often underrated,
not considering that besides provision of critical spare-​parts, cleaning and maintenance are
decisive for a high availability of the plant.
Generally, two cranes are required in the production factory. The cranes should be speci-
fied to have the same or at least similar capacities in order to be able to replace each other
in the case of a failure. Although in this case it is not possible to achieve the same output, a
production running at 50% is even better compared to a complete standstill.
Particularly in the case of field factories, a continuous and enough power supply must
be provided. Each failure of the power grid results in a standstill of the plant. Further, safe
operation may be endangered in the case of a power failure when vacuum lifters transport
segments weighing several tons. The time remaining until failure of the vacuum and crush-
ing of the segment is restricted. Experience has shown that an independent power system
(diesel generator) makes sense for at least several areas. Also, power fluctuations appear
(even when the local supplier provides electricity) that may cause damages and failures of
the frequency converters and control units.
The same as to the carousel also applies to the different concrete supply systems (concrete
hopper, conveyor, truck mixer, etc.) cleaning and maintenance contribute to a high avail-
ability. Here, redundancies can also be created, for example, by providing transport mixers
or switches for the bucket conveyor systems.

8.5 CARROUSEL AND STATIONARY PRODUCTION

One of the most difficult issues is to define which type of production system is most appro-
priate for the project, a simple stationary production (Figure 8.16) which is quickly erected
or a sophisticated carousel production (Figure 8.17)? This question cannot always be
answered unambiguously and “the exception proves the rule”.
 Formwork systems for precast segmental tunnel linings 477

Figure 8.16 Stationary segment production with crane-​b ased concrete distribution system.

Figure 8.17 Carousel system with concreting chamber and curing tunnel.

Arguments for a stationary production:

• Short tunnels, short project period

• Quick assembly, early start of production
• Low costs of local workforce
• Little experience in segment production
• Little experience with tunnel lining segment concrete

Arguments of a carousel production:

• High costs for local workforce/​labor costs

• Slightly smaller space requirement
• Better control of the works carried out at the same place
• Prospect of follow-​up projects; i.e. a re-​use of the plant
• Vast experience with segment production and tunnel lining segment concrete
478 Handbook of Precast Segmental Tunnel Lining Systems

Table 8.1 Segment production facilities –​ alternatives.

Generalized operational comparison between modern carousel and stationary
segment production facilities for the same size precast concrete segmental tunnel
linings

Segment production facilities

Item Operation description Carousel system Stationary system
1 Cycle times 8 to 12 minutes 10 to 16 minutes
2 Mold operations per day 2 to 2.5 cycles 2 to 3 cycles

3 Required personnel 0.8 to 1.3 h/​m 3 1 to 1.6 h/​m 3

4 Number of molds 50 to 60 maximum per No limit No limit
carousel
5 Average output per day 100 to 140 segments No limit No limit
6 Readiness for operation 8 to 10 months 4 to 5 months
following order confirmation (approx.) (approx.)

A stationary production is said to be unable to achieve the same output as a carousel plant.
A worldwide comparison, however, has shown that the same; partly even higher, production
rates can be achieved with a stationary system. Therefore, this argument is not crucial.
Operation of a carousel plant is always subject to the cycle time of the plant/​personnel
and the curing time of the concrete as, in addition, only a fixed number of molds is avail-
able. In the case of a stationary production, however, the number of mold sets can be
increased at any time. Further, the requirements to the fresh concrete are not as high as the
mold does not have to be moved directly after concreting and as demolding of the segment
after curing can be postponed if necessary.
Another unjustified argument is that the quality of the segments is better when using a
carousel plant. It can be confirmed that monitoring of the quality; e.g. cleaning or installa-
tion of the reinforcement cages, is easier, however, this does not mean that quality is higher
in any case.
Maintenance of carousel systems may not be underrated. Choosing this type of produc-
tion, operational safety must be guaranteed at any time. If the carousel cannot move, there
is no output at all. When working with a stationary production, however, production can
be continued with restrictions in the case of a failure of the plant.
The following numbers have become established during the last years as listed in Table 8.1.
Costs for heating and consumption costs are slightly lower with carousel production. The
curing tunnel, which is part of the carousel plant enables the curing temperature to be kept
on a constant level.
Besides the short time required for assembly, flexibility is a further advantage of a sta-
tionary production. The system can be flexibly adapted to existing buildings using available
surfaces optimally.
Yet safety and ergonomic aspects have to be considered when planning a stationary pro-
duction. Special attention should be paid to travel paths for concrete transport and crane
travel with heavy load.
The concrete is often transported using flying hoppers assembled to the hall pillars. These
hoppers are able to transfer the fresh concrete, generally between 2 and 3 m³ (2.6 to 3.9 CY)
 Formwork systems for precast segmental tunnel linings 479

very quickly from the batching plant to the place of concreting (mold). The mold is then
filled with a bucket on a rope or a distribution system attached to a crane bridge.
Depending on the geometry of the building and local safety standards, the concrete can
also be filled into the mold directly from a mixer truck.
When implementing a carousel plant, concrete is supplied via a bucket conveyor system.
In this case, the batching plant is located close to the place of concreting.
Summarizing the above, it can be said that, having experience with segment production
and planning to realize future projects, a carousel can offer cost saving in a region with high
wages. The investment can be depreciated on several projects and a well-​instructed team can
achieve a high output on a high quality level.
A look at the recent years shows that stationary systems are implemented for approx. 60
to 70% of all projects, whereas only 30 to 40% are realized using carousel systems.

8.6 SEGMENT HANDLING EQUIPMENT

When working out the overall concept for the lining rings from their production up to
installation in the tunnel, it is necessary to define logistics of the individual segments and
segment stacks at an early stage.
The following process steps and stages of the segments should be considered:

• Demolding
• Transport in the hall
• Rotation/​turning
• Equipment
• Evacuation from the hall
• Transport to the outside storage yard
• Storage and retrieval of segments
• Transport to the tunnel /​launching shaft /​intermediate storage
• Transport to the TBM
• Transport on the TBM
• Installation in the tunnel

The following customized handling devices have been implemented most during the last
years (Figure 8.18):

8.6.1 Demolding
After having achieved a minimum compressive strength of the early age concrete, in general
a minimum compressive strength of 12 to 15 N/​mm² (1,740 to 2,175 psi) is required, the
segment can be demolded/​stripped.
It is common to use vacuum lifters with suction plates for demolding, enabling the seg-
ment to be lifted safely without damages. However, for the use of a vacuum plate the surface
has to be as even as possible without offsets.
For a safe transport in the hall, safety clamps can be installed which fix the segment after
demolding, thus avoiding a crush of the segment in case of a vacuum failure.
480 Handbook of Precast Segmental Tunnel Lining Systems

Figure 8.18 (a) One-​s egment clamp; (b) multi-​s egment clamp; (c) gasket gluing frames; (d) turning
clamp; (e) tilting tables; (f ) transport trolley; (g) extrados vacuum lifter; (h) intrados
vacuum lifter; (i) turning table.
 Formwork systems for precast segmental tunnel linings 481

8.6.2 Transport in the hall

In most cases, the segments are transported from the place of demolding to the intermediate
storage by crane either with the vacuum lifter or using mechanical clamps (see Figure 8.15).
When planning the transport routes, attention should be paid to avoiding suspended load
passing over personnel workstations.

8.6.3 Rotation/​turning
Before or after intermediate storage, the segments are turned. The intrados side showing to
the bottom (the side that is later visible in the tunnel) must be turned up to the top. There
are two possibilities for turning:

• Turning around the longitudinal axis of the segment

• Tilting over ring joint side

Turning around the longitudinal axis is common when working with partly or fully auto-
mated segment transport and equipment lines. Applying this system, the segment is clamped
on the side, lifted from the transport trolley and turned by 180°.
Tilting tables more and more prevail in practice in the last years. Initially laying on an
“L-​profile” intrados down, the segment is tilted with two movements of 90° each with this
method. This simple and fast system offers different ways of application.

8.6.4 Equipment
Unless using a cast-​in gasket, the EPDM gasket is optimally applied to the segment after
turning. For this purpose, a gluing agent is applied first, on which the gasket is placed
after a defined open time. In the next step, a gasket gluing frame presses the gasket
to the segment for a period defined by the gasket manufacturer. When applying the
gasket frame, attention must be paid to maintaining the gasket correctly in its designated
position.
Generally, plywood sheets are not applied to the segments until only after turning.
Additionally, many producers use the gasket gluing agent also for application of the ply-
wood sheets.

8.6.5 Transport out of the hall

For removal from the hall, three options have emerged:

• Transport by means of tractor and trailers

• Transport on an automotive trolley on rails
• Evacuation line with scissor lifters or chain conveyors

All systems are essential and should be selected depending on segment dimensions and
weights as well as local conditions (position of the outside storage yard, crane of forklift,
etc). In many cases, a combined transport and equipment line is chosen, which offers fixed
working positions similar to a carousel system.
482 Handbook of Precast Segmental Tunnel Lining Systems

Outside the hall, the segments can be loaded on the transport vehicles by means of gantry
cranes, semi-​gantry cranes and overhead cranes or similar. Here, a defined sequence of
stacking can be followed.

8.6.6 Transport to the outside storage yard

When transporting segments with a low weight or individual segments, or in the case of a
storage yard without gantry cranes, forklifts turned out to be an adequate option for trans-
port. There are different possibilities for the forklift to pick up the segments and transport
them to the storage place:

• Using the fork and support pads

• Using an intrados vacuum lifter
• Using a clamp

Fortified and clearly defined transport routes are necessary in this case for reasons of work-
ing safety, on the one hand, and in order to avoid damages and twisted storage of the seg-
ment, on the other hand.
When it comes to the transport of heavy segments or complete stacks, it is recommended
to use a system consisting of a tractor or container towing vehicle and trailers (heavy-​duty
pallets). Multi-​service vehicles (MSVs) can optionally be used.
In general, it is necessary to provide large and stable transport routes. Small turning radii
and too steep inclines must be avoided.

8.6.7 Storage and retrieval of segments

Provided that forklifts are used; they can also be applied for storage and retrieval. When
transporting the segments on a trailer and working with a gantry crane on the outside stor-
age yard (Figures 8.19 and 8.20), segment clamps are used in most cases. There are different
types of clamps on the market:

• Mechanical clamps for one or several segments

• Hydraulically or electrically operated clamps for one or several segments

Figure 8.19 Segment storage yards with gantry cranes –​ ­e xample 1.

 Formwork systems for precast segmental tunnel linings 483

Figure 8.20 Segment storage yards with gantry cranes –​ ­e xample 2.

Mechanical clamps dispose of a mechanical system starting when the clamp becomes in
contact with the segment and closing automatically when the segment is lifted. The same
operation for disengaging the segments. This simple type of clamp is used very often.
If segment stacks with different numbers of segments need to be lifted, a clamp with
hydraulic or electric operation is recommended in most cases. The clamp can be opened
and closed via remote control. Further, the support system of the clamp can be selected
depending on the number of segments to be lifted; e.g. instead of a stack of three segments,
a stack of two segments or an individual segment can be transported.
Besides the stacking sequence, handling of the key segment should be an object of further
investigation, as its smaller size requires special solutions.
A static proof is necessary for segment clamps in any case. The supports (contact of the
clamp with the concrete) have to be executed in a way to avoid impact of the forces in
the marginal area of the concrete. Instead, forces have to be transferred as far as possible at
the center. The support surface should generally be executed as a polyamide support.

8.6.8 General
For any handling device and any state of construction, the designer has to carry-​out static
investigations and calculations considering decisive loads working on the segment. In the
case of an unfavorable stacking or displaced stacking woods, for example, loads may occur
that are not considered in the standard static equilibrium of the segment or ring.

8.7 THREE-​D IMENSIONAL MEASUREMENT OF SEGMENT MOLDS AND

SEGMENTS

8.7.1 Instrumentation and analysis

Due to the complex geometry of tapered rings there are frequent designs that have no par-
allel surfaces from which to measure the dimensional accuracy of either the molds used
to produce the segments or the segments themselves. To completely inspect the individual
dimensional accuracies and particularly the torsion in a manual way, a very time-​consuming
measurement procedure using hand-​held templates with highly skilled personnel to cor-
rectly evaluate the measurement results would be required.
484 Handbook of Precast Segmental Tunnel Lining Systems

Figure 8.21 Semi-​a utomatic 3D Segment measurement.

The most common and economical method to perform 3D checks in bigger scales is to
use a laser tracker or laser scanner instrumentation. While laser scanner technology still
fails to meet the accuracy required in most tunneling specifications, contact measurements
using laser trackers deliver acceptable results and are meanwhile an industry-​wide standard
to proof the dimensional compliance of molds in mold manufacturing factories across the
globe. Figure 8.21 shows a semi-​automatic 3D segment measurement.
Such a system typically comprises a laser tracker to carry out a measurement of all the
contact surfaces, the radii, both the outer and the inner diameters as well as any detail such
as gasket grooves and bolt-​hole positions. These measurements are analyzed and processed
via a traceable 3D metrology software platform and comparison is then made between the
measured object and the nominal design. Any report that is made from the measurement of
the tolerances should be as succinct as possible but as comprehensive as necessary to impart
the necessary information.

8.7.2 Measurement regime
The total quality control should be clearly defined as the responsibility of the contractor
who should supervise the segments supplied by the precast company.
Key points to add according to International Tunnelling Association (ITA); ITAtech
Report No.9: Guideline for Good Practice of Fibre Reinforced Precast Segment–​Volume
2: Production Aspects (2018) include the following:

1. The reference temperature that the segments should achieve the specified tolerances.
2. The time after de-​molding when precision measurement of the segment is acceptable.
3. A clear measurement regime should be specified such as follows:
a. Dimensional measurement control of all molds at manufacturers’ factory including
segment detail such as bolt hole position, gasket grooves etc.
b. Check dimensional measurement control of all molds on the precast production line
(either static or carousel before mass production), this should include a check on any
torsion.
c. The measurement location; e.g. roofed hall with temperatures between 10° and 62°C
(50° and 80°F) should be specified in the construction contract.
 Formwork systems for precast segmental tunnel linings 485

d. Three-​dimensional measurement control of all segments after first pouring.

e. Then after a given number of castings, the exact frequency can be varied according
to the designer’s project specific requirements such as follows:
i. Dimensional measurement control of all segments after the 10th casting from
each mold.
ii. Dimensional measurement control of all segments after the 20th casting from
each mold.
iii. Dimensional measurement control of all segments after the 30th casting from
each mold.
iv. After the start of production, the 50th ring produced from every formwork mold
should be re-​measured. The next measurements should be performed after every
100 castings, unless otherwise provided for in the construction contract.
4. Tolerances on “closed” (built) ring must not be the sum of all individual tolerances.
5. Individual tolerances should be compensated with the mathematical sign.
6. Every controlled segment must be proved by a record sheet (physical and digital). A full
segment documentation system should enable full traceability of the segment produc-
tion; this should include all the constituent components of the segment, the production
methods and durations, personnel responsible, tolerance measurements right through to
the precise place the segment is installed in the finished tunnel.

8.8 MASTER RING ERECTION AND VIRTUAL RING BUILD

When a new tunneling project starts, it has been a vital tradition for decades to cast and
erect master rings horizontally on a flat surface to demonstrate the compliance with dimen-
sional tolerances. Some tenders ask for the erection of two or even three rings upon each
other or repeated master ring erections at intervals throughout the project.
Typical designer specifications define that two test rings shall be erected on a flat and level
base with the top ring rotated one bolt pitch from the bottom ring with the bolts inserted.
The base ring shall be retained as the master-​ring. Dimensional checks shall be made against
the master ring at intervals not exceeding 0.5% of the segment production according to the
tolerances in Table 8.2.

Table 8.2 Master ring dimension tolerances.

Summary of tolerances and allowable deviation dimensions from the master ring for a
built-​u p test ring assembly

Master ring dimensional controls

Deviation
Operation description for master
Item ring build Code Tolerance Metric Imperial
1 Internal diameter of ring
ID ±0.2% ID ±10.0 mm ±0.39”
2 External diameter of ring
ED ±10.0 mm ±0.39”
3 Lip between adjacent segments LI ≤3.0 mm ≤0.12”
on internal diameter
4 Gap between longitudinal G 1.0 mm feeler 0.04”
segment joints gauge not passing
486 Handbook of Precast Segmental Tunnel Lining Systems

Figure 8.22 Example of physical mastering build.

Figure 8.23 Digital twin of the physical ring.

Particularly for large diameters, building trial rings is a challenging and risky task for
the workforce in the segment factory. In some cases, it can take up to a week to complete
a master ring and a truck-​mounted crane is necessary (Figure 8.22). The concept of virtual
rings (Figure 8.23) may have the potential to replace this tradition by assembling master
rings digitally instead of physically based on a best-​fit approach considering the 3D coordi-
nates from laser tracker measurements of segments or segment molds.
Building a virtual ring, a sub-​millimeter accuracy can be achieved, the alignment of neigh-
boring circumferential and ring joints observed, and the fit of bolt holes checked. The mean-
ing of the traditional method, on the other hand, despite the significant amount of work and
risks, is considered controversial in the industry as the horizontal position of the ring and
absence of gaskets do not simulate the reality in the excavated tunnel.
To demonstrate the manufacturing capabilities of a segment supplier including concrete
quality and surface characteristics, a master ring erection is mandatory for both parties the
contractor as well as the owner. Subsequent checks against the master ring on the other
hand should be performed by building virtual rings based on the geometrical results of 3D
measurements of segments or segment molds.
 Formwork systems for precast segmental tunnel linings 487

8.9 TOLERANCES OF SEGMENT MOLDS AND SEGMENTS

Tapered rings enable the tunnel lining to accommodate alignments that contain curves or
to compensate for misalignment of the tunnel due to site-​specific problems. This tapered
shape of the reinforced precast segments necessitates a high degree of geometrical precision
in manufacture and greater care in their installation is needed. The reasoning for specifying
the very tight tolerances in these standards are as below:

• Geometric sensitivity to inaccuracies and distortions of individual segments

• High load effects from earth, water and grouting pressure on the tunnel lining
• Jacking forces during mining
• The load transfer takes place only in limited areas (partial surface load)
• Damage cannot always be detected (for example, on the outer side of the ring)
• Repairing of damaged segments are costly and time-​consuming

Most projects include a requirement in the tender documents that specifies compliance with
the tolerances indicated in one of the several international standards, most of which were
issued prior to the use of fiber reinforcement or the availability of the current measurement
technologies. Many of these international standards already call for sub-​millimeter accuracy
on several key dimensions. Ring designers will frequently modify these values according to
project specific requirements.
All these standards are however incomplete with respect to the requirements of current
precision measurement techniques and fail to include full traceability of the segment, its
constituent components, manufacturing process, dimensional accuracy and installed loca-
tion as part of an ongoing quality assurance/​quality control process. On some projects there
is only a passing reference to the tolerance and quality control requirements, which enable
the contractor or precast company to bypass the checking that is implied in the standards/​
guidelines.
The simple addition of a few clarifying clauses in the tender documents are called for to
significantly improve this. The inclusion of all these requirements in a table or diagrammat-
ically on a single composite diagram will dramatically improve the understanding of the
requirements and assist in their implementation. Please refer to Table 8.3 below.
All tolerances mentioned in Table 8.3 should be subject to the following set of rules:

• It is not allowed that the maximum tolerances add to one another, hence tolerances of
the whole ring are smaller than the sum of the single tolerances.
• All given tolerances refer to a reference temperature of, e.g. 21°C (70°F).

In the case of inconsistencies of two dependent tolerances, the more stringent shall be rele-
vant. Furthermore, tolerances for gasket grooves (TG, WG) shall be in accordance with the
requirements of the gasket supplier.

8.10 SEGMENT TRACKING AND QUALITY DOCUMENTATION SYSTEM

Planners and owners usually design tunnel buildings for life spans of 100 years or more and
it is mainly the precast segments of which a tunnel tube is constructed. Taking this fact into
consideration, the definition of a comprehensive quality control plan and its continuous
monitoring seems inevitable.
488 Handbook of Precast Segmental Tunnel Lining Systems

Table 8.3 Segment and formwork dimensional tolerances.

Summary of segment and formwork dimensional tolerances based on published codes
and industry guidelines
Precast segment and formwork tolerances
(According to BTS Specification for Tunnelling–​3 rd
Edition, Section 24.4.1)
Segment tolerances Formwork tolerances
Segmental lining Item
Item tolerance of fabrications code Metric Imperial Metric Imperial
1 Width of segment W ±1.0 mm ±0.04” ±0.3 mm ±0.012”
2 Segment thickness (on T ±1.5 mm ±0.08” ±2.0 mm ±0.08”
backs)
3 Internal radius of RI ±1.5 mm ±0.06”
segment
4 Radius (position) of RO ±1.0 mm ±0.04”
bolt hole axis, joint
face (contact axis) and
gasket groove axis
5 Evenness ER ±0.3 mm ±0.012” ±0.3 mm ±0.012”
(flatness) of radial (0.1 mm) (0.004”)
(longitudinal) joint: ②
contact zone
6 Evenness (flatness) of EC ±0.5 mm ±0.02” ±0.3 mm ±0.012”
circumferential joint ①③
contact zone
7 External segment arch AE ±1.0 mm ±0.04” ±1.0 mm ±0.04”
lengths: types A1, A2,
A3, A4, A5 An, B, C.
8 External segment arch AE ±1.0 mm ±0.04” ±0.5 mm ±0.02”
length: type K
9 Pitch angle segment: αP ±0.2° ±0.2°
types A1, A2, A3, An,
B, C
10 Pitch angle segment: αP ±0.1° ±0.1°
type K
11 Cross setting angle in βR ±0.1° ±0.1°
radial joint –​ angle of
joint plane
contact zone
12 Cross setting angle in ΒC ±0.1° ±0.1°
circumferential joint –​
angle of joint plane
contact zone
13 Angle of Radial Joint γ ±0.01° ±0.01°
taper
14 Position of bolt ±1.0 mm 0.04”
hole and Plastic dowels
 Formwork systems for precast segmental tunnel linings 489

Table 8.3 (Continued)

Precast segment and formwork tolerances
(According to BTS Specification for Tunnelling–​3 rd
Edition, Section 24.4.1)
Segment tolerances Formwork tolerances
Segmental lining Item
Item tolerance of fabrications code Metric Imperial Metric Imperial
15 Depth of gasket sealing TG ±0.5 mm ±0.02” –​0 .2/​+​ –​0 .008 /​ +​
groove 0.3 mm 0.012”
16 Width of gasket sealing WG ±0.5 mm ±0.02” ±0.2 mm ±0.008”
groove
17 Smoothness of faces ±1.5 mm ±0.06”
other than joints (smooth float) (smooth
±1.0 mm float) ±0.04”
(formed face) (formed face)
18 Mismatch of sealing ≤0.5 mm ≤0.02”
groove at corners
Notes:
① 2.0 mm (0.08”) feeler gauge not passing beneath a 1.00 m (39.37”) long straight edge
② From the theoretical plane, with a rate of deviation not exceeding 0.6 mm/​m (0.0072”/​foot)
• 0.3 mm (0.012”) generally
• 0.1 mm (0.004”) in radial plane
③ From theoretical plane, with rate of deviation not exceeding 1.0 mm/​m (0.012”/​ft)

Traditional quality control methods include pen and paper, or Microsoft Excel based
solutions to document the fabrication of segmental linings in the precast factory. With the
introduction of BIM models (building information modeling) and ongoing digitalization,
the need for reliable and complete data increased significantly to be able to trace from an
individual segment installed in the tunnel back to the raw materials used to produce it.
Traditional methods often fail to fulfil these demands.
Besides the documentation aspect, modern tracking systems support production man-
agers in increasing productivity of their precast factories and connect them to the tunnel
jobsite thus creating comprehensive reporting and dashboards about the logistic and instal-
lation procedures.

8.10.1 Segment labeling
There are various options to label (tag) precast segments and give them a unique identifier
(Figure 8.24). In addition to the limited information (e.g. mold set number and type) pro­
vided by the marking plate typically mounted on the mold’s circumferential joint, the most
common way to tag a segment is using two barcode stickers; one at the leading edge and one
intrados face. For the purpose of long-​term identification in the tunnel or automatization
steps, synthetic marking plates with or without RFID tags recently became more popular
in the industry.
Regardless of which tagging is applied, the unique serial number always links to a
database that saves the complete history of the segment. Using handheld devices such as
smartphones or barcode scanners, new information can be added to this database and the
segment’s vitae can be reviewed.
490 Handbook of Precast Segmental Tunnel Lining Systems

Figure 8.24 Segment barcode labels.

8.10.2 Digital documentation and process monitoring

Tunnel precast segments are produced under a strict quality control regime. For each unit,
various quality documents as in steel certifications, batch tickets, laboratory tests, check
lists, concrete curing temperature records, inspection releases, repair histories and time
stamp protocols for each individual handling step must be created and made available to
different stakeholders.
State-​of-​the-​art tracking systems allow to acquire and maintain all necessary data in a digital
database thus making the segment factory paper-​free. To avoid human errors during the docu-
mentation and ensure that the data is complete and reliable, some suppliers offer customized
software solutions according to project-​specific quality plans that are capable to control the
completion of the documentation of individual production steps only then allowing the oper-
ator to continue with the next work step. Apart from the added effort to customize the software,
this process monitoring approach enables the factory workers to self-​check, which significantly
enhances the quality of the documentation and reduces the need of human supervision.
After a precast segment is produced and released for transportation to the tunnel jobsite,
a comprehensive quality management system also tracks the logistics part. To record the
whole life cycle, the transport, onsite storage and even installation position of each indi-
vidual segment in the tunnel should be recorded. Linking the segments to a specific ring
number is a key aspect to be able to transfer and upload the lifetime quality records (LQR)
into a 3D BIM model.

8.10.3 Production and logistics management

As a side effect of the digital documentation, various visualizations, dashboards and report-
ing can be generated without remarkable effort. The production manager’s work can be
simplified by automatic daily and weekly production reports. Decisions and improvements
in the production workflow are based on clear and transparent performance figures when
working with a digital system.
The graphs in Figures 8.25a and 8.25b show the production output of a segment factory
as well as a shift comparison between the day shift (green) and night shift (blue). In average,
the blue shift’s output is three units higher compared to the green shift. A more detailed ana-
lysis of the data would uncover that the cycle time of the carousel system during the blue
shift is lower due to fewer concrete vibrations in the concreting chamber.
 Formwork systems for precast segmental tunnel linings 491

Figure 8.25(a) and 8.25(b) Dashboard of segment production performance and shift comparison.

Besides production management, particularly in segment logistics with complex ring

designs, a tracking system can also support the warehouse management by providing stor-
age visualizations. Facing yards with several thousand segments of different geometry,
reinforcement structure, concrete mix design and maturing age that all look equal, such a
mapping solution avoids extensive search hours and ensures an optimized use of storage
capacities.
Furthermore, the easy and live availability of stock inventory and deliveries improve
the understanding on TBM ring consumption against production of both the jobsite and
the segment factory. Eventually, this information supports efficient production planning,
improves a balanced workload throughout the year and avoids ring shortages on the TBM.

8.11 SEGMENT MOLD QUALITY CONTROL

The molds for production of high-​precision segments are subject to strong mechanical and
thermal influences. The details are as follows:

• Dead weight of the concrete

• Hydrostatic pressure during concreting
• Different load cases during filling of the mold
• Single frequencies and overlaid frequencies during the vibration process (compacting
of the concrete)
• Extension of the concrete during curing (heat of hydration)
• Heating of the mold in the case of accelerated curing
• Opening and closing of the mold with compressed air operated impact wrench
• Closing of the screws with torque wrench
• Cleaning with abrasive tools

To be able to produce high-​quality segments with dimensional stability, the molds regularly
have to be checked. For this purpose, the mold manufacturer issues an operating manual
including recommendations regarding type and interval of the inspections. Please refer to
Table 8.4 for summary checklist of inspection and preparatory tasks for segment casting
operations.
492 Handbook of Precast Segmental Tunnel Lining Systems

Table 8.4 Summary checklist of inspection and operations tasks.

Detailed summary of preparatory tasks and routine inspection activities of precasting
operations using high precision steel forms for segmental tunnel linings

Operations stage and task

frequency

1x time per month

1x times per week

Decommissioning
Prior to each ese
Commissioning

1 to 2× yearly
Item Activity description operations and verifications
1 Cleaning –​ with rag and spatula W W W
2 Mold release agent –​ application W W W
3 Bolt hole formers grease –​ check gasket (O-​r ing) W W W
4 Bolt hole formers –​ grease threads (if applicable) W W W
5 Spindles –​ grease W W W
6 Adjusting screw –​ check for firm fitting W W W
7 Centering screws –​ cones and grease W W W
8 Sealing/​g askets –​ clean, check, exchange if necessary W W M M
9 Vibrators –​ visual and auditory check W W W M
10 Bolts for vibrator mountings –​ checking fixations M M W
11 Bolts, hinges, joints. axles and pivots –​ M M M
grease visual check of welded seams
12 Limit stops, locking and safety pins –​clean and check M M/​E W
13 Rubber block supports –​ clean and check M M W
14 Tension springs –​ check for damage M M M
15 Welds and welded joints –​ check M M W
16 General check of conditions –​ check M/​E M/​E M/​E
17 Vibrators –​ check direction of rotation M M
18 Compressed air feed –​ check M M
19 Compressed air distribution –​ check M M
20 Anti-​r ust agent –​ apply M W
21 Wheels –​ check M M M
22 Bearings –​ check M M W
Table Legend:
W Worker operator
E Electrician
M Mechanic, Foreman
Special Notes:
1
During the first use of the molds after assembly has been completed, there is an acute danger that
threaded connections can work themselves loose.
2
After commissioning, all threaded connections must be maintained and are the responsibility of the cus-
tomer and/​o r operator both in term of maintenance and costs.
3
Consider additionally, maintenance instructions in suppliers’ documentation package.
 Formwork systems for precast segmental tunnel linings 493

In addition to the above-​mentioned inspections, it is recommended to check the dimensional

stability of the molds in regular intervals.
Typical measurements and inspections during daily production include the following:

• Measuring of the mold width with a width gauge or a micrometer

• Checking of the markings at the joints of the side shutters
• Depth measurement with a caliper gauge
• Checking of the diagonal line with a measuring tape
• Checking for torsion

In case these measures show deviations from the nominal measures given by the mold manu-
facturer, it is recommended to measure the mold in detail using a 3D laser tracker.

8.12 INNOVATION IN TUNNEL SEGMENT PRODUCTION

Almost 200 years ago, courageous pioneers realized the first shield-​supported tunneling
works in Great Britain for construction of the Thames Tunnel in London. With this method,
the tunnel walls have generally been supported with steel beams and wood (ribs and boards)
and have been lined with bricks afterwards.
In the beginning, the tunnel cross-​sections had mostly been rectangular. With further
technical development and mastering of technologies, more and more tunnels had been
excavated with round cross-​sections. They were generally lined with bolted, cast-​iron seg-
ments. Economic reasons led to a replacement of the cast-​iron segment by the reinforced
concrete segments used today.
Reinforced concrete segments are state-​of-​the-​art also today, however, they have been
continuously improved. Recent developments include the following:

• Fiber reinforced concrete (commonly steel fibers)

• Concrete with fire-​protection properties (mainly polypropylene fibers)
• Cast-​in EPDM gaskets with improved anchoring in the concrete
• Dowels in the ring-​to-​ring joint that could accept both shear forces and tensile forces
(replacement of the usual bolting system)

In the field of segment production, innovations and improvements were brought to market
maturity and count among the state of the art in the meantime:

• Production of lining segments in modern, temporary erected on-​site factories

• Use of carousel systems and intelligent concrete distribution systems
• Flexible plants and systems (different type of precast products and diameters)
• High availability of the production plants
• Partly automated modules and components
• Data management/​traceability/​documentation (BIM)

Recent requirements particularly focus on the reduction of labor –​due to the lack of getting
labor –​the costs for plants, wages (labor costs) and operation as well as increasing safety
for the workers. The following innovations are currently investigated and implemented on
the market:

• Automated segment production (“Segment Factory 4.0”)

• Hydraulic opening and closing of molds
494 Handbook of Precast Segmental Tunnel Lining Systems

• Tunnels with permanent concrete protection using HDPE liners (Combisegments®)

• Use of existing data for production and logistics management. Please also refer to
Section 8.10.

8.12.1 Automated segment production (“Segment Factory 4.0”)

In 2019, the first robot supported production plant has been commissioned as a pilot and
development project in Switzerland. The task included automation of the following work
steps on an existing carousel plant aiming to significantly reduce the cost-​intensive personnel:

• Opening of the mold

• Cleaning of the mold
• Application of release agent
• Closing of the mold
• Inserting injection sockets

For this purpose, a bend-​proof steel portal with an overhead traveling crane has been
installed on top of the working line (to be compared with an assembly line in the automotive
industry). On this crane, a six-​axle industrial robot is installed headfirst. The robot and the
overhead traveling crane position themselves according to X-​Y-​Z coordinates and provide
a repetitive accuracy of less than half of a millimeter. The robot has been equipped with a
quick coupling system for the use of different tools. The tools travel together with the robot
and the crane. Figure 8.26 shows an example of a robot-​supported working line.

Opening and closing of the mold

For opening and closing of the mold cover and side walls, a full hydraulically operated
system has been developed. Once the segment mold arrives on a carousel system at a cer-
tain position, a quick coupling system connects the mold with the hydraulic aggregate
(Figures 8.27 and 8.28). Based on a multi-​level pressure system –​regulated by different
types of valves –​all mold operations start automatically.
First the cover lids will be opened, followed by the side walls and end shutters. This
process takes less than 1 minute and no worker impact is needed. Also maximum safety
for the worker is granted. As well, noise impact and a lack of safety –​due to loud impact
wrenches –​is eliminated.
On a carousel system this process can run on the cross transfer table. Means once the
mold arrives at the working line position, all covers and side walls are fully open and the
segment is ready for demolding.
In a stationary (or static) setup, the hydraulically operated molds should be fenced in a
safety cell.
The start of the opening process can be commenced by a timer (for example, all molds
will be opened in the morning before the first shift arrives) or manually.
Closing of the mold can run at different positions, wherever needed.

Cleaning of the mold

After opening of the covers/​side shutters and demolding the segment, the mold moves to the
next position. For coarse cleaning, a rotating brush is used. The rotating brush is mounted
on a six-​axis-​industrial robot and cleans the mold in a pre-​programmed mode.
 Formwork systems for precast segmental tunnel linings 495

Figure 8.26 Robot-​s upported working line.

The pre-​programming is necessary because after the completion of the cleaning by

brush, the robot changes the tool and blows off the remaining concrete and dust from
the mold with compressed air nozzles. Figures 8.29 to 8.31 show mold cleaning and oil-
ing tools.

Application of release agent

After blowing off the remaining concrete, a release agent is applied exactly where it is
necessary.

Inserting injection sockets

A three-​finger gripper tool (see Figure 8.32) finally inserts the injection sockets. The robot
picks up the sockets from a material box and places them exactly into the openings in the
mold. After this procedure, the mold drives into the concreting chamber to be filled for the
next production cycle.
Without use of the robot, six to eight persons were necessary to complete the tasks on the
working line. This number could be reduced to two workers, including the crane operator
with the robot.
496 Handbook of Precast Segmental Tunnel Lining Systems

Figure 8.27 Quick hydraulic coupling.

Figure 8.28 Mold with hydraulic installations.

Figure 8.29 Mold cleaning & oiling with a robot.

 Formwork systems for precast segmental tunnel linings 497

Figure 8.30 Rotating cleaning brush.

Figure 8.31 Tool for oiling.

Figure 8.32 Robot with three-​f inger-​g ripper.

498 Handbook of Precast Segmental Tunnel Lining Systems

Safety installations
The complete working zone of the robot is protected with a fence. The access doors are
integrated in the control system. Within the fenced area, separate cells are defined, which
are protected by means of light barriers; i.e. persons working within the fenced area cannot
enter the zone where the robot is working in that moment.

8.12.2 Tunnels with permanent concrete protection using HDPE

liners (Combisegments®)
Many cities all over the world modernize their sewage systems and have a large demand for
interceptor sewers collecting the inflows of the individual sewage systems and transporting
them on a long distance up to the sewage treatment plant. The interceptor sewers in general
have a diameter of 3.0 to 5.0 m (9’-​10” to 16’-​5”) and a length of more than 5 km (3.0
miles).
Old, mostly stonewalled interceptor sewers often show damages in the top area occurring
due to sulphurous gases. The gases generally emerge in the inflow areas as a consequence
of turbulence. To avoid this, the interceptor sewers have so far been lined with two-​shell
systems consisting of

• Outer shell, generally a segmental lining ring

• Inner shell with HDPE, generally in situ concrete cast

This two-​shell construction method, on the one hand, requires a large tunnel cross section
and a long construction period, on the other hand, as the inner shell cannot be installed
before completion of the outer shell. Further, all joints of the sealing lining require a gas-​
tight welding.
When implementing a one-​pass lining system with so-​called Combisegments®, the tun-
nel cross section (excavation) can be reduced, the time-​consuming and risky welding is no
longer necessary, thus resulting in a significant reduction of the construction time.
The Combisegments® consist of well-​proven products used in the tunneling industry
for years:

• HDPE lining panels with a thickness of 2.5 to 5 mm (0.10 to 0.20”)

• Compressible EPDM gasket with anchoring
• Polydicyclopentadiene (pdcpd) resin for connection of the individual components
• Injection sockets

In the production factory, the HDPE panel and the EPDM gasket are put in a large alu-
minum injection mold and over-​molded with pdcpd. The product, a strong panel, can easily
be inserted in the segment mold.
The panel is concreted with the segment and shows a very good compound with the
concrete and a high pull-​out force after curing. The segments with Combisegments®
panels are stored, transported to the tunnel and installed in the same way as standard
segments.
By compressing the EPDM gasket, on one hand, water extrusion to the outside is avoided
and, on the other hand, it offers a protection of the joint sides towards the sulpurous gases.
The inner surface of the segment itself is protected against corrosion caused by the gases by
means of the HDPE.
 Formwork systems for precast segmental tunnel linings 499

Figure 8.33(a) and 8.33(b) Tunnel lined with Combisegments® and mold with Combisegments®
inliner.

The interceptor sewers are built quite close to the surface in general. Therefore, the
outer water pressures caused by ground water can mostly be neglected (<1 bar) (14.5
psi). However, in case of higher water pressures from outside, the Combisegments® can be
drained with the help of special drain valves. This means that the external water pressing
from outside is led into the tunnel for relief.
An alternative possibility for application of the Combisegments® panels is protection
of the extrados side, i.e. the outer side of the segment. In the case of a high impact of
chloride, for example, a permanent concrete protection could be achieved. See Figure 8.33
for examples of CombiSegments usage)

8.13 CASE HISTORIES OF SEGMENT PRODUCTION

As described in the previous chapters, there are quite different concepts and implementa-
tions of segment production systems adapted to the special conditions of each project. To
be able to get an overview on the practical application of the different concepts and systems
500 Handbook of Precast Segmental Tunnel Lining Systems

presented before, some projects for recent segment production are shown in the following
with facts and figures.

8.13.1 Combined Railroad Line and Express Highway Adler–​Krasnaja

Poljana, Sochi, Russia
In 2008, segment production for lining of two tunnels, 9.400 m (30,840 feet) and ID 11.830
m (38,810 feet) internal diameter were tendered with high requirements: simultaneous seg-
ment production for up to six mechanized tunneling jobsites. The area available for seg-
ment production was located very close to the tunnel construction site on an 180,000 m²
(45 acre) site (including outside storage yard, see Figure 8.37) in Vesyolye (Adler District,
Sochi).
For production of up to 350 segments per day, the project planners decided to work
with three carousel systems with 96 molds in total. Each of the carousel plants included the
following:

• Concreting station with concrete hopper, gates, automatic mold cover and con-
trol system
• Reinforcement supply system with clamp, monorail with lifting hoist, reinforcement
cage feeder
• Curing tunnel with heating system
• Two vacuum lifters with (Figure 8.35) a capacity of 16 mt (18 tons)
• Evacuation line including deposit blocks, a tilting table 2 × 90° with a capacity of 16
mt (18 tons), a gasket gluing frame
• Two mechanical clamps with a capacity of 16 mt (18 tons)

Further, the overall scope included the below:

• Two carousel systems (Figure 8.34) for production of reinforcement cages

• Three batching plants
• Six gantry cranes 2 × 12.5 mt (14 tons)
• Two loading cranes 2 × 12.5 mt (14 tons)

See Figure 8.36 for an example of an integrated turning system.

Figure 8.34 Carousel system with curing tunnel Sochi.

 Formwork systems for precast segmental tunnel linings 501

Figure 8.35 Demolding with a vacuum lifter.

Figure 8.36 Integrated turning system.

The first carousel production line started operation in October 2009, running at full pro-
duction on the 20,000 m² (4.95 acre) site in March 2010. The efficient carousel production
kept the segments, some weighing up to eleven tons, in continuous motion 24/​7 to ensure
just-​in-​time segment production for tunneling. The segment production plant in Sochi was
the world’s largest of its kind at that time.

8.13.2 Riyadh Metro Line 1, Saudi Arabia

Over an 18-​month period in 2014 to 2015, the charged Joint Venture produced approx.
72,000 concrete segments in total for Riyadh Metro Line 1; i.e. for a 16.2 km (9.9 mile)
long mechanized tunneling section.
Using 160 molds (20 sets existing of 8 molds each) for stationary production, the Joint
Venture produced up to 35 rings (280 segments) per day and up to 839 rings (6,712 seg-
ments) per months. A maximum of 2300 rings (18,400 segments) could be stored in the
yard (see Figure 8.39).
To reduce interfaces in planning, design, delivery and installation and to have planning
certainty, the Joint Venture ordered a turnkey segment production plant from one supplier
with the following scope:
502 Handbook of Precast Segmental Tunnel Lining Systems

Figure 8.37 Segment storage yard Sochi.

Figure 8.38 Segment factory with access road.

Figure 8.39 Segment storage Riyadh.

 Formwork systems for precast segmental tunnel linings 503

Figure 8.40 Segment handling Riyadh.

• 20 sets of segment molds for stationary production (concreting with truck mixer)
• Segment handling equipment (Figure 8.40)
• Design of the complete segment production plant
• Factory set-​up including site preparation, steel building (Figure 8.38)
• MEP works and building concrete works
• Overhead cranes, gantry cranes, rails and transport trolleys
• Compressed air station
• Segment dimensional system (SDS): modular segment production and logistics man-
agement system

Due to the careful planning, the segment factory could be erected in the desert sand in
10 months only.

8.13.3 Albvorlandtunnel, Germany
For lining of the of the approx. 8 km (4.9 mile) long railway tunnel in Germany, the pro-
duction company installed a stationary production plant with (Figure 8.41) 77 molds and
an elaborate concrete transport and supply system in an existing building close to the
tunnel portal. The concrete logistics system included a flying hopper running on a gantry
crane along the longitudinal axis of the hall and a concreting bucket movable on a crane
(Figure 8.42) along the transverse axis of the hall to fill the molds, which were equipped
with compressed-​air vibrators.
For a perfect curing result, the mold supplier installed a combined heating system including
direct heating under the molds (finned pipes) with individual control to balance temperature
fluctuations as well as an overall heating system for the hall.
After demolding with a vacuum lifter (with safety clamp), the segments were placed in
the pre-​storage area. After turning of the segments from intrados to extrados by means of a
special turning device the gasket was applied with the help of a special gasket gluing frame
(Figure 8.43). Following this, a mechanical segment clamp put the segments on a trailer
504 Handbook of Precast Segmental Tunnel Lining Systems

Figure 8.41 Stationary production plant for Albvorlandtunnel.

Figure 8.42 Crane-​b ased concrete supply system.

to be transferred to the close tunnel portal, where the segments where loaded on a multi-​
service vehicle to enter the tunnel. See Figure 8.44 for transport of segments.
The segment factory for the Albvorlandtunnel had been built, planned and equipped in
close cooperation between the construction company and the manufacturer of the segment
production equipment. The plant was ready for production after six months of planning,
manufacture and installation only.

8.13.4 Koralmtunnel KAT 2, Austria

Between August 2012 and June 2016, the Joint Venture KAT 2 produced more than 120,000
concrete elements for the 32.8 km (20 mile) long Koralm Tunnel, which is part of the 130
km (79 mile) long high-​speed line from Graz to Klagenfurt in the South of Austria. The
segments for the two tunneling machines in use had to be produced in a very limited space.
For this reason, about 3,000 segments had to be ready in the stock permanently to ensure
continuous tunneling; this represents 1.2 times the average monthly output. See Figure 8.45
for the segment factory. The Joint Venture produced up to 238 precast parts per day. This
 Formwork systems for precast segmental tunnel linings 505

Figure 8.43 Gasket gluing frame.

Figure 8.44 Transport of the segments from the factory to the tunnel.

Figure 8.45 Segment factory Koralmtunnel KAT 2.

506 Handbook of Precast Segmental Tunnel Lining Systems

Figure 8.46 Carousel plant for segment production.

Figure 8.47 Automated segment evacuation line.

includes the segments of 8.2 mt (9.1 tons) as well as the invert segments of 13.4 mt (14.8
tons), which are produced on two carousel systems with 112 molds in total. Despite the
high occupancy of more than 1,300 production cycles, the molds kept to the tolerances of
up to 0.5 mm 0.012”). The tolerances were proven by means of 3D measurement of the
segments using a laser tracker.
The segment production with two carousel plants with 8 sets (48 pieces) of molds each
allowed consistently high production rates. Furthermore, the modular segment production
and logistics management system, Segment dimensional system (SDS), provided a reliable
basis for quality management on the jobsite. With this system, each segment can be moni-
tored from the pouring of the concrete to its installation in the tunnel.
After demolding with a vacuum lifter and pre-​storage in the hall in intrados position, the
segments were turned, fitted with the gasket and transported to the outdoor storage area
by means of an automatic evacuation line with scissor lifts running on rails to transfer the
segments between the different positions. See Figures 8.46 and 8.47 for the carousel plant
and automated segment evacuation lines.
 Formwork systems for precast segmental tunnel linings 507

BIBLIOGRAPHY AND READING LISTS

PUBLISHED CODES AND STANDARDS

AMERICAN CONCRETE INSTITUTE

ACI Committee 533. (2020). ACI 533.5R-​ 20: Guide for Precast Concrete Tunnel Segments.
Farmington Hills, MI: American Concrete Institute.
ACI Committee 544. (2016). ACI 544.7R-​16: Report on Design and Construction of Fiber-​Reinforced
Precast Concrete Tunnel Segments. Farmington Hills, MI: American Concrete Institute.
ACI Committee 544. (2017). ACI 544.2R-​17: Report on the Measurement of Fresh State Properties
and Fiber Dispersion of Fiber-​Reinforced Concrete. Farmington Hills, MI: American Concrete
Institute.
ACI Committee 544. (2017). ACI 544.9R-​17: Report on Measuring Mechanical Properties of Hardened
Fiber-​Reinforced Concrete. Farmington Hills, MI 48331: American Concrete Institute.
ACI Committee 544. (2018). ACI 544.4R-​18: Guide to Design with Fiber-​Reinforced Concrete.
Farmington Hills, MI.: American Concrete Institute.

AMERICAN SOCIETY OF CIVIL ENGINEERS (ASCE)

American Society of Civil Engineers (ASCE), Guidelines for Tunnel Lining Design, Prepared by the
Technical Committee on Tunnel Lining Design of the Underground Technology Research
Council, Edited by T. D. O’Rourke. ISBN: 0-​87262-​402-​1

ASSOCIATION FRANCAIS DES TUNNELS ET DE L’ESPACE SOUTERRAIN (AFTES)

AFTES Guide Specifications, The Design, Sizing and Construction of Precast Concrete Segments
Installed at the Rear of a Tunnel Boring Machine, No.GT32.R2A1, Association Francais des
Tunnels et de L’espace Souterrain (AFTES) (1999).

AMERICAN SOCIETY FOR TESTING AND MATERIALS

ASTM C09.42 Subcomittee on Fiber-​Reinforced Concrete. (2020). ASTM C1609 /​C1609M–​19a–​

Standard Test Method for Flexural Performance of Fiber-​Reinforced Concrete (Using Beam
With Third-​Point Loading). West Conshohocken, PA,: ASTM International.
ASTM A820 /​A820M–​16–​Standard Specification for Steel Fibers for Fiber-​Reinforced Concrete: ASTM
International.
ASTM C1609 /​C1609M–​19a–​Standard Test Method for Flexural Performance of Fiber-​Reinforced
Concrete (Using Beam With Third-​Point Loading). ASTM International.

BRITISH STANDARDS INSTITUTE

British Standards Institute. (2008). BS EN 14651:2005–​Test Method for Metallic Fibre Concrete–​
Measuring the Flexural Tensile Strength (limit of proportionality (LOP), residual) (+​A1:2007).
London: British Standards Institution.
British Standards Institute. (2006). BS EN 14889-​1:2006–​Fibres for Concrete. Steel fibres. Definitions,
Specifications and Conformity. London: British Standards Institution.
British Tunnelling Society (BTS) (2004). Tunnel Lining Design Guide, Institution of Civil Engineers,
Thomas Telford, , London, England. ISBN-​13: 978-​07277-​29-​866, ISBN-​10: 072-​77298-​61
British Standards Institute (BSI) (2016). Tunnel Design–​Design of Concrete Tunnel Lining Code of
Practice, PAS 8810:2016, British Standards Institution (BSI). ISBN 978-​0-​580-​88170-​1
508 Handbook of Precast Segmental Tunnel Lining Systems

GERMAN TUNNELLING COMMITTEE CODES OF PRACTICE (DAUB)

Empfehlung für Dichtungsrahman in Tübbinguaskleidungen (Recommendation for Gasket Frames in

Segmental Tunnel Linings, Studiengesellschaft für Tunnel und Verkehrsanlagen e. V (STUVA)
(2019), Koln, Germany. ISBN 978-​3-​9819842-​2-​4
Recommendations for the Design, Production and Installation of Segmental Rings, Studiengesellschaft
für Tunnel und Verkehrsanlagen e. V (STUVA) (2019), Koln, Germany.

INTERNATIONAL FEDERATION FOR STRUCTURAL CONCRETE (FIB)

International Federation for Structural Concrete (fib), Precast Tunnel Segments in Fibre-​Reinforced
Concrete–​State of the Art Report, No. 83, ISBN 978-​2-​88394-​123-​6, October 2017), ISBN
978-​2-​88394-​123-​6, October 2017)–​pdf format
fib Special Activity Group 5. (2010). Bulletin 55-​ 56: Model Code 2010–​ First Complete Draft.
Lausanne: International Federation for Structural Concrete (fib).
fib Working Party 1.4.1 Tunnels in Fibre-​Reinforced Concrete. (2017). fib Bulletin 83 Precast Tunnel
Segments in Fibre-​Reinforced Concrete State-​of-​the-​Art Report. (A. Meda, Editor) Lausanne,
Switzerland: Fédération International du Béton (fib).

INTERNATIONAL ORGANIZATION FOR STANDARDIZATION

ISO/​TC 17/​SC 17 Steel Wire Rod and Wire Products. (2013). ISO 13270:2013–​Steel Fibres for
Concrete –​Definitions and Specifications. Geneva, Switzerland: ISO, the International
Organization for Standardization.

JAPAN SOCIETY OF CIVIL ENGINEERS

Japan Society of Civil Engineers, Standard Specification for Tunnelling-​2006–​Shield Tunnels.

INDUSTRY GUIDELINES
National Precast Concrete Association (NPCA). (20 Jan 20). NPCA Quality Control Manual For
Precast Concrete Plants 14th Edition. Carmel, IN 46032: National Precast Concrete Association
(NPCA).
Precast/​Prestressed Concrete Institute. (1999). MNL-​116-​99: Manual for Quality Control for Plants
and Production of Structural Precast Concrete Products, 4th Edition. Chicago: Precast/​
Prestressed Concrete Institute.

TECHNICAL PUBLICATIONS
Allen, C. (October 2009). Fibre Decider; Tunnels and Tunnelling International. Tunnels and Tunnelling
International, vol. 2009, pp. 35–​37.
Bakhshi, M. (2015). Design of Segmental Tunnel Lining for Serviceability Limit State, AASHTO
SCOBS T-​ 20: Technical Committee for Tunnels April 2015 Meeting. Saratoga Springs,
New York .
Bakhshi, M., and Nasri, V. (2017). Water Infiltration and Crack Control for Tunnel Concrete Lining.
Proceedings of the World Tunnel Congress 2017–​Surface Challenges–​Underground Solutions
(pp. 1–​10). Bergen, Norway: ITA–​AITES.
Bekaert-​Maccaferri Underground Solutions. (2018). Info Sheet: What About Stray Currents
With Steel-​Fibre Reinforced Concrete. Ghent, Belgium: Bekaert-​Maccaferri Underground
Solutions.
 Formwork systems for precast segmental tunnel linings 509

de Rivaz, Benoit. (2016). Design Approach Assisted by Test. Ghent, Belgium: Bekaert-​Maccaferri
Undergound Solutions.
de Rivaz, Benoit. (2019). Characteristics of Steel and Polymer Based Fibre Concrete. Zwevegem,
Belgium: NV Bekaert SA.
de Rivaz, Benoit. (2020). Tunneling Applications: Concrete Segment Reinforcement. Zwevegem: NV
Bekaert SA.
di Prisco, M., Colombo, M., and Dozio, D. (2014). Fibre-​Reinforced Concrete in fib Model Code 2010;
Principles, Models and Test Validation. Structural Concrete 14 (2013), No. 4, pp. 342–​361.
Edvardsen, C. (October 2010). The Consultant's View on Service Life Design. In K. van Breugel, Guang
Ye, Yong Yuan (eds.), 2nd International Symposium on Service Life Design for Infrastructure
(pp. 249–​ 264). Delft, The Netherlands: RILEM–​ International Union of Laboratories and
Experts in Construction Materials, Systems and Structures.
Gerhard Vitt. (2011). Understanding Steel Fibre Reinforced Concrete: Dramix (Version 1.0 ed.).
Zwevegem, Belgium: NV Bekaert SA.
ITA Working Group 2–​ Research; Bakhshi, M; Nasri, V.; Main Authors. (April 2019). ITA
Report No.22–​Guidelines for the Design of Segmental Tunnel Linings. Longrine, Avignon,
France: AITES/​ITA.
ITA Working Group 2–​Research; Tiberti G.; Chiniottri E.; Pizzari G., Main Authors. (April 2016). ITA
Report No 16–​Twenty Years of FRC Tunnel Segments Practice: Lessons Learnt and Proposed
Design Principals. Avignon, France: AITES/​ITA.
ITAtech Activity Group Support; (2016). ITAtech Report No.7; ITAtech Guidance for Precast Fibre
Reinforced Concrete Segments–​Volume 1: Design Aspects. Longrine: International Tunnelling
and Underground Space Association.
ITAtech Activity Group Support; (2018). ITAtech Report No.9: Guideline for Good Practice of
Fibre Reinforced Precast Segment–​Volume 2: Production Aspects. Longrine: The International
Tunnelling and Underground Space Association.
Latson, J. (22 June 15). The Burning River That Sparked a Revolution. Retrieved from Time.
Com: https://​time.com/​3921​976/​cuyah​oga-​fire/​
Meda, A., and Rinaldi, Z. (July 2017). Beam Tests on Fiber Reinforced Concrete According to EN
14651 and ASTM C1609. Rome, Italy: University of Rome–​Tor Vergata.
Meda, A., Rinaldi, Z., Spagnuolo, de Rivaz, B., and Giamundo, N. (January 2019). Hybrid Precast
Tunnel Segments in Fiber-​Reinforced Concrete with Glass Fiber Reinforced Bars. Tunnelling
and Underground Space Technology 86, pp. 100–​112.
Meson, V. M. (2019). Durability of Steel Fibre Reinforced Concrete in Corrosive Environments–​Ph.D.
Thesis. Kongens Lyngby, Denmark: Department of Civil Engineering, Technical University of
Denmark.
Pascal Guedon (Arcadis) Working Group Leader. (Juillet/​Août 2013). Recommendation of AFTES
No.GT38R1A1–​ Design, Dimensioning and Execution of Precast Steel Fibre Reinforced
Concrete Arch Segments. Tunnels et Espace Souterrain–​No. 238, Association Française Des
Tunnels et de L’espace Souterrain, pp. 312–​324.
Professor Meda, A., and Professor Rinaldi, Z. (2015). Tests on Precast Tunnel Segment in Concrete
Newly High Tensile Strength Steel Fibers Dramix 4D 80/​60BG. University of Rome "Tor
Vergata", Civil Engineering. Rome: TERC–​Tunnelling Engineering Research Center.
Rack, J. (20 August 14). A Brief History of Sustainability. Retrieved from The World Energy
Foundation: https://​thewo​rlde​nerg​yfou​ndat​ion.org/​a-​brief-​hist​ory-​of-​sus​tain​abil​ity/​
Rossi, P. C.-​P. (18 April 16). SFRC Durability in a Chlorinated Environment. Tunnel Talk (Direct by
Design), pp. 2.
Tang, K. (March 2020). Corrosion of Steel Fiber Subjected to Stray Current Interference. ACI
Materials Journal, vol. 117, no. 2, pp. 99–​112.
Tunnel Talk Reporting. (25 April 19). TunnelTech: First Use of Macro-​Synthetic Fiber Reinforced
Concrete in Segments in the USA. TunnelTalk.
510 Handbook of Precast Segmental Tunnel Lining Systems

TunnelTalk Reporting. (March 2015). TunnelTech: Lab Comparison of Steel vs. Synthetic FRC
SegmentsTests and Results; by University of Rome Tor Vergaata. TunnelTalk.
United States Environmental Protection Agency, U. S. (17 September 20). Learn about Sustainability.
Retrieved from EPA.Gov: www.epa.gov/​sus​tain​abil​ity/​learn-​about-​sus​tain​abil​ity
Vandewalle, M. (2005). Tunnelling is an Art. Zwevegem, Belgium: NV Bekaert SA.
VMT–​Segment Documentation System SDS. Modular Production and Logistics Management: VMT
GmbH GB | Segment Documentation System SDS (vmt-​gmbh.de).
Chapter 9

Fabrication of precast segmental tunnel

linings
Lead author: Carlo Cattelan
Contributing authors: Marc Drouin, Fred Esch, and Van Tuan Chau

9.1 INTRODUCTION
The goal of this chapter is to introduce and sensitize the reader to the key elements of the
precasting of tunnel liners –​a series of complex multidisciplinary activities that require
planning, rigor, discipline –​it is a lot more complex than meets the eye. Production of pre-
cast segmental tunnel liners is to TBM mining operations as tires are to Formula 1 racing
motors and automobiles; a less glamorous and spectacular element, as well as relatively
modest in price compared to the car and motor (the TBM), but everything rides on them
and the success or failure of the TBM operation is heavily dependent on them. Within this
chapter, we will try to briefly touch on select topics of interest to provide a background for
the reader’s information.
Note to the reader, that the following chapter regarding the process of fabricating seg-
ments, is written in an informal style, as the subject matter discussed is done so from a
production person’s point of view, skimming and touching many topics that could deserve
a chapter to themselves, and this producer’s point of view is one that the designers and
stakeholders in the industry must account for to guarantee successful future projects. There
is rich information contained in this chapter, segment manufacturing information that does
not exist in this holistic form in the current literature. This chapter is geared towards inform-
ing the young upcoming generation of tunnel industry workers, with the goal of transferring
knowledge about the precasting operations and knowledge leads to confidence and com-
petence benefiting the industry. Typical precast segments are shown below in Figures 9.1
and 9.2.

9.2 GENERAL SUMMARY
For an overview on estimating costs and schedule for precast tunnel operations and running
a precast tunnel liner operation, there are many systems involved that require many talented
people with skills and judgment and it is a tour-​de-​force to keep operations smooth. Bidding
a job and the consideration of the depth and complexity of the various systems required will
humble the most experienced senior estimator because on top of everything to study, most
of the purchases, system set-​ups and operations must be predicted, with risks assessed over
the course of two to three years with volatile prices of equipment and raw materials.

DOI: 10.1201/9781032675541-9
512 Handbook of Precast Segmental Tunnel Lining Systems

Figure 9.1 Modern precast concrete segment –​ ­e xample 1.

Notes: A typical modern precast concrete segment used for mechanical tunnel excavation meth-
ods. Tunnel is 12.65 m ID (41’-​6 ”) requiring 4.2 m 3 (5.5 yd 3) per segment.

Figure 9.2 Modern precast concrete segment –​ ­e xample 2.

Note: Typical large tunnel diameter modern bolted and gasketed precast concrete segments used
for tunnel excavation method.

9.2.1 Schedule affects and impacts

With so many cost-​sensitive elements to deal with, one must remember that any change in
design and schedule can have major impacts. A job delay in 2020 may mean a production
being pushed to several unscheduled months in 2023 and may mean that there are three
extra months of production at 2023 prices in lieu of 2022 prices and the project will incur
three months of extra overhead costs. One must understand that certain suppliers provide
year-​based incremented pricing, and with an extended schedule, purchases for these materi-
als fall into a higher bracket when jumping to the subsequent year. Also, certain supplies
may be linked to an index and delays often have the index play against you. In aggregate,
these two material supplier conditions can have a significant impact on material heavy jobs,
as tunnel liner precast jobs can be.
 Fabrication of precast segmental tunnel linings 513

A schedule delay, when factoring in the extra costs of overhead, indexed material costs
and incremented material costs, can add up to hundreds of thousands, and even millions
of dollars of extra costs. The precaster must ensure to measure risks, and account for them
or qualify them in the quoting process. This is an important element that contractors and
owners must understand when they run their own risk analysis.
In 2022, we cannot forget to mention the impact that Force Majeure events have on a
precast tunnel liner manufacturing supply chain. The effects of pandemics and wars (which
will continue to plague humankind) are touching most aspects of the production as tunnel
liner jobs are material heavy, they rely on international suppliers, who source their raw
materials worldwide (some sources of raw material are disappearing and the law of supply
and demand, with the scarcity, increases costs and limits opportunity), which in turn means
a dependence on sea-​shipping, in which we have seen container costs doubling and tripling.
All major domestic products suppliers (cements, steels, aggregates) are feeling crunches and
breaking contractual agreements and increasing prices. Energy costs are increasing drastic-
ally, affecting transport costs and operation costs.
Many suppliers are breaking contracts, refusing to honor contracts, knowing that they
are guaranteed to lose money on a supply. The other important impact that the Force
Majeure events has is on the delivery time of equipment, materials, supplies and even com-
modities. Supply chain delays due to shortages of manpower, vital parts/​components, com-
mon parts, get compounded during the delivery/​transportation stage with the world logistic
systems, a once well-​oiled machine now tilted.
The number of shipping vessels or sea containers have decreased, and the result is that
shipping confirmations are cancelled, and rescheduled several times last-​minute. If that were
not enough, the situation in the major ports is affected with delays to unload and process
goods, and then finally, further delays await with trucking shortages backing up the ports.
Trucking and driver shortages have a tremendous impact on precast operations. This
includes everything that comes in or out of a precast plant, unless a rail line or barging is
available, is transported by truck. These Force Majeure events are a serious issue that the
owners need to address in the present moment and in future contracts upfront. The world
events happening at the writing of this chapter in 2022, and their effects on the precasters of
tunnel liners will most probably be the new normal in years to come with continued global
conflicts, probable global pandemics, prolonged climate change and coming demographic
changes. Understanding causes and effects helps to better plan major projects, to manage
them efficiently, to limit risks (and equitably allocate or share risks) and surprises, and to
avoid potential litigation.

9.2.2 Precast segment casting operations –​complex pop-​up set-​up

Everyone is aware of the recent concept of pop-​up stores where a company will rent a
bricks and mortar store and set up in it and occupy it for a fixed period, several weeks to
several months. Well, segment-​ring plants are often a complicated behemoth version of a
pop-​up precast operation, the reason being that existing precast operations are ongoing
businesses that are optimized for the products they serve in the local markets they serve.
The plants that are operating do not necessarily have the extra capacity in terms of batch-
ing capacity, aggregate storage capacity, materials staging capacity, floor layout spaces, and
most important, yarding spaces where the finished product is stored, as certain jobs require
up to 4.0 hectares (10 acres or more) of storage space depending on the contractor mining
schedules, which may all be different at bidding time.
514 Handbook of Precast Segmental Tunnel Lining Systems

Bidding precast tunnel liners is like bidding moving targets. And the challenge is always
to find a suitable place to set up operations. At times, one can copy and clone from existing
operations, at times one must set up from scratch. The costs of each set-​up vary tremen-
dously and thus it makes some of us chuckle when some people brandy about costs per liner
foot for segmental ring. The answer is always; it depends.

9.2.2 Historical background
A good place to start would be to take a moment to peruse the book, History of Tunnelling
in the United States, by David Klug, Michael Roach, Colin Lawrence and Brian Fulcher. In
this book, one will come to realize that that as shown in a borrowed slide from this book
that what we think is a recent technology, actually was developed in the United States
more than 100 years ago. Note that in 1906, a US Patent No.830,345 by A.F Mattson was
recorded for what in effect are precast concrete tunnel liners. Please refer to Figure 9.3 for
the original patent document.
There are many other examples of early tunneling techniques in Europe and Asia. For
example, the Metrosud tunnel in Naples, Italy in 1982 was the first application of steel fiber
reinforced concrete (SFRC) segments used on a significant public works project.
The rebirth of modern use of precast concrete tunnel liners –​installed with an erector
in a modern TBM was pushed forward in the 1980s by American tunnel contractors who
used them as first pass primary lining, commonly called “junk liners”. The contractors
were successful with the use of this multiple pass lining (as initial support), waterproofing,
cast-​in-​place final lining, method that was commonly used with geological conditions that
permitted this type of tunneling.
When owners and associated system programs required that the tunnel alignments be
driven in soft running soils or in mixed face conditions, the North American tunnel industry
looked to Europe and Asia, to where they were building tunnels in these challenging geo-
logical conditions using one-​pass precast tunnel linings with bolted connections and water-
proofing seals in conjunction with tunnel boring machines (TBMs) that controlled the
geological conditions while provide safe working conditions.
Within the one-​pass tunnel liner systems, there are two subcategories:

• Transportation tunnels, which will see a high volume of human or vehicular traffic
• Utility and water/​wastewater tunnels, which are part of infrastructure systems

Because transportation tunnels present a direct threat on human life, as opposed to the fluid
transport tunnels, transportation tunnels tend to have stricter design parameters, but in
reality, both types of systems merit the same attention and there is no reason to not ensure
that the specifications and fabrication of tunnel liners for either system be to the same
quality standards.
The forming systems used to make the segments have evolved over the years. Originally
the segments were formed on curved concrete bases that had movable sidewalls (made of
steel or wood as shown below in Figures 9.4 and 9.5). These are still used in certain applica­­
tions. The accuracy and precision of these systems, in the right hands, was good but even-
tually came the evolution to steel formwork systems. Steel forms are resilient, offer better
precision and accuracy, allow a larger variety of cast-​in elements, and permit quicker set-​
ups during plant operations. See Figures 9.6 and 9.7 and refer to Chapter 8, for additional
details on this subject.
 Fabrication of precast segmental tunnel linings 515

Figure 9.3 A.F. Mattson 1906 Patent for precast tunnel lining segments.
Note: This patent describes one of the earliest precast concrete segmental lining designs devel-
oped in the United States for shield-​d riven tunnels in soft soils.
516 Handbook of Precast Segmental Tunnel Lining Systems

Figure 9.4 Segment mold with concrete base and wood sidewalls (1980s).
Note: Example of an early, high precision precast mold used for bolted and gasketed segmental
tunnel liners.

Figure 9.5 Segment mold with concrete base and wood sidewalls with rebar cage.
Note: Example of an early segment design with rebar reinforcement placed in the mold.

Figure 9.6 Steel segment forms –​ open.

Note: Example of the modern fabricated steel high precision mold for bolted and gasketed segments.
 Fabrication of precast segmental tunnel linings 517

Figure 9.7 Steel segment forms –​ closed.

Note: Example of the modern fabricated steel high precision mold for bolted and gasketed
segments.

9.3 SEGMENT FABRICATION PLANT LAYOUT

A segment plant is a collection of systems, many computer-​controlled, that must be set-​
up, commissioned, operated, calibrated and maintained tightly. See in the sketch below
the layout of a complete tunnel liner manufacturing plant system and note the number of
systems (some independent and some dependent/​interconnected). And the systems are run
by a well-​orchestrated team of lead personnel, engineers, technicians, mechanics, laborers,
operators, quality control and quality assurance personnel, truckers and support staff. One
can note in Figures 9.8 to 9.10 below or refer to Chapter 8. This is an incomplete list, but
one gets the picture.

• Office trailers • Batching equipment • Door systems

• Quality control quarters (mixer, silos, screw con- • Water pumping and distri-
and equipment veyors, belt conveyors, bution systems
• Waste management batching plant, aggregate • Water pumping systems –​
systems storage) clear and wastewater
• Water treatment units, • Storage of cast-​in • Welding machines
• Heavy wheeled equip- elements • Building maintenance
ment (loaders, fork- • Steel fiber feed systems, • Facilities maintenance and
lifts, extended reach • Steel reinforcing assembly improvements (electrical)
units, etc.) jigs and areas • Storage yard leveling
• Overhead crane capacity • Air compression systems • Maintenance
• Segment molds and distribution • Infrastructure preparation
• Carousel systems • Electrical distribution • Lighting systems
• Finish line systems systems • Evacuation and fan systems
• Curing equipment –​kilns • Pneumatic systems and • Monitoring systems
(primary and secondary) controls • Security systems
• Vapor generators, vapor • Water heating systems • Release agent storage, dis-
lines and valves • Water chilling systems tribution and application
• Water softeners • Natural gas entries or equipment
• Control systems propane gas tank farms
and systems
518 Handbook of Precast Segmental Tunnel Lining Systems

Figure 9.8 Example of a carousel precast tunnel liner fabrication plant.

Note: Traylor-​Technopref Precast, LLC., Tacoma, Washington; Sound Transit Project U220–​U 230
Seattle.
 Fabrication of precast segmental tunnel linings 519

Figure 9.9 Example of a carousel precast tunnel liner fabrication plant.

Note: Technopref Industries, Inc, Minneapolis–​S t. Paul LRT Project.
520 Handbook of Precast Segmental Tunnel Lining Systems

Figure 9.10 Example of a carousel precast tunnel liner fabrication plant.

Note: Bay State Precast LLC, Brandywine, Maryland; Blue Plains, CSO, Anacostia River Tunnel
and the Northeast Boundary Tunnel.
 Fabrication of precast segmental tunnel linings 521

9.3.1 Observations and conclusions

Many of these systems are noisy, greasy, hot, create water/​moisture, create many different
types of dust, constantly create dirt and rubbish, system components wear out with time
and use, and the consequences are:

• It is a major challenge to keep the housekeeping under control during the operations
and maintenance of systems
• Filtering systems have trouble keeping up. Layers and coating of dust and debris make
equipment and parts overheat or run hotter than design temperature. As a result,
system components often wear out prematurely with heavy use, and equipment that
often sits in idle mode is often the greater target of the corrosive environment
• These roughest of environmental conditions for equipment (mechanical and elec-
tronic) are rarely met with such variety in any industrial setting and at times military
grade components are the only way to go, and even with that, equipment break-
downs, failure and troubleshooting and fixing-​on-​the-​fly are part of everyday life in
a precast plant.

Keep in mind that it takes quite a good level of competence and expertise to run a precast
tunnel liner operation including the following essential activities.

• Set-​up the production line and supply chain on time

• Produce segments on time
• Facility and equipment maintenance to produce with the expected quality (the highest
standards of the industry)
• Deliver high-​quality finished segments on time.

Precast plant operators get a little offended when someone criticizes an operation without
consideration of all the issues and potential causes.

9.4 CAROUSEL AND STATIONARY SEGMENT FORMWORK

PRODUCTION
Two types of production methods are used depending on the size of job, and speed at which
one needs to produce (predicated on the TBM contractor installation schedule, and yard
space/​project storage space available).

9.4.1 Stationary mold production

Stationary mold production is the simplest layout of production (not necessarily any sim-
pler to operate since each type of production has its advantages, disadvantages, and requires
different skill sets). Stationary production lines typically have a shorter and less expensive
set-​up time.
Stationary mold productions, in order to be efficient with a tight schedule, are generally
required to cycle the molds twice per day or require a giant production hall that will accom-
modate multiple sets of molds (and these large production halls are quite rare in certain
countries).
522 Handbook of Precast Segmental Tunnel Lining Systems

Figure 9.11 Stationary mold production setting in a covered plant enclosure.

Note: This is a very customary modern segment forming operation where all molds remain fixed
in place and are cast and demolded once or twice per day depending on the operational schedule.

With small diameter tunnel liners, it is easier to fit many mold sets on a floor layout and
they are relatively cheaper than large diameter tunnel molds that also gobble up two to three
times more floor space.
There is a calculation that needs to be made regarding working with fewer mold sets on
two shifts or more mold sets (which have a considerable capital cost) on one shift.
It should be noted that if producing on one shift is possible, the cost of the concrete can
be reduced in the sense that one would have 24 hours to reach the stripping strength. With
a normal fixed-​mold operation or a carousel operation where one would turn the molds
twice (or more per day), the concrete design mix is driven by the early strength that must
be achieved in about 6 hours or so, hence a richer mix (more costly) is required. However,
a richer mix well designed, has better durability characteristics.
The stationary mold production arrangements as shown in Figures 9.11 and 9.12 have
smaller capital costs (no semi-​automated carousel system expense) but there is one disad-
vantage; the output of the plant is limited to at best cycling the number of molds on the
plant floor twice a day. In certain circumstances, this system may not meet the TBM con-
tractor’s schedule or peak mining schedule. The segments are typically cured by covering
the stationary molds with tarps and then injecting steam under the tarps from steam pipes
located under the molds.
Please refer to Chapter 8, for in-​depth details and photos of both systems of production.

9.4.2 Carousel segment production

Carousel segment production can come in a semi-​automated format and more recently in a
quasi-​fully automated format. The disadvantage of a carousel is the upfront capital cost and
the maintenance costs of operation/​downtime but in certain circumstances it is the only way
to go. Please refer to the fabrication plant layout images in Figures 9.8 to 9.10.
One disadvantage of the carousel system is that it requires a minimum size of floor
space that is predicated on a kiln size that is in turn predicated on the amount of min-
imum hours that the segments must remain in the primary steam curing kiln (remember
that early kiln ambient temperatures must not affect the concrete preset –​and that
 Fabrication of precast segmental tunnel linings 523

Figure 9.12 Stationary segment molds placed in a covered plant enclosure.

Note: This is a very customary modern segment forming operation where all molds remain fixed
in place and are cast and demolded once or twice per day depending on the operational schedule.

depending on the cementitious materials used, there are maximum internal concrete tem-
peratures that must be respected by North American codes –​between 70 and 77°C (158
and 170°F). A primary kiln will usually have two or more temperature zones, with zones
having higher set point temperatures as the molds travel through. Thus, when preset and
internal temperature restrictions are factored into the design, along with the fact that
the specifications in North America call for a slow acting type-​II cements (now being
replaced by environmentally favorable type-​1L cements) and steam curing, a carousel
will require more molds and a larger primary kiln so that the segments are metered out
at about a 6-​hour interval.
The advantage of the carousel system is that it can be run continually, if need be 24 hours,
7 days per week, with segments spending about 6 hours in the primary curing and coming
out at a regular interval (say between 10 and 15 minutes). It must be understood that 24/​7
is a theoretical principle because real life gets in the way, and there is equipment mainten-
ance, break-​down downtime, manpower exhaustion and inefficiency, etc. to consider, but
the possibility of running in a quasi-​continuous manner exists. Please refer to Figure 9.13
for an image of a modern carousel arrangement.
Another advantage of a carousel is that it is possible, with considerable effort, to integrate
two different productions on the same manufacturing line/​operation. The logistics of run-
ning two jobs on the same line are demanding as different mix designs may be involved, but
the carousel system offers this flexibility.
The capital cost of carousels gets offset by the volume of segment-​rings required for
the project and the possibility of running several projects out of the same plant. Also, the
additional capital cost of a carousel must be measured against the savings on labor that
524 Handbook of Precast Segmental Tunnel Lining Systems

Figure 9.13 Carousel segment production operation running two different projects.

Note: This is a unique carousel segment forming operation where all molds are mobile and in this
case two separate projects and segment designs are being cast concurrently.

a carousel system has as an advantage on a fixed-​mold production. Well run, a carousel

should produce more segments per manhour that a fixed-​mold (fixed-​pour) operation.
A general weakness of a carousel is that of any manufacturing line operation, one is only
as fast as the work at the slowest workstation because one cannot move the line until each
station has finished its work. But the critical weakness of a carousel is that if you have an
equipment problem –​the whole manufacturing line is down. It is very important to keep
the carousel maintained and be able to quickly troubleshoot issues or else it is easy to lose
entire days of production.
A carousel production has many important advantages and some disadvantages, it is
essential to understand them, when trying to predict an output. The authors are not try-
ing to be negative with respect to the carousel systems, they are wonderful solutions in the
proper circumstances. Refer to Chapter 8 for in-​depth details and photos of both systems
of production.

9.5 REBAR REINFORCED CONCRETE

Rebar reinforcing components are assembled in precise jigs to form rebar cages, and then
the resulting cages are transferred into the molds, supported on reinforcing chairs (typically
of concrete that has similar durability properties such as the concrete, or plastic chairs),
and secured in the mold. There are two main ways to reinforce segments with rebar-​type
reinforcing:

• With deformed bar (rebar) type reinforcing

• With deformed wire type reinforcing

Below we will provide a brief introduction to both main types of reinforcing materials.
 Fabrication of precast segmental tunnel linings 525

9.5.1 Rebar-​t ype reinforcing

Rebar reinforcing comes in many varieties including the following in most markets.

• Classic rebar reinforcing –​standard reinforcing –​for example ASTM Grade 60

reinforcing steel (deformed –​with the ridges), f’y =​60,000-​psi (60 ksi) [415 MPa],
which can come in the normal grade and the weldable grade. This type of rebar can
come in higher grades –​such as Grade 80, f’y =​80,000-​psi (80 ksi) [550 MPa], but
beware that availability of higher grades with specific characteristics may be limited
because they may be rolled only by certain mills in certain parts of the country, there
may be limited factory rollings per year, and there may be minimum order quantities
(a tunnel liner job may require thousands of tons of steel but this a minor quantity
when considering a mill’s daily rolling capacity). If an engineer is considering higher
grade specialty steels –​careful coordination with steel mills is a must.
• Deformed wire products that come in various grades anywhere from the following (it
is possible to draw the wire to a specific grade (i.e. tensile strength), where the wire
is drawn out in dies and then deformed, which look like regular rebar products, but
they are of high carbon and are weldable.
• Grade 60 f’y =​60,000 psi (60 ksi) [415 MPa]
• Grade 80 f’y =​80,000-​psi (80 ksi) [550 MPa]
• Grade 85 f’y =​85,000 psi (85 ksi) [585 MPa]

9.5.2 Classic rebar reinforcing

The engineering base drawings in the USA are usually calculated and shown as classic rebar –​
in Grade 60 [415 MPa] –​following ASTM A615 Standard Specification for Deformed and
Plain Carbon-​Steel Bars for Concrete Reinforcement. These bars can be rolled into round
bars, hooked bars, stirrups, straight bars and assembled with tie wire. However, the dimen-
sional stability of the assembled cages is paramount because the concrete covers must be
respected, and a deformed rebar cage will infringe on covers and possibly come apart during
the concreting operation where the mold undergoes high frequency vibration (unless tack
welds are permitted). Please refer to Figure 9.14.
The classic rebar is sized in the USA as Nos.4, 5 and 6, etc., each number depicting the
diameter of the bar relative to one inch –​for instance, a No.4 is 4/​8 of an inch =​0.50 inches
(12.7 mm) diameter. A No.7 bar is 7/​8 of an inch =​0.875 inches (22.1 mm) diameter. The
bar sizes of classic reinforcing are fixed, and one can only play with the spacing of the bars
to obtain a desired area of steel for a concrete section. The metric equivalent diameters are
as follows:

• No.10 (10 mm diameter)

• No.15 (15 mm diameter)
• No.20 (20 mm diameter)

Note the release agent sheen on the steel formwork elements. Mastering the choice and
specific/​correct application of the release agent is crucial to achieve a good durable concrete
finish and aids in the maintenance of the molds. Several different release agent products
for specific areas and steel treatments are part of the precaster’s release agent arsenal. This
little-​known subject could merit its own chapter because the esthetics are always an element
526 Handbook of Precast Segmental Tunnel Lining Systems

Figure 9.14 Classic rebar welded cage placed in a segment mold.

Note: In this example, deformed bar cages are arranged for the perimeter of the segment with
some “prefabricated ladder” reinforcing used for the interior.

of visual concern for precasters, engineers and clients, and concrete finish may in some
instances be an aspect of the overall durability of the precast elements.
Cages assembled with tie wire must be well designed and tied with great care. If one
uses type-​706 weldable rebar (in USA market), per ASTM A706 Standard Specification for
Deformed and Plain Low-​Alloy Steel Bars for Concrete Reinforcement), one can tack-​weld
the components in strategic locations that will provide good rigidity. One can assemble the
components solely with weld tacks or with a hybrid solution of tacks and tie-​wire. At times,
standard ASTM A615, Deformed and Plain Carbon-​Steel Bars for Concrete Reinforcement,
rebar may be allowed to be tack welded for assembly purposes. In Europe and Asia and
other parts of the world, steel wire products are the norm, with high carbon drawn out steel
that is weldable, such as the ASTM A1064, Standard Specification for Carbon-​Steel Wire
and Welded Wire Reinforcement, Plain and Deformed, for Concrete.

9.5.3 Deformed wire products

Deformed wire products are fabricated in North America to ASTM A1064/​A1064M,
Standard Specification for Carbon-​Steel Wire and Welded Wire Reinforcement, Plain and
Deformed, for Concrete. Note that ASTM 1064 pertains to the deformed wire itself, and
ASTM C 1064M pertains to Deformed Wire that is fusion welded into a mesh (or a fusion
welded shape as a ladder). See Figure 9.15 for examples of wire and welded wire products
and cages.
 Fabrication of precast segmental tunnel linings 527

Figure 9.15 Wire and welded wire products.

Note: Examples of several wire and welded wire products commonly used for prefabricated
reinforcing elements–​a ssembled in cages.

Generally, the cages that result may be a combination of components of different grades,
shapes, loose bars, welded components. The components are assembled in precision jigs by
tack welding (or they could also be tie-​wire assembled). The great advantage using deformed
wire are the following:

• We can save on steel and still provide the same performance

• We can tailor the bar sizes to the exact needs because bar sizes can be custom made
and with these, one can play with the spacing (if one wants 10 bars across an area, the
exact area of the steel required can be converted into 10 bars of equal area)
• Wire products can be welded, and this is useful for cage assembly speed and cage
rigidity

Savings on steel with deformed wire include the following:

• If using Grade 60 reinforcing steel (60 ksi) (415 MPa) steel; (European most used
equivalent is 500 MPa steel) and comes up with a desired number of required reinfor-
cing bars, one will be designing with a specific cross-​sectional area of reinforcing
steel. But in the case where the reinforcing is available in Grade 80 reinforcing steel
(80 ksi) steel; (European equivalent 550 MPa steel) grade, the equivalent area will
be 60/​80 =​0.75, resulting in a 25% reduction in the amount of reinforcing steel
required.
• By using the deformed wire products, 25% of the steel weight required to reinforce
the segment rings is saved and on a typical project and it could mean saving thousands
of tons of steel and such savings go to the owner and taxpayer on public projects. And
it could be argued that it is an ecological advantage too because our natural resources
can be put to the better use. Many but not all reinforcing designs and configurations
lend themselves to these savings by reducing weight afforded by the increasing wire
strength.
528 Handbook of Precast Segmental Tunnel Lining Systems

Figure 9.16 Prefabricated welded wire in mold.

Note: Prefabricated welded wire reinforcing elements pre-​a ssembled into a cage and placed in
the mold for as prefabricated reinforcing.

Wire bar sizes are sized as per the final area of the wire that is drawn. The nomenclature
is designated as dXX. For example, a d25.4 bar would have an area of 0.254 in2
(163.87 mm2). The maximum size of bar that can be made is a d31, 0.31 in2 (200 mm2).
Please refer to Figures 9.15 and 9.16 below that illustrate deformed wire products in the
precast plant and placed in a segment mold. Additional welded wire products are shown in
Figure 9.17.

9.5.4 Reinforcing design
It is usually best for the Tunnel Liner Design Engineer to show the reinforcing as classic
reinforcing and let the precaster come up with the reinforcing equivalence scheme as there
are many ways to achieve the intended result. Each precaster may have preferences on which
type of assembly systems they prefer, based on a variety of factors, such as facility layouts,
expertise with the assembly methods, industry supply chain issues and opportunities, etc.

9.5.5 Rebar cage design vs. segment mold design

To properly design steel reinforcing cages, one must start with a 3D drawing of each mold
and design the steel cage in 3D, carefully sizing each of the components, and deciding the
tolerances of each of the components (some have +​zero, –​x tolerances and some have –​zero,
+​x tolerances, and some simply have +​or –​tolerances). In North America, classic rebar
manufacturing typically has larger tolerances with plus +​minus tolerances whereas wire
products often have plus zero and minus zero tolerances available because of the nature
of fusion welding machines that ensure fusion assembly using wire position guides which
account for the tighter tolerances.
In turn, the 3D drawings of the components and molds will be used to create the 3D
design of the reinforcing assembly jigs required. Remember that the ring is tapered, and the
cages need to be designed to fit every mold, so one designs the cage width to respect the
narrowest mold.
 Fabrication of precast segmental tunnel linings 529

Figure 9.17 Examples of various wire and welded wire products.

Notes: Typical wire and welded wire reinforcing steel products commonly include the following;
(a) Prefabricated welded wire curved ladders. (b) Straight and loose wire reinforcing products.
(c) Prefabricated welded wire straight ladders. (d) Prefabricated and bundled welded wire curved
mats (commonly referred to as mesh).

There are many embedded elements in a mold and these embedded elements are attached
to steel components that protrude into the mold sidewalls and bottom; there are tapered
hole formers for bolt holes and shear cone reservations to contend with and covers must
be respected at each of these pinch points and thus the design of a reinforcing cage is very
complicated. There is also the fact that the sidewalls and end walls are not at a perfect 90°
to the mold bottom surface and this also adds complexity to the design of the cages and
cage assembly jigs.

9.5.6 Rebar chair systems

Rebar chairs are typically required to be of a concrete quality that is of a similar durability
characteristic to that of the segment concrete. Polymer-​based chairs (at times with steel
cores) are now more accepted, but historically there was some concern with the polymer
dilating under heat and shrinking after the concrete cools down and thus allowing an ingress
530 Handbook of Precast Segmental Tunnel Lining Systems

for water to get to the rebar, however, nowadays, there are many different chair geometries
and polymer type systems, with chemistries and behaviors that are available to control these
effects, in consideration of the relatively low concrete curing temperatures. The tunnel liner
design engineer decide on this issue, case by case. In some circumstances with extremely
high reinforcing ratios, the weight of the cages is such that there is little choice but to use
polymer chair systems. Other times, because of the geometry of the cages, the plastic chair
systems offer better stability of the cage during casting vibration. In general, there is a wider
range of choices of polymer chair systems, with a more limited range of choices for concrete
chair systems.

9.5.7 Securing instrumentation and sensors in molds

A practice that is becoming more and more in-​vogue is the integration of embedded
sensors in precast segments and the industry’s enthusiasm for the new technology is
not being matched by the thought to the less complex elements required for securing
this high-​tech equipment in the mold. The ideal is to design adapters/​chairs that can
be attached to the reinforcing and especially if there are locations without reinforcing,
special chairs or small sacrificial reinforcing cage elements need to be thought out and
carefully designed.
At the moment, there is often reliance on site-​improvised support techniques that often
create long line-​of-​fire discussions with the technicians that supply and coordinate the sen-
sors. As can be seen in Figures 9.18 and 9.19 below, although these improvised systems get
the job done, they are very time consuming to put in. Additionally, there is always second
guessing as to confirming if the instrumentation devices stayed in place under the influence
of concrete compaction with vibrators located under the mold intrados skin. There must be
a better way to achieve the goal of securing the sensor systems rapidly. Installing this type
of arrangement on several molds negatively impacted the segment production rate by more
than 85%…

Figure 9.18 Securing sensing equipment inside a mold.

Note: Example of securing sensor equipment with improvised site made supports to reinforcing
and intrados skin–​b etter ways must be developed.
 Fabrication of precast segmental tunnel linings 531

Figure 9.19 Securing extrados and intrados sensing equipment inside a mold.

Note: Example of securing sensor equipment on both extrados and intrados areas with the pre-
fabricated rebar cage and supports and conduits to intrados skin.

Additionally, the engineers and new technology suppliers must keep this in mind and
design systems that can easily be put into the molds so that the industrial flow of work is
not impacted by the punctual operations related to the discrete installation and monitoring
of the specialized embedded sensors. Moreover, many of these embedded sensors require an
activation procedure that requires a piece of electronic equipment to be rubbed against the
area over which the sensor is embedded. With the precaster’s production segment stacking
procedures and arrangements, this extra instrument activation process can easily impact the
regular production rates because of the following considerations.

• Segments may need to be kept separate for this to occur.

• This operation by a third-​party technician may take up to 15 minutes or more and
would thus interfere with workflow.

Understandably, a lot of creative engineering and design are going into the addition
of these systems in the quest for Big Data Collection, but little thought is being given on
how to implement this technology in an industrial production setting. Without proper
planning, a precast segment production can be reduced as much as 80% due to the dis-
continuous additional operations: this is very costly, and the extra costs must get passed
onto the owner. Recent experience has shown that the client, owners and engineers are
often astounded at the actual additional costs for installation of embedded instruments at
the precast plant.

9.5.8 Securing rebar cage in molds

It is very tricky to secure a cage in a mold with very few elements to secure to. Basically, if
one is not allowed round polymer spacers that keep the cages from mold sidewalls (circum-
ferential joints and radial joints), one can only secure a cage with wire ties to the polymer
inserts for bolts or dowels, and against the bolt hole forming pins. It is an art and science to
532 Handbook of Precast Segmental Tunnel Lining Systems

Figure 9.20 Securing rebar to a mold–​s pear bolt.

Note: Example of securing the prefabricated rebar cage to the mold using the spear bolt hole
former unit.

Figure 9.21 Securing rebar to a mold–​d owel.

Note: Example of securing the prefabricated rebar cage to the mold using the dowel socket
insert.

get it right and keep the cages secure. See Figures 9.20 and 9.21 for examples of how a steel
reinforcing cage is tied down.

9.5.9 Rebar management and assembly shop

The challenge with a rebar cage assembly shop is the management of the multiple different
types of components, the staging of the raw material, the staging of the finished cages, how
one gets the cages to the molds. The demands in terms of real estate for shop floor layout
and staging areas can be important and is always a challenge for the precaster. An assembly
 Fabrication of precast segmental tunnel linings 533

floor layout will typically consist of a minimum of six to eight jigs with the accompanying
welding machines and storage racks.
The storage of steel reinforcing components alongside of the assembly jigs is critical and
there is always the complexity of the large number of components with certain jigs: for
instance, with a counter-​key segment that has one square end and one skewed end where
one can have up to 25 to 30 bars of different lengths to deal with for just one cage. In these
circumstances, at times it is more efficient to order several bars of the same length and cut
them to size and field fit, but this is going to increase the tonnage on the job but may offset a
combination of labor, staging area needs, and operation costs. It may raise the overall price
a little but simplify the operation and may lower the overall cost of the operation and it
would be a choice that the precaster makes.
Special considerations in terms of classic steel reinforcing schemes
A word of caution to those who want to estimate the cost of steel classic reinforcing designs
simply by measuring theoretical lengths of bars on a design concept drawing: this can get
you close but you may also be far off in terms of the actual quantity, costs, and price for the
job. There are a few factors to give consideration to that affect the total final cost of reinfor-
cing steel, such as the following:

• Waste factors coming from the manufacturing process (i.e. cutting and bending and
steel cutouts to avoid interference with embedded items).
• Application of CRSI (Concrete Reinforcing Steel Institute) industry practices based on
its Manual of Standard Practice, as it relates to detailing and the calculation finished
product billing.
• Waste and costs associated with the Engineer of Record (EoR) special non-​industry
standard tolerances.
• Waste and efforts required to build, test and modify assembly jigs in order to assemble
reinforcing cages to EoR special non-​industry specifications.
• Waste and efforts to run full size test steel cages, and extra costs related to small pro-
duction runs of specialty steel products.

With certain designs and large volumes/​tonnage, as is the case in the repetitive cage tight
tolerance design configurations of tunnel liners, the consideration of these factors can sig-
nificantly increase the overall cost relative to the cost number derived from actual manufac-
tured bar length calculations.
In every operation there is cut-​off waste (drops) from cutting and bending of reinfor-
cing and some shaped reinforcing elements have more than others. There is no universal
“one-​size-​fits all” waste factor. As an example, if reinforcing is coming in 40-​foot lengths
(≈12 meters), and stirrups are being made in an automated straightening/​bending machine,
because of the lengths of the bar required for a particular stirrup, one could be left with
39 inches (≈1 meter) of a useless piece of reinforcing if the minimum bar length required for
the stirrup is 42 inches (≈1.066 meters).
One must consider that in certain markets, the supply of reinforcing steel may be in
lengths that are disadvantageous with respect to being left over with cut-​off “drops” that
must be considered scrap steel. There may be bars that are only available in 20-​foot (6.1
m), 40-​foot (12.2m) or 60-​foot (18.3 m) lengths and the lengths available may create more
waste than if there was a choice in the supply. At times, the drop or left-​over short bar can
be re-​used in certain instances (with further handling and work such as cutting and bend-
ing), at times, it cannot. If coils or bobbins of wire reinforcing are available, cut-​off waste
534 Handbook of Precast Segmental Tunnel Lining Systems

is reduced because of fewer drops afforded by the greater overall lengths of bars that are
wound up around spools. These details matter when comparing the cost of reinforcing steel
designs. It is not a matter of just counting the theoretical lengths and applying a random
2% waste factor!
There is also a waste factor and premium to pay that can come with special geometries
where the design engineer insists on maintaining zero tolerances (combinations of plus X;
minus 0, or plus 0; minus X on covers for instance) with regard to concrete covers. There
are some shapes of bars that are difficult to maintain in the fabrication process because of
the mix of:

• Larger diameter of bars

• Location of the rib/​deformations on the bars with respect to the bend
• Chemistry of the bar (high strength steel bars may vary in strength, they will meet the
minimum strength but can be higher)

These factors combined may mean that the there is an inconsistent bend or bounce back
in the bar as it comes off the pin that the bar is bent around, and the final shape becomes
out-​of-​tolerance and the bar/​reinforcing element is considered rejected or scrap. Certain
bar sizes, bar grades, and shape combinations may be prone to more geometry tolerance
errors and thus create more rejects and scrap. And certain bar and shape combinations may
require more set-​up trials in order to find the correct machine settings and precise set-​points.
Since these learning curve trials create scrap, these quantities must be addressed and paid
for. At the same time, everyone must understand that there are tolerances that accumulate
and compound along the process with different parts of the process needing the plus/​minus
allowances, some requiring minus 0; plus X tolerances, some requiring plus 0; minus X
tolerances, etc.
The final achievable tolerances are a function of numerous factors including the following:

• Combination and cumulative of tolerances of a component (i.e. a stirrup, for example)

• Tolerances of fabrication of the machine that makes the stirrup, or the jig that the
stirrup requires for final acceptable shape, coupled with tight tolerances on possibly
edge beams (a subcomponent)
• Jigs the edge beams are fabricated in
• Eventual compounding effect from the final assembly of a cage
• Jigs that are used to assemble the segment rebar cages
• Large-​scale potential deformation of the assemblies due to the cumulative thermal
deformation effects of the tack welds (the tack welds themselves having negligible
effects on the steel bar intersections but the overly long assemblies may distort slightly)
• Installation/​placement in a mold, which has its own set of plus or minus tolerances.

There is a reason why there are standard plus/​minus specifications in governing standards
such as ACI, and when we get more restrictive, the costs go up with the unprecedented and
astounding levels of precision that are being demanded. The achievement of extreme dimen-
sional tolerances is often an iterative process with trials and readjustments to elements, the
process, and the reinforcing steel jigs used, and the costs of these must be accounted for and
compensated by the entities demanding great precision in the final product.
With regard to what is meant by the CRSI effect, it should be noted that bar lists in
almost all of the United States and Canada are developed in accordance with the Concrete
 Fabrication of precast segmental tunnel linings 535

Reinforcing Steel Institute (CRSI) theoretical lengths. This is the practice of using the sum
of the individual out-​to-​out lengths of a bent bar to determine its overall length. This intro-
duces the phenomenon of “bend curve deduction”, which has an effect on the length of the
bent reinforcing steel that is listed and the invoiced weight on unit price projects.
For straight pieces there is no impact. For long single-​end or double-​end bent members,
however, “bend curve deduction” is relatively minor, but when a large percentage of the
project is short members such as stirrups with mid-​size diameter bars such as No.5, 6, and
7, the listed and thus invoiced lengths of the stirrups can be significantly longer than the
length of bar sheared to fabricate the item.
The phenomenon of “bend curve deduction” is best illustrated by an example, such as
a simple L-​shaped bent bar, a 24-​inch (610 mm) by 24-​inch (610 mm) that is bent at 90
degrees. For such a shape, one will be invoiced for 48 inches (1,220 mm) of bar, no matter
the diameter of bar used, but there is a difference to note. If a small bar diameter is used,
such as a No.4 bar (No.15 metric bar equivalent), the total length of the actual bent bar will
be shorter than 48 inches (1,220 mm) because of the rounded corner, it will be 46.5 inches
long (1,180 mm). If a large bar diameter is used, such as a No.8 bar (No.25 metric bar
equivalent), the total length of the actual bent bar will be shorter than 48 inches (1,220 mm)
because of the rounded corner, it will be 45.5 inches long (1,155 mm). But note that the
L-​shaped bar made with a larger diameter bar has a shorter actual length than an L-​shaped
bar made with a smaller diameter bar but in both cases, the invoiced L-​shaped bars will
be based on 48-​inches (1,220 mm). In both cases, one pays for the theoretical out-​to-​out
dimension of the L-​shaped bar and not the actual length of bar used, and the “bend curve
deduction” factor is bigger with the larger bar.
One may ask, is this practice and reasonable for measurement and payment consider-
ation? In reality, it is standard industry practice for the manner in which materials are meas-
ured and paid for. It may be that one of the reasons for the practice of using the out-​to-​out
dimensions to define the length is to keep the process simple and consistent. All single-​end
bent bars that span 2 feet × 2 feet are listed as 4 feet long on all of the documents on the
project. The design engineer’s drawings might show this. The detail drawings, shop bar list
and item tags certainly do. The only place that the reduced length is shown is for the cut
length listed on the shear cutting list. Everywhere else, the theoretical length and total of the
out-​to-​out rebar lengths are used.
So, estimating the cost of steel based on theoretical design concept lengths can be mislead-
ing and it is unjust and unreasonable to not consider the reality of what it takes to execute
each particular custom design, and the corresponding correct price must be paid. At times a
design that seems on the surface to be less expensive on the basis of a quick look based on
a rapid calculation of the concept drawings theoretical bar lengths may turn out to be more
expensive. And as a final note, certain classic reinforcing designs with special grades of steel
may look good on paper but it may not be possible to guarantee a supply of the steel grade
in certain markets or parts of the country.
Reinforcing steel mills in certain sectors of the country may not roll certain grades of
steel, or if they do, they may have limited rollings, which are off-​timed with the precast
production schedules, or the mills may have large minimum rollings/​order quantities that
remove the possibility of the supply. For example, large tunnel jobs have thousands of tons
of steel but this is a relatively small quantity that a mill can churn out in a matter of hours.
Moreover, a local mill may have the ability to roll the special grades of steel but the mill
would want to roll out the tonnage of the job in one rolling and ship it immediately, thus
creating potential cash flow and storage issues, making it a show stopper for the precaster
536 Handbook of Precast Segmental Tunnel Lining Systems

and the owner and contractor who may not be in position to pay the whole steel for the job
upfront.
Careful coordination is required with the engineers, contractors, owners and precasters in
regard to the materials specified. For reinforcing steel, and as well as for all other materials
specified, care must be taken when finalizing the design to make sure that all such materials
are readily available in the specific market and that they will continue to be for the entire
duration of the project, which may span several years, with added attention to safeguard
against the specification of materials from unique suppliers that can fail to have or maintain
the capacity to deliver the materials at the pace required for the precast production (result-
ing in costly segment production shutdowns).

9.6 FIBER REINFORCED SEGMENT PRODUCTION –​ PRECASTER’S

CONSIDERATIONS
Fiber reinforced concrete is a relatively new practice in the reinforcing of tunnel liner seg-
ments. The first true efforts in implementing this new concrete reinforcement technology
occurred in the last 10 to 15 years in Europe (and Asia). It is a means of reinforcing that has
gained acceptance for utility tunnels but not so much transport tunnels due to code stipula-
tions. There is a tendency to use a hybrid solution with reinforcing and steel fiber in some
circumstances.
And when we say fiber reinforced tunnel liners, we must also consider the definition of the
fiber material since there are both polymer fibers and metallic fibers. Both types will behave
similarly in testing but there seems to be a reticence with designers to consider the polymer
fibers with certain applications.
There is a natural tendency for people to think that steel fibers are easy for a precaster to
deal with compared to classic rebar cage reinforcing but it is not so. Part of the bias prob-
ably starts from a comparison of the size and geometry difference between a fiber and a
piece of rebar (because one is small and light and rather dainty compared to the other one
that is heavy, unwieldy to handle and rusty…). Dealing with steel fibers has its whole new
set of challenges, and in a sense is a more technical and delicate thing to handle, and meter
into the mixer. Both reinforcing systems (rebar-​based and fiber-​based) have their advantages
and disadvantages in terms of design and both systems have their advantages and disadvan-
tages in terms of manufacturing.
The classic steel fibers are designed in Europe to standard EN 14889-​1, “Fibers for
Concrete”. Steel fiber definitions, specifications, and conformity in North America are to
standard ASTM A820 Standard Specification for Steel Fibers for Fiber-​Reinforced Concrete.
Chapter 5, goes into detail regarding fiber manufacturing specifications. When dealing with
fiber reinforced concrete, the subjects of interest to the precaster are the following:
As noted in Chapter 5, there are various types of steel fibers that are available in the
market. ACI references the ASTM A820/​A820M standard specification for steel fibers for
use in concrete. ASTM A820/​A820M provides classification for five general types of steel
fibers, based primarily on the product or process used in their manufacture: type I: cold-​
drawn wire; type II: cut sheet; type III: melt-​extracted; type IV: mill cut; and type V: modi-
fied cold-​drawn wire. However, not all of the ASTM types of steel fibers are appropriate for
use in this application.
The appropriate ASTM A820 fibers for use in precast segments are manufactured from
cold drawn wire, corresponding to a type-​I, or type-​V classification. The corresponding
appropriate European (BS-​EN) and International Standards (ISO) are the BS-​EN 14889-​1
 Fabrication of precast segmental tunnel linings 537

standard with a CE marking system 1 (steel fiber for structural use) or the ISO 13270 (class
A) as a minimum. The fibers in the appropriate dosage are homogenously mixed into the
concrete and then provide a network of individual pieces that are oriented in multiple direc-
tions and reinforce the entire cross-​section of the segment.
The following are the recommendations and considerations for steel fibers used in precast
segments:

• Fibers out of drawn wire, with a tensile strength of steel wire >1,800 MPa (261 ksi)
minimum
• Precast segments typically have compressive strengths well in excess of the min-
imum that is required for their structural design
• Typically, 35 to 40 MPa (5,000 to 6,000 psi)
• Because of high early strength requirements for production efficiency in the seg-
ment manufacturing plant, higher compressive strengths in the 65 to 70 MPa
(9,000 to 10,000 psi) are common
• To maintain ductility of the SFRC, higher tensile strength wire is recommended
• Hooked ends that have been optimized to ensure proper anchorage in the con-
crete matrix
• Fiber length: in the range of 50 mm (2”) to 60 mm (2.4”)
• Aspect ratio (length/​diameter): a minimum of 65, 80 is recommended
• Minimum fiber length: 2.5 times the maximum coarse aggregate size
• Fibers glued in clips to promote homogenous distribution and eliminate balling of
fibers since it takes more of a mixing effort to integrate loose fibers.
• Introduction of fibers into mix using a fully integrated automatic dosing system

9.6.1 Logistics and storage of steel fibers

Steel fibers are delivered in “big bags” of various sizes; for example, they can range from
800 kg to 1,000 kg (1,770 to 2,210 lbs) per bag (these can be custom sized). The big bags,
if shipped by sea container are really limited by their weight per container rather than by
volume. In a typical sea container, up to 20 bags can be shipped (depending on the weight
of the bags). They can also arrive by transport truck, covered or on a flatbed truck but well
covered. It is essential that the steel fibers remain dry otherwise they will rust, or if they are
glued fibers, the glue, which is water based and will dissolve with water will ruin the fibers.
Rusted fibers or fiber with glue that was deactivated are considered unusable and must be
discarded.

9.6.2 Reception of steel fibers

Upon their arrival at the precast plant, the big bags, typically in one metric ton sizes, must
be unloaded from trucks or sea containers. Care must be taken to avoid damaging the big
bags and they must be stockpiled in the plant or a storage building. The delivery of the fib-
ers must be timed with the rate of production, and this can be quite challenging if there are
delays in production due to the precaster or the contractor.
The steel fiber manufacturer fabrication schedule and supply chain are quite complex,
and it is quite difficult to suddenly switch deliveries on or off. Whatever was in transit or
in production at the time of the halting request is probably going to make its way to the
precast plant in the schedule that was originally agreed upon. This means that the precaster
538 Handbook of Precast Segmental Tunnel Lining Systems

may struggle to find room to store the big bags of steel fiber if precast production is slowed
down or halted. In any precast plant arrangement, the management of steel fiber, pertaining
to logistics and plant storage, is very challenging.
The steel fiber bags must be stored under roof in a dry environment. The logistics of pro-
viding this floor area in a precast plant must not be ignored.

9.6.3 Metering of steel fibers

The most critical part of steel fiber reinforced precast concrete is the metering (feeding)
of the steel fiber into the mixer. This is the most challenging element, and it is difficult for
many reasons, and it determines to a great extent the successful dispersion of fibers in the
fresh concrete. The complexity of metering in fiber is related to batch plant geometry and
arrangement: because every batch plant geometry is different, the successful fiber feeding
solution is always unique to the arrangement, and the solution is part science, part cookery
and part art. See Figures 9.22 to 9.26 for examples of fiber dosing systems that are used for
moving fiber/​metering fiber.
Getting it correct means that you will be able to batch consistently, rapidly, without any
“balling”. Please refer to Figures 9.27 to 9.30 for examples of balling. Balling occurs when
the fibers that are metered in clump-​up and mysteriously form balls that can range in size
between a golf ball to a baseball or softball.

9.6.4 Challenges of working with different types of steel fibers

There are challenges in the metering of the two different types of fibers available in the
market; loose fibers and glued fibers.

Glued steel fibers

Glued fibers come in clips where fibers are linked together with water soluble glue. There is
a time delayed reaction in the glue dissolving and thus the release of the fibers in each clip
comes at a different time, and thus the dispersion of fibers in the wet concrete mixture being

Figure 9.22 Fiber dosing and metering system –​ ­e xample 1.

Note: Typical helicoidal vibratory doser/​feeder for steel fibers.
 Fabrication of precast segmental tunnel linings 539

Figure 9.23 Fiber dosing and metering system –​ ­e xample 2.

Note: Plant steel fiber dosing system consisting of multiple fiber dosers linked by a belt conveyor.

Figure 9.24 Fiber dosing and metering system –​ ­e xample 3.

Note: Typical helicoidal vibratory doser/​feeder for steel fibers.

batched comes along quite well –​balling is generally rare with glued fibers (but there could
be instances of it, with some adjustments required).
As an example, with certain plant feeding arrangements, the metering in of glued fibers
could be done by dumping all the glued fibers in at once. With a dosage rate of 35.6 kg/​m3
(60 lbs/​CY) of fiber, for a 1.5 m3 (2 CY) concrete batch, one could easily think of dumping
55 kg (120 lbs) of fiber almost instantaneously (in a skip hoist setting), or at a very fast rate
per second or minute. If one has a 4.2 m3 (5.5 CY) batch of concrete, getting 160 kg (350
lbs) of fiber in rapidly poses a different challenge.
One way to dump all the glued fibers in at once, at times the easiest way, would be to
introduce the fibers with the pre-​weighed aggregates in the aggregate skip hoist. Introducing
540 Handbook of Precast Segmental Tunnel Lining Systems

Figure 9.25 Fiber dosing and metering system –​ ­e xample 4.

Note: Plant steel fiber dosing system consisting of multiple fiber dosers linked by a belt conveyor.

Figure 9.26 Fiber dosing and metering system –​ ­e xample 5.

Note: High-​c apacity steel fiber dosing and metering system used in a modern segment precast-
ing plant.
 Fabrication of precast segmental tunnel linings 541

Figure 9.27 Steel fiber balling –​ example.

Note: Actual example of the steel fiber “balling” occurring at the precast plant in the concrete
mixer equipment.

Figure 9.28 Steel fiber balling –​ during casting.

Note: Actual example of the steel fiber “balling” occurring, resulting fiber balls rising to the
surface.

the fibers into a skip hoist system is quite simple because one would do this almost at ground
level; and the fibers would travel up to the mixer with the sand and stone. The layout of the
plant would have to allow this, and this is not always the case!
Assuming the layout of the plant allows sending in the steel fiber in with the aggregates,
the solution would seem simple enough but perhaps too good to be true in certain instances
because of one thing –​some high-​tech concrete mixes and batching systems work with
moisture probes in the mixer and these moisture probes are microwave-​based and when the
steel fiber gets introduced with the aggregates, the mixer moisture probe, whose job it to
542 Handbook of Precast Segmental Tunnel Lining Systems

Figure 9.29 Steel fiber balling –​ large ball.

Note: Actual example of the steel fiber “balling” occurring at the precast plant in the concrete
mixer equipment.

Figure 9.30 Steel fiber balling –​ multi-​b alls.

Note: Actual examples of the steel fiber “balling” occurring at the precast plant in the concrete
mixer equipment.

measure what the water content of the combined constituents that are in the mixer drum
(so the batch software can subtract this amount of water free water from the water that it
intended to add to reach the design water content based on the water cement ratio) does not
read properly. The presence of steel fiber with the aggregate will falsify or affect the reading
of the moisture probe and create havoc. There are innovations in the sensitivity of moisture
meters to iron-​based interference.
 Fabrication of precast segmental tunnel linings 543

So, with certain batch plant arrangements it may not be possible or desirable to introduce
the steel fibers with the aggregates and this is a big challenge. What does this mean? It means
introducing the fibers, metering them in, after the aggregates are introduced. Metering the
fibers after the aggregates means that the fibers get introduced directly into the mixer that
is possibly 9.1 m (30 feet) off the ground with all sorts of equipment in the way. Conveyors
may be required.
The other disadvantage is that metering in fibers directly in the mixer drum, in a batch
system that uses a probe in the mixer, lengthens the batch mixing cycle, and whatever that
extra batch time number is, 30 seconds, 60 seconds, etc., in the grand scheme of things,
when one has a high volume job and the mixer needs to supply a certain number of batches
per hour, it matters! So, with glued fibers, the introduction of fibers is more a straight-​
forward matter, but there is quite a bit of logical thinking and system design involved and
it is not that simple.

Loose steel fibers

The metering of loose fibers is more challenging because there is also cookery and art involved
in finding a solution. Loose fibers are more prone to “balling” if introduced too rapidly as
measured in metering rates measured in pounds, or fraction of pounds, per second. What
this means is that the batch mixing cycle can get extended by several minutes, if only to
reduce exposure to fiber balling. Again, this is extremely penalizing if one has not taken this
into account because the hourly output of the mixer can be much less than you thought.
Of course, with the loose fibers, as it is with glued fibers, in a system that relies on a probe
device mounted in the mixer drum, one needs to meter in the fiber after the introduction of
the aggregates (and after the mixer moisture probe has taken its reading). This lengthens the
batch cycles as was explained in detail above.
The mixer drum in most arrangements will most probably be several dozens of feet off the
ground and will require a mechanical means of getting the fiber close to the mixer. When
the fiber gets close to the mixer, this is when the black magic starts, and an intricate meter-
ing solution must be developed to introduce the fiber into the running mixer at a specific
rate and with a specific sprinkling spread. This must be thought out, designed, tested and
implemented and at times through trial and error. There are many different styles of mixer
drums; planetary, twin-​shaft, horizontal shaft, turbin, and they each have their peculiarities
and characteristics to deal with.
Loose fibers are much more difficult to master and work with in terms of dispensing/​
metering than a glued fiber.

9.6.5 Dispersion of steel fibers in fresh concrete

The dispersion of fibers in the fresh concrete is the most important part of the operation.
Ensuring good dispersion is essential. We have touched on this in the previous paragraphs
and will elaborate more on this.
Apart from the visual cue of witnessing or not witnessing the balling phenomenon (keep-
ing in mind that balls of fiber tend to float up when the concrete is vibrating in the mold),
there is a test that is performed on the fresh concrete to measure the amount of fibers per
volume of concrete, i.e. the wash-​out test. This is an indicator test since it gives you a general
indicator of the way the fiber is distributed in your concrete more than the proof that your
concrete has the required number of pounds of fiber in it. This is misunderstood by many.
544 Handbook of Precast Segmental Tunnel Lining Systems

We have to remember that the batch plant controls the components that are added to
each batch cycle, and these components are added within set tolerances (example ±1% to
3%) and the batch cycle stops if the tolerances are off. So, if your batch plant is calibrated
following required industry specifications, each batch of concrete should have the proper
amount of fiber in it, within tolerances and if the mixer is kept in proper working order, the
fiber should be well distributed in each batch of concrete. You will have an indication of this
through the washout test statistics (and it is the statistics that one must pay attention to and
not the measured value of each test).
This test is typically performed on the volume of the concrete that is found in the air
testing pot, e.g. 0.25 ft3 of concrete (two air pots for 0.5 ft³). One can easily do the math
and figure out the pounds of fiber that should be present in the specific volume of the pot.
Hence the test consists of washing out the paste and aggregate from the fibers, baking the
fibers until all the moisture is gone and then weighing the fibers. Please refer to Figures 9.31
to 9.33.
One must understand that the dispersion of the fiber in the concrete shoveled into an air
pot will have some variance and typically the acceptable limits of fiber presence in air pot

Figure 9.31 The making of fiber concrete cylinders and beams.

Note: Example of equipment used to make the sample cylinders and beams.
 Fabrication of precast segmental tunnel linings 545

Figure 9.32 & 33 Fiber wash-​o ut test.

Note: A pre-​d etermined amount of concrete is washed-​o ut on a magnetic table to determine
the quantity of fibers in sample.

samples can vary by; minus 20% and plus 20%, from batch to batch but it is the monitoring
of the running averages that is of value.
There are now new technology alternatives to the wash-​out test, i.e. induction machines
that can calculate the amount of fiber that is in a wet or hard concrete sample.

9.6.6 Hardened concrete testing of fibers –​preparation of samples

The fresh concrete testing of fibers is, as discussed in the previous paragraph, the washout
test, but the concrete technicians must also prepare the samples that will be used for the
hardened concrete testing –​the beam tests.
First, the beam molds must be cleaned, oiled and assembled before getting them ready
for filling. The testing beams must be prepared with the concrete in its fresh plastic state.
To effectively make concrete beams, once must use a proper high-​frequency vibrating table
with ample power and take care to fill the beams with a generous amount of concrete; a
little more than the molds can hold.
546 Handbook of Precast Segmental Tunnel Lining Systems

Once filled, the lab technician must use judgment to vibrate them with the proper amount
of time to guarantee proper consolidation, and to avoid entering a state of segregation: if
the large aggregates drop, they can affect the orientation and position of some of the fibers
and the testing results will be skewed.
Special robust vibration tables with high-​frequency are a must. Not just any vibrating
table will consolidate the concrete properly and keep the aggregates and fibers sus-
pended in their natural initial state. We do not want segregation; aggregates and fibers
to drop, or to have fibers change their orientation (which could possibly help or hinder
the results).
As with all concrete vibration activities, good judgment and good crews are required to
do it correctly. It is important to ensure consistency in the making of samples. The previ-
ously mentioned washout test is really more a measure of the consistency of the making of
the samples, and not a measure of how much fiber is in the concrete batch. There is a specific
protocol for the preparation of beam samples in the American (ASTM) and European (EN)
specifications, but there are some steps that are not detailed and there are efforts underway
to provide more guidance in the specifications.

9.6.7 Curing of concrete fiber beams

Concrete beams can follow the primary cure and then get sent to the standard water cure
with the test cylinders or can be left to mature normally and then be sent to water cure
with cylinders (which makes more sense because the beam testing results are sensitive to the
proper curing of the beams).
See Figure 9.34 below of the tanks necessary to cure beams. Note that the water
level should be kept so that it matches as much as possible the top level of the beams
and there is some adjusting to do as one loads and empties the tanks because the water
displacement.

Figure 9.34 Sample beam and cylinder curing tanks.

Note: Concrete sample curing tanks at the plant materials laboratory.
 Fabrication of precast segmental tunnel linings 547

9.6.8 Hardened concrete testing of fibers

The hardened concrete testing of beams occurs with a very sophisticated beam breaking
machine. There are two different formal tests available for quality control testing of speci-
mens of steel fiber reinforced concrete used for precast segmental tunnel liners.

• American test
• ASTM C1609, Standard Test Method for Flexural Performance of Fiber-​Reinforced
Concrete (Using Beam with Third-​Point Loading)
• European test
• EN 14651, Test Method for Metallic Fibre Concrete–​ Measuring the Flexural
Tensile Strength (Limit of Proportionality (LOP), Residual)

They are both equivalent tests within 34.5 to 41.3 MPa (5,000 to 6,000 psi) concrete range,
but when concrete strengths reach 58.6 to 82.7 MPa (8,500 to 12,000 psi), the European
test shows more consistent results. This matter is explored in detail in a previous chapter
of this book.
It is important to note that there are some new tests that are being introduced and proofed
that would replace or complement the classic beam tests such as the EN 14651 test and the
ASTM C1609 test. One such test, as an example, is the Barcelona test that is used to evaluate
FRC post-​cracking behavior. The Barcelona test is a double-​punch test that can be seen as
offering important advantages over other types of tests. It can be relatively easily carried-​out
on regular compressive testing equipment, and it uses light and easy to handle specimens. It
is said that its use reduces the dispersion of results, which is welcome. One idea is to use it
to complement existing tests. There is also a notched beam test being developed by ASTM.

Preparation of fiber beam samples

The beam samples must be prepared before being loaded in the beam machine as shown
in Figures 9.35 and 9.36. They must be at times ground down, if the beam molds have left
certain surfaces with protrusions that make the plane surfaces uneven. With the European
test (EN 14651), a saw cut must be made in the beam.

Testing of fiber beam samples

The proper loading of the beam machine is critical. Details matter. Improperly installing
a beam in a beam break machine can lead to lower results than reality, result in a faulty
or a failed test. The beam machine does all the work and calculating but if the beam was
installed improperly on the beam machine, it may produce a test data outlier. The techni-
cian must take care to load the beam into the fixture properly and ensure that the fixture is
tightened up properly. The testing time for a beam sample can last from 30 minutes to one
hour. Sampling from the beam testing machine takes time with a sampling rate of ±10 per
second; after one hour, one is left with 10,000 readings. If one prints-​out the raw data for
one test, it will use over 200 sheets of 8.5 × 11-​inch paper (a significant amount of data is
collected). See sample beam test report in Figure 9.37.

9.6.9 Concrete durability testing considerations with steel fiber

reinforced concrete
One must consider that there are many durability tests that measure chlorides or measure-
ments of electrical current and these types of tests simply cannot be considered on samples
548 Handbook of Precast Segmental Tunnel Lining Systems

Figure 9.35 Example of beam breaking machine for ASTM C1609 test protocol.
(a) This machine is set-​u p to perform the ASTM 1609 beam bending test.
(b) General arrangement and specific beam set-​u p apparatus (yoke and transducer).
 Fabrication of precast segmental tunnel linings 549

Figure 9.36 Example of beam breaking machine for EN 14651 test protocol.

(a) This machine is set-​u p to perform the EN 14651 beam bending test.
(b) General arrangement and specific beam set-​u p apparatus with notched beam.
(c) Clip-​o n gauge mounted to the underside of the test beam to measure the crack opening.
550 Handbook of Precast Segmental Tunnel Lining Systems

Figure 9.37 Example of beam test report and parameters.

Note: Example of a typical beam testing report that defines the test sample parameters and
procedures. Note the graphic representation of the stress and strain values.

containing steel fiber. So, one can proof the mix during pre-​production with the absence
of fibers, but during production, it will not be possible to test the fibered samples, for tests
measuring, for instance, chloride ion diffusion (one would have to use a sacrificial batch
that lacked fiber and there are costs related to this but there are significant costs to sacrificial
batches in terms of wasted concrete and the disposal of wasted concrete, and the fact
that the batching of sacrificial concrete may affect production time schedules). The other
solution is to develop a system to extract the fibers from the wet concrete (which comes at
a significant cost depending on the frequency) but if a few fibers are left behind, the tests
could be compromised.

9.6.10 Hybrid reinforcing considerations

There is a tendency for precast segmental tunnel lining designers to look at hybrid designs
where a light steel reinforcing cage is paired with steel fiber reinforcing. However, the con-
tribution of the fibers and the steel rebar-​type reinforcing are generally not considered in
aggregate for the design in the resistance against the effects of the acting forces, nor in con-
sideration of enhanced durability and hopefully this will change in the future.
There are some engineering advantages to this type of design that has gained in popularity
outside of North America. From a precaster’s perspective and project economic analysis, it
is best to consider a single type of reinforcing because when one doubles up the reinforcing
types, one also doubles up the plant set-​up areas for assembly, storage, distribution and
metering, and the sum of the operations for managing two types of reinforcing is costly and
can be extremely costly.
A light steel cage is sometimes only different in the amount of steel loaded in the jigs but
all the operations and logistics that lead up to making it and then handling it and loading it
in the molds is practically the same in manhour requirements and encumbrances.
 Fabrication of precast segmental tunnel linings 551

In the end, a hybrid solution is more than the sum of the parts in terms of organization,
logistics, operations and testing –​it is at times three or four times the work and effort; i.e.
higher costs to implement at the project level.

9.6.11 Steel fiber reinforced concrete –​conclusions

There is a tendency for engineers to assume that working with fiber is much easier for the
precaster. As mentioned before, there are advantages and disadvantages to each system.
Here are a few considerations to ponder:

• Fiber metering equipment, vibratory feeders, conveyors have up front capital costs
and can have significant costs of installation.
• There are costs related to the electronic controls that are required to link the metering
systems to the batch plant software –​so that one can control with assurance the true
amount of fiber that is fed in the mix.
• There are metering system calibration costs (that occur on an established frequency).
• Fiber metering systems must be fed throughout the shift by manpower, at times with
specialized equipment.
• Fiber must be unloaded from delivery trucks, at times with special handling equip-
ment, on a regular frequency.
• Fiber must be handled and stored protected from the elements. This means a roofed
area for storage and there are costs to this.
• Fiber metering systems are quite robust, but they still need to be constantly monitored
because they can be prone to mechanical issues, or jams if not properly designed.
• Fiber metering systems need to be maintained and repaired occasionally.
• If fiber metering systems go down, the production stops.
• One must be aware that batching time can be increased by fractions of a minute to
several minutes with these processes.
• There is a labor-​intensive aspect to the testing of fiber concrete, there is expensive
testing equipment, and there are large expenses related to shipping and testing with
an independent lab, or even with in-​house testing.
• Hence it is important to understand that fiber reinforced concrete is not a cakewalk
for the precaster –​there are many considerations and costs to deal with, depending
on the numerous factors outlined above.

9.7 CONCRETE BATCHING, MIXING AND CURING

It can be argued that segments have two major quality markers that define them; high dur-
ability and tight geometric tolerances. The concrete is the bones of the product and proper
concrete batching/​mixing is required to achieve excellence in the durability camp.

9.7.1 Concrete batching and mixing

One of the most important aspects of precast segment manufacturing is to develop a manu-
facturing process that provides a consistent quality of concrete.
There are several ways of providing concrete to a segment casting plant, the standard practice
is to use a dedicated batch plant, which offers precision, accuracy and traceability (documen-
tation). Using ready-​mix concrete from an existing plant works under certain circumstances,
552 Handbook of Precast Segmental Tunnel Lining Systems

typically when the batching operation is close to the casting molds. It is standard industry
knowledge that the concrete mix designs used for making segments are typically high-​strength
high-​tech concretes that are “hot” meaning they are rich in cements and admixtures and can
react relatively rapidly –​not necessarily exhibiting an abnormal early set but they may lose
workability rapidly (so timing in getting it to the molds quickly matters).
A modern dedicated batch plant will have several silos of cementitious materials such
as cement, slag, fly ash and silica fume and several tanks of admixtures. It will need to be
designed for the output required for continuous, high-​volume production in varying wea-
ther conditions –​hence attention to the aggregate storage (stone and sand bins/​silos) and
aggregate handling systems (conveyors) is important. The plant will most probably require
hot water heaters, chillers, and possibly aggregate heating systems to meet certain climatic
conditions. The batch plant needs to be sized to the project, meaning that the mixer can
produce sufficient concrete to fill one mold completely in one or two cycles, otherwise
manufacturing efficiencies are challenged.
While modern batch plants rely on moisture meters for aggregates and possibly for the
mixer, there is a substantial amount of technology in a modern casting operation. Just the
same, there are three important factors that can impact plant performance:

• Good housekeeping (clean mixer drum with properly adjusted paddles)

• Proper calibration
• Knowledgeable batch plant operator, since many decisions and adjustments in the
casting operation are influenced by visual observations

9.7.2 Concrete curing –​ precaster’s considerations

Proper concrete curing is a very important element to ensuring concrete durability and it is
a subject that precast tunnel liner engineering specifications are often vague about by virtue
that they often include multiple different curing specifications. With multiple specifications
on the same subject, it could be argued that some curing specifications conflict with each
other, or there can be arguments on the precedence of different specifications. With multiple
specifications called for on the same subject, there is also a potential for the engineers to
cherry-​pick parts of each of the different specifications and ask for the simultaneous appli-
cation of the handpicked elements and this can be very challenging for the precaster.
In most operations where the molds are cycled at least twice per day, there should be
relatively clear directives with some flexibility regarding, depending on the subject, what is
permitted or what is not permitted. Curing protocols must have some flexibility, but at the
same time have some rigidity and formalized structure. Some of the subjects that might need
directives are the following:
Primary curing period (approximately the first 6 hours after casting):

• Setting-​
out the allowance of ambient steam temperatures during the concrete
preset period
• The maximum internal temperature that is allowed in the concrete itself

Secondary curing period (the bid specifications should have clear directives regarding
this period)

• Is there a secondary curing required? If so, how long?

• Is there a method that is required?
 Fabrication of precast segmental tunnel linings 553

• If there is a requirement to reach 70% f’c before yarding the product

• If there are a minimum amount of degree-​hours that the concrete must have before
yarding (this may be ghostwriting the spec for steam curing as a secondary curing
process)
• If a curing compound application is sufficient
• If a curing compound is sufficient, what is the surface coverage required by the curing
compound

Clarity is required to even the playing field. Confusion in the specification language is not
easy for the precaster bidding work, all of the competition must know what is going to be
required for the job and be held to the same standards. There is a world of financial, oper-
ational and durability performance differences between applying a curing product to seg-
ments versus curing segment-​rings in steam in computer controlled vapor kiln enclosures
(there is the cost of the vapor generation systems, the gas costs, the kiln costs, the real estate
costs, the costs of handing, among many more, versus the cost of a Hudson garden sprayer
and relatively inexpensive curing compound liquids –​the difference being multiple hundreds
of thousands of dollars).

9.8 SEGMENT CONNECTIONS, ACCESSORIES AND GASKET

INSTALLATION
Segments need the proper connections and accessories to facilitate proper handling and
assembly, plus to facilitate waterproofing requirements for long-​term service life. The seg-
ment manufacturer must incorporate many products into the segment casting operation to
meet specific project requirements.

9.8.1 Connection systems –​ radial joints

The standard industry practice is to use steel bolts for connecting the radial joints or the
segment-​to-​segment faces in the segmental ring. To facilitate segment alignment, plastic
guiding rods are installed on the radial joint face to assist in providing joint shear resistance
(information provided by the segmental ring design engineer) and to facilitate a more effi-
cient ring build and quality of assembly. Bolt diameter, grade and coating are determined
by the project’s specific engineered requirements. Boltless radial joint connectors are being
evaluated and developed by various industry manufacturers.

9.8.2 Connection systems –​ circumferential joints

The past practice was to also install steel bolts for the circumferential joint connections, but
the industry has advanced to the more common practice of using dowel connection systems
for circumferential joint or the ring-​to-​ring faces in the segmental ring. Dowel systems con-
sist of two or three female embeds cast into the circumferential face of each segment into
which a dowel is inserted in the female embeds in the tunnel then the TBM shove jacks are
used to push the segments together locking the dowel system into place. Dowel systems are
designed to provide proper ring-​to-​ring alignment, improve ring erection time, and to meet
project specific shear and pull-​out requirements.
554 Handbook of Precast Segmental Tunnel Lining Systems

9.8.3 Grout/​l ifting sockets

To facilitate a proper ring build, a cementitious grout is placed behind the assembled ring
to provide ring circularity and long-​term lining stability. Grout ports for placing the grout
behind the segments can be cast into each segment, and these consist of a plastic sleeve
through which the grout is pumped and later sealed with a plastic screw cap. This same
plastic sleeve can be engineered to for the use of centrally located steel TBM erector lifting
balls, for the mechanical handling of the segments, if the contractor decides to use this seg-
ment handling system. Blind grout-​lift inserts should be used as much as possible since these
inserts can be used for grouting (after drilled and perforating hole through the segment)
and thus it limits the opportunities for potential water egress when not used for grouting
operations.

9.8.4 Waterproofing gasket seals

Every project owner wants a dry tunnel. To accomplish this, continuous “door frame”
gasket seals are typically manufactured from EPDM rubber and are installed at the precast
segment manufacturing plant. When installed in the tunnel as an integral part of the lining,
the gaskets will prevent ground water from infiltrating the tunnel, and in some cases the
fluid being transported within the tunnel from exfiltrating to the surrounding strata. There
are two primary types of waterproofing seals that are installed at a precast segment manu-
facturing facility.

9.8.5 Anchored gaskets
Anchored gaskets have “feet” that extend into the cast concrete of the segment, and these
feet anchor the gasket to the final segment without any secondary gluing operation, that was
the past practice. Anchored gaskets provide safer working conditions in the casting plant
as typical gasket adhesives contain ingredients that give off volatile gases during the curing
process, which require the work area to be properly ventilated for worker safety. Anchored
gaskets are attached with care to a groove machined into the steel mold’s circumferential
and radial walls, after oiling of the mold. See Figure 9.38 that illustrates the anchored
gasket installation work in progress. The sequence of installing the gasket must be followed
for consistent results.

9.8.6 Glued gaskets
Glued gaskets are installed using a contact-​cement type adhesive system that connects the
EPDM rubber gasket to the gasket groove that is cast into the segment. This operation is
typically performed after the segment has been removed from the initial curing process as
shown below in Figure 9.39. The preferred industry system and more secure method is to
apply the gasket glue to the groove area only and then use a gasket press to apply pressure
for a few minutes. This process mimics the process of applying glue to two surfaces, the
gasket and the gasket groove, and then tapping the gasket into the gasket groove. There is
less human skill and judgment required when using the gasket press and the results are typ-
ically better quality than hand application methods.
 Fabrication of precast segmental tunnel linings 555

Figure 9.38 Anchored gasket installation.

Note: There is a specific sequence for clipping in key sections of the gaskets before proceeding
to clipping in others.

Figure 9.39 Glued segment gaskets –​ gasket gluing press station.

Note: Example of a modern, semi-​a utomated gasket gluing station within a segment precast
plant. In this view, the following equipment is in operation; segment rotating device (left), pneu-
matic gluing frame (center), and the segment handling clamp (right).
556 Handbook of Precast Segmental Tunnel Lining Systems

9.9 SEGMENT DEMOLDING, HANDLING AND STORAGE –​PRECASTER’S

GENERAL OVERVIEW
Demolding needs to be done carefully, this operation is done when the concrete has not yet
achieved its full 28-​day compressive strength. The industry has developed many procedures
and designed equipment that minimizes segment damage from the time of demolding from
the casting form to handling by the TBM erector arm in the tunnel. A proper quality control
procedure details these various phases and procedures.
The segment handling equipment used, at all phases of operations, needs to be designed
to prevent putting any strain on the segment, as to prevent cracking prior to assembly in
the tunnel.
The location of the dunnage in the segment stacks as shown below in Figure 9.40 needs
to be carefully studied and it must be positioned with precision and accuracy and within
defined tolerances. See as an example the following stacking diagram that defines a par-
ticular phase of segment storage.
Considerable detail needs to be paid to how the segments will be stored outside, in
the storage yard. Any storage area needs a stabilized base to prevent the stacked seg-
ments from listing during storage, which would thus potentially cause edge damage to
the segments.
There needs to be adequate outside land space to facilitate any delays in tunnel segment
consumption by the tunnel contractor, a common industry occurrence.

Figure 9.40 Segment stacking arrangement with dunnage placement.

Note: Example of a stacking arrangement for stacking and handling a full ring –​ or a ring with
two picks.
 Fabrication of precast segmental tunnel linings 557

9.10 SEGMENT DIMENSIONAL MEASUREMENTS AND

DOCUMENTATION –​ PRECASTER’S VIEWPOINT
There are several ways to measure segments –​but the best way is to control the dimensions
of the molds to guarantee the segment tolerances. This is addressed in detail in Chapter 8,
but for clarity, we add to the subject because there are today too many misunderstood ways
to interpret mold and segment measurements and it is creating havoc for everyone (precast-
ers, contractors and engineers).
As mentioned in Chapter 8, there is new international laser scanning technology that can
measure both the steel molds prior to casting, and the segments at the time of demolding.
Several elements must be understood with regard to the high-​tech 3D measurement systems
for molds and segments:

• There is a cost to owning and/​or operating such systems.

• There may be interference with production flow, with timing of measurements and
length of time required to take measurements.
• There may be incompatibilities with the available plant floor layout space available in
certain areas of country.
• There are advantages and disadvantages to each system and although the new tech-
nology is interesting and can be very effective when interpreted properly, the large
amount of data produced by these systems, data that is subject to interpretation, can,
in the wrong hands, or incompetent on-​the-​subject hands, create unnecessary polem-
ics that paralyze projects.
• Tracking and trending of the molds (and the mold components that are subject to
be off tolerance) may not be easy to follow with the generous but inappropriate-​
for-​the-​subject data generated by laser scanning systems. It is important for the
precaster to have a feel of how certain specifically located mold tolerances may
be drifting from the normal in time, and the old-​school proven methods with pre-
cise and accurate instruments whose measurements have compensations for atmos-
pheric conditions have value here. In the right hands, and without interference
from people who know little about this complex subject, the new technology is
viable.
• The 3D measuring system provides a best fit plane of the face of the sidewall measured.
It averages the locations of the spots where data is acquired and then the software
creates a surface. This “as found” surface is judged against the theoretical surface of
the mold sidewall or endwall. It is important to understand this.
• The old-​school measurements with a vernier can be looked at locally at the different
measuring points and one gets an idea of how each specific measurement location
varies in time and the average of the measurements gives one an idea of the planarity
of the sidewalls from theoretical.

Segment dimensional measurement comes with a cost and the owner/​design engineer must
determine in advance of the bidding of the project if this level of quality control is required
and if the owner willing to pay this additional cost. This may be required for a 14 m (46-​
foot) diameter transportation tunnel but not required for a 3.0 m (10-​foot) diameter sewer
tunnel.
558 Handbook of Precast Segmental Tunnel Lining Systems

9.10.1 The basic fool-​proof way to control segment tolerances –​

through control of molds
The reality is that at minimum, in the present day, we have the means to merge traditional
analogue and new digital methods of measurements to ensure production of excellent prod-
ucts. The control of the segments is a combination of using the new 3D technology and
the traditional methods of measurement with both the molds and segments and thus when
rigorous measurement protocols are followed, basic logic can be used to deduce the state of
tolerance of the segments. The proofing process is to establish a “baseline mold tolerance
set” that will guarantee a complete segment geometry tolerance.

Establishing a “baseline mold tolerance set” (collection of specific mold tolerances)

This is all based on the premise that nowadays, the molds are designed in 3D by the latest
computer aided design software with the mold design generated directly off the 3D geomet-
rical design of the segment: this method of designing/​fabricating molds revolutionized the
manufacturing and removed the potential for errors that existed with the traditional 2D
way of designing. Here are the steps required to establish a baseline:

Building the first set of molds

• The mold manufacturer builds the first set of a series of molds using their fabrication
jigs and templates.
• These are carefully checked against segment design dimensions and allowable
tolerances.

Proofing the mold fabrication templates and jigs

• The geometry of the first set of molds is checked with the 3D scanning technology to
certify that the tools used to build the molds (e.g. manufacturing fixtures and jigs) are
within tolerance with the theoretical geometry and in particular, to confirm that the
manufacturing fixtures and jigs meet the theoretical dimensions and geometry within
tolerance. Tolerances between the actual mold surfaces and the 3D generated surfaces
must be compared and respected. (See Figure 9.41.)

Casting the first set of molds

• Next, the first set of molds are cast at the mold fabricator’s shop.

Proofing the molds (and by correlation with the segments)

• Each one of the segments cast are now measured by the 3D scanning technology to see
if the mold has produced a segment that is within the geometrical tolerances allowed.
If this is the case, you now are assured that the mold (the resultant of all its tolerances)
yields a segment that is in geometric tolerance (the aggregate of all its tolerances).
• At this point, the “baseline mold tolerance set” is established. See Figures 9.42 to 9.46
for examples of the mold-​segment tolerance comparisons.

Only once the mold manufacturer is convinced by the previous 3D measurement testing
that the mold fabrication templates and jigs are of the sufficient precision and accuracy to
 Fabrication of precast segmental tunnel linings 559

Figure 9.41 Segment mold under fabrication with 3D measurement machine.

Note: Large segment mold under shop fabrication prepared for check (conformity) measurement
checking by 3D machine. The Faro unit is mounted in the middle of the steel segment mold and
can measure the fabrication to within tolerance settings (pre-​d etermined, for example ±0.5 mm
(picture from CBE GROUP).

build molds that will in turn produce segments that are in tolerance, the mold manufacturer
continues the assembly of the other sets of molds in the production sequence. If discrepan-
cies were seen, adjustments are made to the first set of molds that was fabricated and these
adjustments are in turn applied to the following sets of molds being assembled. Each set of
molds gets checked with the 3D instruments to certify each mold.
By this process of interactive checking and logical deduction, we can, at this point, with
great confidence measure molds with either traditional methods or 3D methods with the
reassurance that the segments produced from these molds will be in tolerance. Hence, we
can confidently say, using this protocol, that the geometry checks of the molds equate the
geometry checks of the segments during production.
The documentation related to these methods of measurement comes in two forms:

• In digital raw data form and summary tables for the 3D laser scanning methods and
• In manual check sheet data that is converted to excel tables whereby statistics can be
followed and analyzed.

Related notes on controlling mold and segment tolerances

• Hence, once we have established a “baseline mold tolerance set”, one can use 3D
measurement systems for the validation of segment tolerances, or the traditional meas-
urement systems (calibrated vernier instruments), both types of systems to measure
only the critical measurements determining geometrical tolerances.
 newgenrtpdf
560
Handbook of Precast Segmental Tunnel Lining Systems
Figure 9.42 Segment mold-​s egment tolerance comparisons –​ widths.
Note: Examples of segment mold width tolerance comparison reports and corresponding graphic images listing the actual measurements and
apparent deviations.
 newgenrtpdf
Fabrication of precast segmental tunnel linings
Figure 9.43 Segment mold-​s egment tolerance comparisons –​ half diagonals.
Note: Examples of segment mold half diagonal tolerance comparison reports and corresponding graphic images listing the actual measurements
and apparent deviations.

561
 newgenrtpdf
562
Handbook of Precast Segmental Tunnel Lining Systems
Figure 9.44 Segment mold-​s egment tolerance comparisons –​ fatness.
Note: Examples of segment mold “flatness” tolerance comparison reports and corresponding graphic images listing the actual measurements
and apparent deviations.
 newgenrtpdf
Fabrication of precast segmental tunnel linings
Figure 9.45 Segment mold-​s egment tolerance comparisons –​ thickness.
Note: Examples of segment mold thickness tolerance comparison reports and corresponding graphic images listing the actual measurements
and apparent deviations.

563
 newgenrtpdf
564
Handbook of Precast Segmental Tunnel Lining Systems
Figure 9.46 Segment mold-​s egment tolerance comparisons –​ plane orientation.
Note: Examples of segment mold plane orientation tolerance comparison reports and corresponding graphic images listing the actual measure-
ments and apparent deviations.
 Fabrication of precast segmental tunnel linings 565

• The critical measurements are the canary-​ in-​

the-​coal-​
mine measurements that
by themselves are harbingers of the total geometrical tolerances of the mold –​an
example being the 10 mold width measurements–​if these are within say ±0.5 mm
(0.02 inch), you can rest assured that the radial and circumferential joint orientations
are in tolerance.
• There are many geometric tolerances of the molds that do not vary with time and use
of molds –​such as the radii of the intrados, extrados, and circumferential joints. The
non-​critical geometrical tolerances are verified early-​on with the molds and segments
with the factory manufacturing 3D scans. These are not subsequently verified during
production unless a catastrophic event occurred with a mold.
• If there are worrying trends in the mold width tolerances and tolerance issues that are
not related to mold nesting pocket cleanliness, the precaster verifies the orientation of
the radial and circumferential joints with the templates and jigs that the mold manu-
facturer has provided (the templates and jigs that were used to assemble the molds).
The results of the measurements will determine actions to take. With certain mold
manufacturers, checks are not made with templates and jigs.
• A mold that is severely out of tolerance will be difficult to open and this is a telltale
sign that something was damaged. Damaged molds that are out of tolerance must be
brought back to spec by a technician from the mold manufacturer (or someone spe-
cifically trained).
• For those who have only faith in the digital-​mechanical systems, remember that
the machines that are used to make these digital mechanical machines are analogue
proven old-​school vernier or similar built machines, it can be argued that the new
technology is still at the origin built with the old technology! And the first machine
that was used to make machines, was made by a human hand with a human eye.

9.11 TEST AND DEMONSTRATION RING –​ PRECASTER’S

CONSIDERATIONS
The test ring is a demonstration ring/​control ring that is built vertically, without gaskets, as
they would be impossible to compress when mating segments, due to the high reaction force
from the gaskets. Test rings are typically comprised of one to three rings stacked vertically
on top of each other at the casting plant. Two types of test rings exist in current practice for
precast segmental tunnel liners.

• One that is built at the mold manufacturer’s plant as a control of the first ring set
of production molds, before launching into production of the complete set of molds
required for the project.
• This is the most useful type and is usually only one ring high.
• It may be measured with high precision with a FARO-​type 3D surveying instru-
ment and provides useful comparative geometric data between the molds and the
final product, the segments.
• The other type of test ring is the one usually called for at the project’s precasting plant,
as the first production rings come out of production.
• This test ring (or rings if they are stacked on top of each other) are mostly cere-
monial in nature if the test ring at the mold manufacturing plant already conveyed
critical geometric tolerance information.
566 Handbook of Precast Segmental Tunnel Lining Systems

Figure 9.47 Test ring at precaster’s yard –​ side view.

Note: Example of test ring set up at precaster’s yard; elevation view of segments mounted
on jacks.

Figure 9.48 Test ring at precaster’s yard –​ view from above.

Note: Example of test ring set up at precaster’s yard; top view of segments mounted on jacks.

Please refer to Figures 9.47 and 9.48 for photos of a test ring set that is set-​up at the pre­­
caster’s yard.
Test rings, being built vertically, especially multiple level rings, are extremely challenging
to build and are prone to move and creep with time because of the awkward forces that
can deform certain connectors, and because the circumferential joint connectors are ground
down to allow for dismantling (these ground-​down connectors have slop in them). The ver-
tical installation arrangement requires forcing segments unnaturally into position, and this
may force the plastic inserts. There is also the “slop” (loose fitting) in the tapered bolt holes
to contend with which permits movement between the segments. Finally, the temperature
will have an effect on the ring geometry; e.g. if the sun is beating on one side of the ring and
the other side is cool, there will be differences in measurements depending on the time of
day they are taken. However, the roundness and fit of the segments can be witnessed and
this is the important aspect.
Historically it made sense to build a test ring at the precasting plant to verify if fit was
good and if the bolt pockets or inserts had been located improperly in the molds. Please
recall that 30 years ago, the mold shop drawings were created by draftspersons and engin-
eers using a pencil and paper and drawn in two dimensions. One can brazenly say that a
segment-​ring is just a cylinder, but it is a challenging and frustrating exercise to study a ring
 Fabrication of precast segmental tunnel linings 567

and its fine and subtle segment geometrical details and to manually draw it up because every
segment looks alike, it was easy to make a mistake. The test rings generally caught these
mistakes and subtle misalignments.
Today we have 3D computer-​aided design software where the segment drawings are
made by the modeling of a ring and its segments directly with the 3D software and then the
molds and their detailed parts are designed off these digital 3D segment models The risk of
bolt pockets and elements being located and/​or assembled in the wrong location is almost
non-​existent. Hence, nowadays, the checking of a test ring is different than the original
intention of a test ring. Whereas originally and using the technology available at the time,
the test ring was used to verify general fit-​up, today, a test ring is used to delve into the fine
details of dimensional tolerances that can be measured with theoretical surface models that
are generated with the available computing power.
The other issue that a test ring or set of two or three test rings does not address is the
different permutations of say the situation with a production using seven sets of molds with
seven segments per ring, and possibly three circumferential bolt/​dowel locations per seg-
ment. It would be impractical to interchange the seven sets of segments from seven sets of
molds and clock them in each possible position that the rings could be mated in. This is why
the test ring at the mold manufacturer is so important.

9.12 SEGMENT TOLERANCES –​ PRECASTER’S CONSIDERATIONS

There are segment tolerances and there are mold tolerances, as we explored in detail earlier.
They are different, as the molds must start off with tighter tolerances to make up for losses
of tolerance due to the mechanical closing of the mold sidewalls and how they react to
vibration and loading. Hence the allowable segment tolerances are always larger than the
allowable mold tolerances –​and how they vary depends on the size of the segments: bigger
segments/​bigger molds require larger tolerances and the engineers, precasters, and mold
manufacturers must at times all come to a mutual agreement as to what is possible, regard-
less of what the various international standard guidelines recommend.
Naturally, the molds will move slightly under loading (concrete and vibration), and certain
dimensions will open up, while certain dimensions will close-​up, especially with long side-
walls that may micro buckle (so slightly). During production, certain mold measurements
may vary in tolerance but must stay within the upper and lower limits set or agreed upon, but
on average they have been noted to hover around or slightly higher than the factory settings.
There are many different nomenclatures for mold and segment tolerances that the industry
is trying to standardize, but different authorities have their own ways of naming things, and
it is difficult to relate between them. Or some of the authorities use tolerance nomenclatures
that mean something slightly different to other authorities, or to the mold manufacturers.
The industry owes Nasri and Bakhshi a debt for their effort to take all the world specifica-
tions and compare them in a table, as can be seen in the ACI 533.5R-​20, “Guide for Precast
Concrete Tunnel Segments, Table 12.1b”. However, there is critical engineering judgment
that needs to be exercised in interpreting the table of this guide, as there are nuances in the
way certain authorities define certain technical terms. This can lead to arguments between
Engineers, Contractors as well as precasters and mold manufacturers. Presently, some of the
authorities that have listed tolerances are the following.

• British Tunnelling Society

• BTS Specification for Tunnelling 3rd Edition 204.4.1
568 Handbook of Precast Segmental Tunnel Lining Systems

• BTS ICE 2004, Tunnel Lining Design Guide

• German Railway Standard for Rail Tunnels
• Ril 853-​2012
• German Transport Ministry Standard for Road Tunnels
• ZTV-​ING-​2007
• Deutscher Ausschuss für Unterirdisches Bauen e. V. (DAUB)
• Empfehlungen für den Entwurf, die Herstellung und den Einbau von
Tübbingringen
• Lining Segment Design: Recommendations for the Design, Production, and
Installation of Segmental Rings (2013)

At times, engineering specifications refer to various international specifications simultan-

eously and call for certain dimension tolerances that are confusing to the precaster because
he/​she is trying to relate these to the mold manufacturer’s achievable tolerances (what the
mold manufacturers can achieve with the prescribed ring geometry and diameter).
The tolerances that the mold manufacturer uses to manufacture the molds are the ones
that we should prioritize because these are the ones that the precaster can control and reset
and modify if the mold goes out of tolerance. An example of the tolerances that a mold
manufacturer tracks are shown in Figure 9.49. The mold manufacturing/​assembly toler­
ances are dear to the precaster because they are the ones that can be tracked and studied for
trends, and more importantly, these are the ones that the precaster can quickly verify to see
if the mold is severely off tolerance.
The tolerance of the finished extrados surface is a subject that must be agreed upon based
on what is reasonably achievable in consideration of the mold geometry (size). Expecting
±1.5 mm (0.06 inches) on a finished surface is not reasonable. The extrados should be con-
trolled with a maximum and minimum depression or protuberance allowed, coupled with
the concept of average thickness across the section.
In certain cases, it is fair to consider three levels of tolerances:

• Very tight mold manufacturing tolerances (at the apex of manufacturing ability –​to
provide a buffer)
• Tight mold production setting tolerances
• Segment production tolerances

Please refer to Chapter 8, for a comprehensive comparison of different specifications about

mold and segment tolerances.
Note that some of the different standard specifications consider the concept of evenness
(flatness) of radial and circumferential joints and this choice of words can be misleading
because a single value is normally provided to describe this tolerance and in actuality, it
involves three or four different measurements.

• Two surface orientation measurements given in linear measurement offsets and angles
(which are incompatible with large mold dimensions). See Figure 9.50 below.
• True flatness –​planarity of a surface measurements. See diagram in Figure 9.51 below.

The data listed in the table, shown in Figure 9.49, is an example of the tolerances that are
of importance to the mold manufacturer and precaster vs. the segment tolerances. The table
 Fabrication of precast segmental tunnel linings 569

Figure 9.49 Example of tolerances that the mold manufacturer tracks.

Note: For every job, the engineers, clients, contractors, precaster and mold manufacturer,
depending on the diameter of the tunnel, determine acceptable geometric tolerances.
570 Handbook of Precast Segmental Tunnel Lining Systems

Figure 9.50 Orientations of mold circumferential joints and radial joints –​ measurable/​a djustable
values.
Note: It is important to have a diagram related to the tolerance nomenclature to avoid
misunderstandings.

also gives an indication of how the measurements can be taken (method) at the three time
intervals

• When the molds are fabricated

• When the molds are in production
• Production segments

Diagrams in Figures 9.50 to 9.53 provide information for the elements of the tabular sum­­
mary illustrated in Figure 9.49.

9.13 SEGMENT QUALITY CONTROL –​ PRECASTER’S CONSIDERATIONS

QA/​QC technicians and inspectors are not the only ones involved in the quality of the pre-
cast segments: everyone involved in the production of the segments has their part to play in
the finished product. The quality of segments is first and foremost built-​in and then checked.
Quality is built-​in, not check-​in.
Segment quality control is guaranteed by a robust quality control system that usually
follows the known reputable industry standards, and precast industry accreditation bodies.
In North America, the precast plant quality control systems for tunnel liner manufacturing
are at minimum based on PCI (Precast Concrete Institute) PCI Manual 116 recommenda-
tions (NPCA has their equivalence). Also, some specifications require the tunnel liner
 Fabrication of precast segmental tunnel linings 571

Figure 9.51 Radial and circumferential joint tolerances –​ flatness.

(a) Diagram for the understanding of the flatness criteria in the segment mold.
(b) Diagram for the understanding of the flatness criteria in the finished segment.

Figure 9.52 Mold (or segment) width dimension check points.

Note: Mold width dimension check points. Typical casting punch marks located on the circle joint
surface of the mold and are used to locate where the inside micrometer must be set to take
measurements. These punch marks become apparent on the concrete segment and be used for
corresponding measurements.
572 Handbook of Precast Segmental Tunnel Lining Systems

Figure 9.53 Diagrams of mold and segment measuring points.

(a) Mold dimensional tolerance diagram showing typical diagonal check measurements using an
inside micrometer (or a steel tape).
(b) Segment dimensional tolerance diagram showing typical diagonal check measurements that
can only be made with a 3D measurement device.

manufacturing to be certified PCI or NPCA (National Precast Concrete Association) and

thus automatically the precast productions would follow the quality system requirements
of these organizations and consequently their independent auditing systems, which entail
several quality system audits per year.
The quality of segments requires that production equipment and dimensional checking
instrumentation be calibrated and follow a recommended calibration frequency (to the
appropriate related industry standards). Each segment should have traceability. This is the
essence of segment manufacturing. There are many ways of achieving an efficient quality
control traceability: with a mix of “old-​school” paper and pen driven data records and
documents, combined with data from the different computerized systems or completely
paperless driven computerized systems, as described in Chapter 8.
It could be argued that the ideal situation is a setting where a mix of manual input track-
ing is mixed in with more recent digital quality control and tracking systems. Both manual
and digital systems have their advantages and limitations –​and they should be viewed as
complementary to one another.
Each element of the quality system should provide useful data if the quality control
system is designed correctly. The main categories of data that are usually collected in a pre-
cast operation are as follows:

• Pre-​pour inspection
• Post-​pour inspection and repair (if it occurs)
• Batch tickets for the concrete used
• Fresh concrete testing
• Hardened concrete testing data (cylinder breaks, beam breaks when steel fibers are
used), mold dimensional checks, and steel cage inspections
• Concrete constituents’ material testing data (mill tests, aggregate testing)

There is a considerable cost to using the latest technology in terms of traceability. For
example –​software development (usually plant customized), hardware requirements
­
 Fabrication of precast segmental tunnel linings 573

(scanning guns, etc.), plant Wi-​Fi, repeaters and internet technology required, personnel
to manage the data and systems, system crashes affecting production schedules, etc. Often,
dealing with the large amount of data generated by these systems can turn into polemic ses-
sions of interpretation and argument which in turn, can lead to unnecessary further inves-
tigations on unimportant elements, unrelated to the product’s performance, which directly
or indirectly severely penalize the precaster (most frequently because of the burden of proof
applied by the tunnel contractor and owner). In the age of modern computerized systems
that generate tremendous amounts of data, it is important to maintain attention on the ori-
ginal intent and that more data may not mean more knowledge and intelligence if it is not
managed properly and if the persons involved do not act responsibly. The intelligent and
responsible use of technology is great and powerful, but its misuse is a waste of resources,
money and brings a false sense of assurance orsecurity.

9.13.1 The old school paper trail method of tracking quality control

Basically, there are two main stop points in precasting operation workflows, as shown in
Figure 9.58: the pre-​pour inspection and the post-​pour inspection.
Traditionally, with pen and paper quality control systems, once the segment is cast, it is
generally considered as part of a ring, and each ring is given a ring number and thus in some
tunnel liner precast quality control systems, the tracking of product is done with a docu-
ment based on the ring number.
The paper quality control documents are filled in by hand and can be commented on with
quality control inspector personal notes. The automated digital systems have the same fea-
ture, where the quality control inspector can add a comment but it must be noted that cer-
tain individuals are “detail-​oriented” with pen and paper, whereas others are more verbose
with the digital technology. A sign of a healthy quality control system is that the inspectors
leave comments behind so that apart from checking of boxes and the necessary corrections
taking place, a deeper understanding on root causes and preventive actions are initiated.
The tracking of segments and rings is done with the identification of segments with a
stamped ID number. Each segment is stamped with a code that indicates the date, shift, and
ring number. An example segment stamping would be 2022-​04-​18–​1–​1222 [the first three
numbers being the date, the fourth number being the shift number (one or two) and the last
number being the ring number].
In paper and pen quality control systems, a final separate quality control document is in
the form of a shipping report/​manifest (again, ring-​number based) that is filled-​out and sent
with the transport vehicle to the jobsite for a counter signature that confirmed the product
received is damage free.
It must be noted that many precast operations do not track “rings” (a set of the segments
required to make-​up one ring). There is simply a “segment-​based” tracking system. In these
situations, the segments of a type are often stored and stacked separately.

9.13.2 Automated digital quality control tracking system of quality

control
The difference between the pen and paper “old school” method and the digitized method
is that the digitized method is segment-​number based, as opposed to ring-​number based.
The main advantage to the segment-​numbering base that the digitized system has is that
a big database is created, and different reports can be gleaned of it. The one disadvantage
that the segment-​based system is less intuitive than the ring-​based system that the pen and
574 Handbook of Precast Segmental Tunnel Lining Systems

paper method is based upon. People generally love the greater data capabilities but crave
the more tangible quick dashboard view that the pen-​and-​paper reports give with the ring-​
based reports.
The digital quality control tracking systems are based on bar code and/​or QR code scan-
ning using hand-​held scanners. The molds each have a bar code and/​or QR code and when
a mold undergoes a QC check, this bar code or QR code is scanned. When the mold is cast,
again, the mold codes are scanned, linking it to the batches that went into the specific mold.
When the mold is stripped, again the mold QR or bar code is scanned and linked to the
new tag that will be printed and glued onto the freshly stripped segment. At this point, the
segment is officially born and has its identity linked to the mold it came from and quality
controls that were done in the pre-​pour inspection. Figures 9.54 and 9.55 shows an example
of a mold QR code tag. An embedded number and bar code for segments and a stick-​on QR
code for segments as shown in Figures 9.56 and 9.57.
As the segment progresses through the rest of the operation, as it may possibly get
repaired, or ready for shipping, it gets scanned and quality control inspections get recorded
into the growing database. Please refer to Figure 9.58, which shows a detailed flow chart of
the segment production process.
One of the main advantages with the databases is that one can easily track defects by type
and one can more easily see how their frequency evolves with time, although the quality
control team usually sounds the alert when certain issues arise in repetition. There are many
reports that can get generated from the segment quality control database, but most are
standardized from the onset in conjunction with the software developer. One issue to note
is that not everyone has the same idea of what a “dashboard report” should show. What is
important to the precaster may not be what the owner’s engineer may want or expect to see.
There is evolution to come with these new systems for digital capture of data and all of the

Figure 9.54 Mold QR code tag.

Note: Example of segment mold QR code tag. Individual tags are placed on each mold to provide
a unique identifier.

Figure 9.55 Segment tag.

Note: Example of segment QR code tag. Individual tags are placed on each segment to provide
a unique identifier.
 Fabrication of precast segmental tunnel linings 575

Figure 9.56 Segment mold with QR identifier tag.

Note: Segment mold in a precast plant with a QR code tag. Individual tags are placed on each
segment mold and become part of the QC “life story” for the segment.

Figure 9.57 Segment with QR identifier tag.

Note: Segment in a precast plant with a QR code tag. Individual tags are placed on each segment
and become part of the QC “life story” for the segment.

parties involved; the contractor, the owner, the owner’s engineer and the precaster. Software
and hardware developers must work together to make sure everyone’s expectations are pos-
sible, reasonable, cost effective, and that they can be met.
It is also apparent that there is a cost in software development in order to produce custom
reports, special written report formats as well as specific dashboard formats, few of which
are embedded in advance in the contract specifications. And there is often the case, because
this is a new innovation, where one party desires to add or change parts of the data or report-
ing formats, these are not readily possible once the segment precasting has commenced. And
finally, if these systems get too intense, data acquisition and reporting becomes a significant
576 Handbook of Precast Segmental Tunnel Lining Systems

Figure 9.58 F low chart of the segment manufacturing process –​ overview and quality control
stop points.
Note: Technopref Industries, Inc, Cape Charles, Virginia; Hampton Roads Bridge and Tunnel
Project, Norfolk, Chesapeake, Virginia Beach, Hampton, VA.
 Fabrication of precast segmental tunnel linings 577

task that requires an experienced management team as big as the production quality control
team. Few understand the energy that these data reporting systems demand for the limited
benefit that may result (i.e. tracking data with little value or follow-​up). On the other hand,
there are significant benefits to come in regard to these new digital data systems but more
industry collaboration required to bring these systems on-​line and make them more func-
tional and productive.
With the digital quality control tracking systems, there is an important learning curve
with the personnel. Learning to input the quality control checking data is not that easy in a
bustling plant environment where the learning is going on at the same time as a production
kicks-​off with a learning curve for everyone (laborers, lead people, client inspectors, owner’s
inspectors), and often debugging systems and procedures. One of the main issues is not the
gathering of data, it is the registering of data once the checking is done. The inspector may
run all the checks and click the boxes but there is always a last step where a final approval
must be given and at times this is forgotten and going back to correct this step is compli-
cated and time consuming.

9.14 LOGISTICS OF SUPPLY –​ PRECASTER CONSIDERATIONS

The logistics related to a precast segment manufacturing plant are one of the most challenging
aspects of the work –​often, these operations require the involvement of the most rigorous and
talented personnel because there are hundreds of ways that one can drop the ball and add costs
to the operations, paralyze or hinder the tempo of the operations at the plant, or affect opera-
tions at the TBM site. Typical precasting plant facilities are shown in Figures 9.59 and 9.60.

Figure 9.59 Precast segment manufacturing plant.

Note: Aerial view of a large precast segment manufacturing plant, complete with concrete batch-
ing and aggregate processing and storage facilities.
578 Handbook of Precast Segmental Tunnel Lining Systems

Figure 9.60 Precast segment manufacturing plant.

Note: Aerial view of a large precast segment manufacturing plant, complete with concrete batch-
ing and aggregate processing and storage facilities.

For example, the logistics of the precast plant encompasses the coordination of the
following:

• All the components of the plant mechanical systems (including, at times, the erection
of a batch plant and its ancillary systems), and of rented equipment.
• Consumables (for example, release agents).
• Cast-​in components (­example –​sockets, grout/​lifting inserts, guide rods).
• Arrival, staging and coordination of the accessories that need to be sent to jobsite for
ring erection in the TBM (i.e. steel bolts, washers, dowels).
• Concrete constituents: aggregates, cement, slag, fly ash, silica fume, admixtures, steel
fiber big bags, reinforcing steel.
• The staging and storage of the above-​mentioned elements within the precast building
and yard is complex because poor timing results in literally nowhere to put things. If
there is lengthy downtime in production, scheduled deliveries that are in ocean vessels
or coming from continental suppliers do not stop their journeys and the plant gets
overwhelmed.
• Storing of the final product in the yard:
• If there are delays with tunneling, managing a precast yard to maximize storage is
very complex. These are large elements and every time you handle them comes the
risk of damaging them.
• If the precast storage yard gets filled-​up, there may be a need to transport the
product to temporary remote yards which entails renting land, renting haul trucks
for the transfers, extra labor, and renting extra costly heavy equipment in the new
yard to unload and eventually load-​out. These are costly energy-​intensive situ-
ations. Please refer to Figures 9.61 to 9.67.
• Transporting segment-​rings to jobsite with inconsistent mining schedules.
 Fabrication of precast segmental tunnel linings 579

Figure 9.61 Highway transport of segments.

Note: Example of a partial precast segmental ring being transported by highway truck from plant.

Figure 9.62 Highway transport trailor –​ loaded.

Note: Example of a fully loaded highway trailer with six partial stacks of precast rings loaded at
precast plant.

• In certain instances, transport to the jobsite may be achieved by truck, train,

barge, or a combination of these. With trucking, ring weight considerations are
important because rings are generally shipped in full stacks or half stacks, not by
type of segment and the weights of the divided loads must be studied with respect
to allowable DOT load restrictions and transport costs can easily go up with cer-
tain indivisible load combinations.
• There are also areas of the country that have seasonal load limits for certain roads,
and these situations also increase the logistics coordination workload and trans-
port costs because during these periods, the haul trucks will be handling partial
580 Handbook of Precast Segmental Tunnel Lining Systems

Figure 9.63 Highway transport trailers –​ loaded.

Note: Example of fully loaded highway trailers with full stacks of precast rings loaded at precast
plant. Segments have hydrophilic gaskets that needed protection from the weather.

Figure 9.64 Highway transport trailor –​ loaded.

Note: Example of fully loaded highway trailer with partial stack of a precast ring loaded at pre-
cast plant.

loads. In these situations, the number of deliveries for the job will go up and
the jobsite will also have to manage different types of deliveries, resulting in dis-
continuous reception routines, oddball jobsite storage and staging patterns, thus
potentially creating havoc and errors.
To be able to have a good handle on logistics, one needs to plan schedules properly, main-
tain excellent communication with your suppliers, subs, transporters, trucking companies
and client. A deep understanding of the lead times and developing plan B’s, C’s, and D’s for
each of the elements is required to adapt and adjust to changes. Please see Figures 9.62 to
9.65 for segments in transport tot eh site from the precasting plant.
 Fabrication of precast segmental tunnel linings 581

Figure 9.65 Precast segments in storage at the manufacturing plant.

Note: Typical view of a stacked precast segments in storage at the manufacturing plant. In this
case, the segments are stacked in a manner to allow for rapid handling and loading by forklift
onto highway transport equipment.

Figure 9.66 Precast segments in storage en-​r oute to the site.

Note: Typical view of a stacked precast segments in storage after leaving the manufacturing plant,
complete with engineered dunnage materials.

9.15 PRECAST PLANT LABOR CREWS

We have discussed numerous important subjects in this chapter up to this point, showcased
many unique topics that featured technology, machinery, complex systems, but at the core
of a precast operation are people, no matter the level of automation that a tunnel liner pre-
cast plant has, precast work is a people job. People are the heart of the operation; i.e. lead-
ers, superintendents, foremen, maintenance people, system operators, laborers. It is exciting
fast-​paced work where one must think and be alert. The most important aspect of segment
production are the crews that make things happen.
582 Handbook of Precast Segmental Tunnel Lining Systems

Figure 9.67 Precast segments in storage on-​s ite.

Note: Typical view of a stacked precast segments in storage at the manufacturing plant. In this
case, the segments are stacked in a manner to allow for rapid handling and loading by forklift
onto highway transport equipment.

9.15.1 Crews for precast shifts

No matter if the precasting operation is the carousel type or a fixed pour type, there is a
minimum number of laborers that are required to keep the operation running. We will not
address the precise manpower crew sizes since there are many factors to consider but as an
estimate, we may have 15 to 35 active workers during a production shift. This would include
superintendents, foremen, maintenance crews, general laborers, finishers, batch plant oper-
ators, carousel operators, forklift operators, overhead crane operators and reinforcing steel
crews, with the understanding that in a precast plant, the personnel are trained and cross-​
trained and able to handle many positions, systems, and pieces of equipment.

9.15.2 Skills required to work in a precast segment plant

People with skills or skillsets are always sought after and welcome in a precast plant set-
ting. However, the reality is that in today’s world, and the way our education system is set
up, with few vocational education programs, people skilled in crafts or trades are rare.
However, the precast industry has traditionally been an industry that accepted green walk-​
ins and developed them in-​house, starting people in simpler tasks and moving them up as
experience and skills set in.

9.15.3 Plant safety considerations

The people reading this book are well accustomed with the safety programs, customs, rules
and systems of the big industrial TBM construction sites but may not know how things are
run safety-​wise in precast plants. Precast operations are very similar in terms of safety as
construction site settings as they are subject to the same national government regulations.
The plants have, for instance, safety programs, toolbox meetings, lock-​out-​tag-​out systems,
with orientation programs for workers. There is an effort made to ensure that plant layouts
and equipment such as molds are designed as ergonomically as possible and adjustments are
made to process and procedures as needed with the input of everyone involved, just like on
the construction sites. There are many positions that are working positions (working fore-
men) and the input from hands-​on people is important.
 Fabrication of precast segmental tunnel linings 583

9.15.4 Craft training –​ general concepts

The basic training for a stationary mold operation and a carousel operation is very similar.
Both types of manufacturing systems can be started with green crews with seasoned leaders.
The skills that will be taught and learned for either manufacturing system are basically the
same. What changes is that with a carousel system, the work (molds) moves to the worksta-
tions, whereas in a fixed pour system, worker move to the work (to the molds).

9.15.5 Craft training –​ carousel casting system

With a carousel system, a semi-​automated mechanical system, the molds move to pre-​
determined workstations –​for example, a stripping station, a cleaning station, an oiling
station, a gasket installation station, a rebar cage installation, an insert installation station,
a pouring station, and a finishing station. There are usually 10 to 12 workstations (or less)
on a carousel work line and some of the workstations that are listed above are usually com-
bined –​every carousel is different. New laborers are paired with more experienced workers
at a station until they have been trained for the work. New laborers are usually cross trained
at each of the stations so that they can get an understanding of the operation and develop
process and flow awareness and be able to fill in at any position if need be. However, with
time, some laborers will gravitate to certain positions/​stations.
The work on a carousel is linear in the sense that the work (the molds) is on rails; coming
at you and leaving you on a horizontal surface. There may be some overhead crane action
on the line, but it is usually in a fixed position, so the focus of the workers is mainly at
eye-​level.

9.15.6 Craft training –​ stationary mold system

With a fixed-​mold system, the molds have a set position, and everything is brought to each
mold (i.e. all cast-​in components) and the cast segments are pulled-​out of molds with the
help of overhead cranes. This type of operation is a highly coordinated dance of many over-
head cranes operating simultaneously in different part of the crane runway. So, in this type
of production, the worker’s focus is in all directions, with attention above, attention at eye-​
level in a 360° direction and downwards, at the floor level because there is a lot of walking
and moving, and many moving parts.
The same tasks that are done in a carousel system are done in a fixed-​mold system,
so the new laborer gets paired with more experienced workers and follows the different
crews down the production hall. There are dedicated crews for tasks such as stripping
molds, cleaning molds, oiling molds, installing inserts and gaskets, reinforcing cages if
there are, closing molds, etc. Again, as with a carousel system, it is important to cross-​
train crews as much as possible, but the nature of the operation lends itself less to this.
Workers here too gravitate early-​on to positions that are more pleasant to them or more
suited to them.

9.15.7 Precast labor crews –​ conclusions

The subject of labor in this chapter got the least amount of attention, mainly because there
was so much that we wanted to explain about the whole process, but it is most important,
one of the most complex and demanding for the precast tunnel liner operation. The amount
584 Handbook of Precast Segmental Tunnel Lining Systems

of energy and resources that are required to recruit, train, sustain, keep safe, and support
a workforce with high turnover is significant. At the end of the day, it is the boots on the
ground that win wars, not all the fancy equipment.
The precaster will need to set-​up a plant in record time and when the whistle blows,
a green crew will need to be trained and kept safe with seasoned leaders and procedures
that will be constantly adapting, and very quickly will have to produce product of the
highest quality (often with zero tolerances) while de-​bugging systems, as all eyes from
the contractor, owner’s engineer and owner are scrutinizing work over the shoulder of the
precaster, even as some technical specifications are still being debated, argued or inter-
preted. In these environments, the seasoned owners, owner’s representatives, inspectors
and contractors understand the enormity of what is being achieved and leave the pre-
caster with the breathing room to take care of his most important assets; e.g. the labor
that is being trained and kept safe, while the precaster debugs, adapts and improves the
process and machinery. High quality becomes a natural by-​product of the operations, for
every segment produced.

REFERENCES AND ADDITIONAL READING MATERIALS

For preparation of this chapter, the following references were used which are categorized into two
categories of authority published documents and authored documents.

CODES AND REGULATIONS

AMERICAN CONCRETE INSTITUTE (ACI)

ACI 201.2R-​16, Guide to Durable Concrete.

ACI 211.1, Standard Practice for Selecting Proportions for Normal, Heavyweight, and Mass
Concrete.
ACI 214.3R, Simplified Version of the Recommended Practice for Evaluation of Strength Test Results
of Concrete.
ACI 222R-​19, Guide to Protection of Reinforcing Steel in Concrete against Corrosion.
ACI 224R, Control of Cracking in Concrete Structures.
ACI 301, Specifications for Structural Concrete.
ACI 304R, Guide for Measuring, Mixing, Transporting, and Placing Concrete.
ACI 305R, Guide to Hot Weather Concreting.
ACI 306R, Guide to Cold Weather Concreting.
ACI 318-​19, Building Code Requirements for Structural Concrete and Commentary.
ACI 503.4, Standard Specification for Repairing Concrete with Epoxy Mortars.
ACI 517, Recommended Practice for Atmospheric Pressure Stream Curing of Concrete.
ACI 533.5R-​20, Guide for Precast Concrete Tunnel Segments.
ACI 544.1R-​96, Report on Fiber Reinforced Concrete.
ACI 544.3R, Guide to Specifying, Proportioning, and Production of Fiber Reinforce Concrete.

AMERICAN SOCIETY FOR TESTING MATERIALS (ASTM) INTERNATIONAL

ASTM A1064/​A1064M, Standard Specification for Carbon-​Steel Wire and Welded Wire Reinforcement,
Plain and Deformed, for Concrete.
ASTM A307, Standard Specification for Carbon Steel Bolts, Studs, and Threated Rod 60,000 PSI
Tensile Strength.
ASTM A36/​A36M, Standard Specification for Carbon Structural Steel.
ASTM A615, Deformed and Plain Carbon-​Steel Bars for Concrete Reinforcement.
 Fabrication of precast segmental tunnel linings 585

ASTM A706/​ A706M, Standard Specification for Deformed and Plain Low-​ Alloy Steel Bars for
Concrete Reinforcement.
ASTM A820, Standard Specification for Steel Fibers for Fiber-​Reinforced Concrete.
ASTM C1012/​C1012M, Standard Test Method for Length Change of Hydraulic-​Cement Mortars
Exposed to a Sulfate Solution.
ASTM C1064/​C1064M, Standard Test Method for Temperature of Freshly Mixed Hydraulic-​Cement
Concrete.
ASTM C1116/​C1116M, Standard Specification for Fiber-​Reinforced Concrete.
ASTM C1157/​C1157M, Standard Performance Specification for Hydraulic Cement.
ASTM C1202, Standard Test Method for Electrical Indication of Concrete’s Ability to Resist Chloride
Ion Penetration.
ASTM C1218/​C1218M, Standard Test Method for Water-​Soluble Chloride in Mortar and Concrete.
ASTM C1240, Standard Specification for Silica Fume Used in Cementitious Mixtures.
ASTM C138/​C138M, Standard Test Method for Density (Unit Weight), Yield, and Air Content
(Gravimetric) of Concrete.
ASTM C143/​C143M, Standard Test Method for Slump of Hydraulic-​Cement Concrete.
ASTM C150/​C150M-​18, Standard Specification for Portland Cement.
ASTM C1556, Standard Test Method for Determining the Apparent Chloride Diffusion Coefficient of
Cementitious Mixtures by Bulk Diffusion.
ASTM C1567, Standard Test Method for Determining the Potential Alkali-​Silica Reactivity of
Combinations of Cementitious Materials and Aggregate (Accelerated Mortar-​Bar Method).
ASTM C1604, Standard Test Method for Obtaining and Testing Drilled Cores of Shotcrete.
ASTM C1607, Standard Test Method for Determination of “Microwave Safe for Reheating” for
Ceramicware.
ASTM C1609 /​C1609M–​19a, Standard Test Method for Flexural Performance of Fiber-​Reinforced
Concrete (Using Beam With Third-​Point. Loading)
ASTM C231/​C231M, Standard Test Method for Air Content of Freshly Mixed Concrete by the
Pressure Method.
ASTM C260/​C260M, Standard Specification for Air-​Entraining Admixtures for Concrete.
ASTM C31/​C31M, Standard Practice for Making and Curing Concrete Test Specimens in the Field.
ASTM C33/​C33M, Standard Specification for Concrete Aggregates.
ASTM C39/​ C39M, Standard Test Method for Compressive Strength of Cylindrical Concrete
Specimens.
ASTM C496/​C496M, Standard Test Method for Splitting Tensile Strength of Cylindrical Concrete
Specimens.
ASTM C595/​C595M, Standard Specification for Blended Hydraulic Cements.
ASTM C618, Standard Specification for Coal Fly Ash and Raw or Calcined Natural Pozzolan for Use
in Concrete.
ASTM C666/​C666M, Standard Test Method for Resistance of Concrete to Rapid Freezing and
Thawing.
ASTM C920, Standard Specification for Elastomeric Joint Sealants.
ASTM C989/​C989M, Standard Specification for Slag Cement for Use in Concrete and Mortars.
ASTM D1149, Standard Test Methods for Rubber Deterioration–​Cracking in an Ozone Controlled
Environment.
ASTM D1171, Standard Method for Rubber Deterioration–​ Surface Ozone Cracking Outdoors
(Triangular Specimen).
ASTM D2240, Standard Test Method for Rubber Property–​Durometer Hardness.
ASTM D395, Standard Test Method for Rubber Property–​Compression Set.
ASTM D412, Standard Test Method for Vulcanized Rubber and Thermoplastic Elastomers–​Tension.
ASTM D471, Standard Test Method for Rubber Property–​Effect of Liquids.
ASTM D573, Standard Test Method for Rubber–​Deterioration in an Air Oven.
ASTM D638, Standard Test Method for Tensile Properties of Plastics.
ASTM F436/​F436M, Standard Specification for Hardened Steel Washers Inch and Metric Dimensions.
586 Handbook of Precast Segmental Tunnel Lining Systems

AUSTRIAN SOCIETY FOR CONCRETE AND CONSTRUCTION TECHNOLOGY (ÖVBB)

ÖVBB 2011, Guideline for Concrete Segmental Lining Systems.

BRITISH STANDARDS INSTITUTION (BSI)

BS PAS 8810:2016, Tunnel Design –​Design of Concrete Segmental Tunnel Linings –​Code of Practice.

BRITISH TUNNELING SOCIETY (BTS) AND THE INSTITUTION OF CIVIL ENGINEERS (ICE)

BTS ICE 2004, Tunnel Lining Design Guide.

CANADIAN STANDARDS ASSOCIATION (CSA)

CSA A23.1-​19, Concrete Materials and Methods of Concrete Construction.

CSA A23.4-​16(21), Precast Concrete –​Materials and Construction.

EUROPEAN STANDARDS/​E UROPEAN NORMS (EN)

EN 196-​2, Method of Testing Cement–​Part 2: Chemical analysis of cement.

EN 206-​1:2013, Concrete–​Specification, Performance, Production and Conformity.
BS EN 1992-​1-​1:2004, Eurocode 2: Design of Concrete Structures –​Part 1-​1: General Rules and Rules
for Buildings.
EN 14651, Test Method for Metallic Fibre Concrete–​Measuring the Flexural Tensile Strength (Limit
of Proportionality (LOP), Residual).
EN 14889-​1, Fibres For Concrete–​Part 1: Steel Fibres–​Definitions, Specifications And Conformity.

FRENCH TUNNELING AND UNDERGROUND ENGINEERING ASSOCIATION (AFTES)

AFTES:2005, Recommendations for the Design, Production, and Installation of Segmental Ring
Recommendation for the Design, Sizing and Construction of Precast Concrete Segments
Installed at the Rear of a Tunnel Boring Machine (TBM)s.

GERMAN TUNNELING COMMITTEE (DAUB)

DAUB:2013, Lining Segment Design: Recommendations for the Design, Production, and Installation
of Segmental Rings.

INTERNATIONAL FEDERATION FOR STRUCTURAL CONCRETE (FIB)

fib bulletin 34, Model Code for Service Life Design.

INTERNATIONAL ORGANIZATION FOR STANDARDIZATION (ISO)–​ISO CENTRAL

SECRETARIAT

ISO/​TC 17/​SC 17 Steel Wire Rod and Wire Products. (2013). ISO 13270:2013–​Steel Fibres for
Concrete –​Definitions and Specifications. Geneva, Switzerland: ISO, the International
Organization for Standardization.
 Fabrication of precast segmental tunnel linings 587

ADDITIONAL READING MATERIALS

fib Special Activity Group 5 (2010). Bulletin 55-​ 56: Model Code 2010–​ First Complete Draft.
Lausanne: International Federation for Structural Concrete (fib).
fib Working Party 1.4.1 Tunnels in Fibre-​Reinforced Concrete. (2017). fib Bulletin 83 Precast Tunnel
Segments in Fibre-​Reinforced Concrete State-​of-​the-​Art Report. (A. Meda, Editor) Lausanne,
Switzerland: Fédération International du Béton (fib).
Gerhard Vitt. (2011). Understanding Steel Fibre Reinforced Concrete: Dramix (Version 1.0 ed.).
Zwevegem, Belgium: NV Bekaert SA.
ITA Working Group 2–​ Research; Bakhshi, M; Nasri, V.; Main Authors. (April 2019). ITA
Report No.22–​Guidelines for the Design of Segmental Tunnel Linings. Longrine, Avignon,
France: AITES/​ITA.
ITA Working Group 2–​Research; Tiberti, G.; Chiniottri, E.; Pizzari, G., Main Authors. (April 2016).
ITA Report No 16–​ Twenty Years of FRC Tunnel Segments Practice: Lessons Learnt and
Proposed Design Principals. Avignon, France: AITES/​ITA.
ITAtech Activity Group Support. (2018). ITAtech Report No.9: Guideline for Good Practice of
Fibre Reinforced Precast Segment–​Volume 2: Production Aspects. Longrine: The International
Tunnelling and Underground Space Association.
ITAtech Activity Group Support. (2016). ITAtech Report No.7; ITAtech Guidance for Precast Fibre
Reinforced Concrete Segments–​Volume 1: Design Aspects. Longrine: International Tunnelling
and Underground Space Association.
National Precast Concrete Association (NPCA). (20 Jan 20). NPCA Quality Control Manual For
Precast Concrete Plants 14th Edition. Carmel, IN 46032: National Precast Concrete Association
(NPCA).
Precast/​Prestressed Concrete Institute. (1999). MNL-​116-​99: Manual for Quality Control for Plants
and Production of Structural Precast Concrete Products, 4th Edition. Chicago: Precast/​
Prestressed Concrete Institute.
Roach, M. F., Lawrence, C. A., Klug, D. R., Fulcher, W. B. (editors) (2017). History of Tunnelling in
the United States, Society of Mining Engineers (SME), Littleton, CO, United States, pp. 564.
ISBN-​10: 0873354303, ISBN-​13: 978-​0873354301
Chapter 10

Transportation, handling and

installation of precast segmental
tunnel linings
Lead author: Shane Yanagisawa
Contributing author: Werner Burger

10.1 INTRODUCTION
The logistical challenge to transport, handle, and install segmental tunnel liner rings has
existed ever since the invention of the prefabricated tunnel liner. In the first tunnel built
under the St. Clair River linking Sarnia, Ontario with Port Huron, Michigan (1889 to
1891) the tunnel was lined with cast iron segments forming rings 6.4m (21 feet) in outside
diameter and 460 mm (18 inches) wide. Each ring had 13 curved segments weighing about
460 kg (1,000 lbs). The tunneling shield had a crane with an arm (Figure 10.1) that pivoted
on the center of the shield with a counterweight on the other end to hoist the segments into
position. The arm could grab and maneuver a segment in three dimensions. The segment
was positioned and bolted in place by a 15-​man crew that could assemble a complete ring
in 15 minutes.
In a modern tunnel, heavy precast concrete tunnel liner segments must be delivered on
time, in the correct order, and without damage to the segments. The assembled segmental
liner ring is the final product on most projects, so ring assembly must be done with pre-
cise placement of individual segments. Gasketed precast segments provide a watertight seal
at pressures up to 20 bar (290 psi) and keep the groundwater from entering the tunnel.
Efficiency of the segmental liner installation is important due to the work being on the pro-
ject schedule’s critical path schedule and the activities on a TBM are repeated thousands of
times like a production line at a factory.

10.2 SEGMENT SUPPLY LOGISTICS

10.2.1 Segment production location

Precast concrete segments may be fabricated at a remote location or at the tunnel site.
A shorter tunnel may not justify the set-​up of a segment factory. Often a segment fac-
tory is initially set up for a nearby project and then follow-​on contracts in a tunneling
program provide enough business and economic justification to keep the segment factory
open. Segment factories are sometimes set up at a quarry site where the aggregates are at-​
hand and there is enough space to stockpile finished segments. A segment factory may be
set up as an expansion of an existing precast concrete operation where good infrastructure
and transportation options exist.
For long tunnels or tunnels with large heavy segments, setting-​up an on-​site segment pro-
duction facility can make sense if there is enough room to store the segments. An on-​site
facility can eliminate size and weight limitations allowing for larger segments and save on

DOI: 10.1201/9781032675541-10
 Transportation, handling and installation 589

Figure 10.1 Erecting cast iron tunnel liner segments at the St. Clair Tunnel.
Note: Illustration courtesy of Lambton Heritage Museum, Grand Bend.

the cost of shipping segments to the jobsite. The handling of segments is reduced and thus
the chance of segment damage is also reduced. The reduced handling also provides a direct
safety benefit with the reductions in handling and inspection.

10.2.2 Logistics regarding ring type

Common ring types for one-​pass linings are the “universal ring” and the left-​right-​parallel
rings. As the design of these rings is described in other chapters, only a very brief descrip-
tion of the ring types is mentioned here. As the name implies, the universal ring can be used
anywhere in the tunnel. The ring is rotated to produce the amount of horizontal and/​or
vertical curvature in the tunnel lining required to satisfactorily follow the TBM. The key
segment for the ring can be located at any allowed position around the tunnel perimeter.
The left-​right ring system requires two ring types, and the left-​right-​parallel ring system
requires any one of three ring types depending on whether the TBM is on a tangent or
curves. The key piece for the left-​right ring system is always located near the crown of the
tunnel. Modern TBM guidance systems can include software that predicts the sequence of
ring types required for delivery to the heading relative to the tunnel alignment and the cur-
rent location and orientation of the TBM.
590 Handbook of Precast Segmental Tunnel Lining Systems

10.2.3 Stacking segments in correct order for installation

With either ring type mentioned above, the goal is to provide the correct precast segments
in the correct order to the heading when needed. The correct order of segments within a
ring is determined by the contractor’s preference for the ring building sequence. Some con-
tractors prefer to build rings in the tunnel by starting with the segment opposite the key
and placing subsequent pieces alternating on either side of the first piece. For example, a
ring with six pieces plus a key marked “A” through “G” where “A” is the key, the erection
sequence would be E-​>D-​>F-​>C-​>G-​>B-​>A-​>(key). The rings would be delivered to the site
and stacked with that order (Figure 10.2). For larger diameter rings where a single complete
ring set of segments would overload a truck, the segments will arrive in a split shipment,
which is assembled on-​site into an order.
Other contractors prefer to start with the first piece next to the key and build the ring all
the way around to the key piece in which case the erection sequence would be B-​>C-​>D-​>E-​
>F-​>G-​>A-​>(key) and the segments for a single ring would be stacked in that order with the
key being placed on top of the E segment.

Figure 10.2 Segment stack in storage yard.

Note: Segment stack organized for delivery to the tunnel heading. Forklift access points shown.
Drawing courtesy of Technopref Industries Inc.
 Transportation, handling and installation 591

The erection sequence needs to be determined before the segment molds can be ordered.
The precast plant needs to organize the molds for a ring so that as the segments are removed
from the molds and stacked, the segments get stacked and stored in the order requested by
the contractor. If a delivery is split because of weight considerations, the stacks need to be
delivered with the bottom half of the stack first. Depending on the location of the precast
plant to the job site there are some creative ways to cut down on trucking costs. If the pre-
cast plant is hours away, consideration for adding an additional stone to the truck load if
room and weight permits may be a possibility. The stacking of the added segments would
be completed at the job site and would have to be managed logistically.
With a universal ring, all the rings are the same so the site yard only stages one type of
ring. Delivery from the precast plant to the job site needs to meet daily average production.
With a left-​right-​parallel type rings, segments sent to the tunnel need to be coordinated with
the heading to make sure the correct ring is sent to the heading. This also means that the
yard needs to store and have access to three different types of rings. The different type of
rings cannot block-​off access by storing one type of ring in front of a different type, unless
rings are stored and retrieved by a crane. There may be more than one ring in the process of
being loaded or in transit to the heading, careful coordination is required to make sure the
correct sequence of rings is supplied to the heading.

10.3 SEGMENT HANDLING ON THE SURFACE

10.3.1 Initial inspection of segments received on-​site

The hand-​off of the segments from the manufacturer to the tunnel contractor normally
occurs at the point of delivery at the jobsite. Even if the manufacturer is the same entity as
the tunnel contractor, there is a need for a well-​defined hand-​off and transfer of responsi-
bility for the quantity and condition of the segments. Once the segments are unloaded, they
become the responsibility of the jobsite. As part of the hand-​off there must be a thorough
inspection of the segments at the delivery point. This inspection should take place before the
segments are unloaded from the delivery vehicle. A check-​off sheet such as the one shown
in Figure 10.3 should be used to begin tracking the delivery of the segments on-​site and the
confirm that the segments have been received in good condition. The inspection must be
done accurately and promptly as the delivery company will charge demurrage for vehicles
held longer (i.e. idled) than a specified time.
A ring with seriously damaged segment(s) should be returned for assessment and dispos-
ition by the manufacturer. This is a rare occurrence. More often, there may be some minor
damage to the segment edges and the manufacturer is formally notified of said occurrence
via the segment inspection and receiving report. Repairs on the jobsite using an approved
method will be done at the expense of the manufacturer. Any repairs on site must be docu-
mented and approved by the contractor’s Quality Control Department. The owner or con-
struction manager may also have inspection personnel on-​site that may participate in repair
decisions.

10.3.2 Segment storage yard layout

The entire process of receiving segments in the site yard, staging them, and delivering them
to the heading must be designed with the goal of eliminating damage to the segments while
making a timely delivery to the heading. The yard must be laid-​out to minimize the handling
592 Handbook of Precast Segmental Tunnel Lining Systems

Figure 10.3 Segment receiving report form.

Note: Inspect and record segment condition when delivered to the jobsite.
 Transportation, handling and installation 593

and movement of tunnel segments. More handling generally leads to more damaged seg-
ments, less handling of segments is preferred. The edges on precast concrete liner segments
near the gaskets are very delicate. The segment edge below the gasket groove is easily
chipped or spalled resulting in an exposed gasket. Sometimes the spalling can extend below
the gasket seat. Damaged segments require costly labor-​intensive repairs, and the space to
make the repairs.
The site yard layout needs to consider the flow of traffic, the size of the yard, the need
for access to different types of liner rings, and the number of rings to be stored on site
based on production numbers and location of the precast plant. There will be a constant
flow of trucks into the yard delivering segments and other supplies along with trucks
hauling away tunnel muck. If possible, the site should separate the traffic for segment
deliveries and muck hauling so that one does not hold up the other. If the yard is large
enough, there may be room to separate the muck pit from the segment storage area
and prevent muck trucks from stacking up while a load of segments is inspected and
unloaded.
On a small site, segments may be unloaded from trucks parked momentarily on an active
street with flaggers directing traffic around the unloading zone.
Segment handling and staging does not happen in isolation but is part of an overall
system that supports the TBM. There must be an overall site plan that addresses not only
the segments, but muck handling, support equipment, materials storage, plant and equip-
ment, and utilities. All these pieces have to be arranged until they fit together like a jigsaw
puzzle to provide the optimum solution for the site. This is not necessarily the ideal solution
but the best that can be achieved given the constraints of the site.
Ideally, the yard should be large enough to accommodate enough rings for a day of
peak production, which is usually about two times the average production. There should
also be enough storage to handle interruptions in supply due to weather and traffic.
When the contractor is in control of the segment plant and the plant is nearby, storing a
minimum number of rings at the tunnel site may be reasonable. If the segment plant is an
outside supplier far away, prudence requires the storage of more rings on site to guard
against shortages, which would shut down the entire tunneling operation. An additional
storage yard for segment storage may be a consideration by the contractor to control the
segment inventory at the job site. Keep in mind that double handling usually results in
more damage to segments.
The site segment storage yard needs to have a flat all-​weather surface capable of handling
high bearing loads from a stack of segments. Truck unloading areas need to be flat so that
the lifting device picks-​up the segment stack evenly. An uneven surface can damage the seg-
ment lifting device and the segments as the load is being picked-​up by the device.
Small sites are the most difficult because all the support equipment, muck storage and
tunnel supplies compete for space. The smallest footprint will come from a 100% crane
set-​up with a tower crane picking the segments directly from a truck and storing them in a
tightly packed configuration (Figure 10.4). The same crane will be used to lower the seg­
ments down the shaft and load the transporter. The contractor should ensure that the seg-
ment stack spacing leaves enough room for workers to maneuver the segment picker over
the stack.
If there is room for more segment storage at the site, there may be room for a storage yard
beyond the reach of the shaft crane. This segment yard might be serviced by another crane
or by a forklift. If the segments are handled by forklift, then a paved surface is needed to
prevent rutting and out-​of-​level operation by the forklift.
594 Handbook of Precast Segmental Tunnel Lining Systems

Figure 10.4 Storing segments on-​s ite using crane and segment picker.
Note: Maximizing segment storage in small yard on-​s ite.

10.3.3 Dunnage needed for stacking finished segments

Segment stacks need to be stored on a flat surface to ensure stability of the stack. Each
ring is generally stored as a single stack with a dunnage scheme approved by the segment
designer. The dunnage is typically wood timber laid across the width of the segment to
evenly distribute loads from the segments above. The dunnage provides a gap between seg-
ments for forks or a segment lifter to reach under prior to lifting the stack.
Depending on the clearance required, dunnage can vary from small to large timbers. The
support base is sometimes a precast cradle instead of wood dunnage. Dunnage can be a
substantial expense when thousands of segments are stored. The dunnage scheme must be
checked by the design engineer to prevent overstressing of segments and a drawing prepared
to show approved dunnage and fork locations. The precast plant must be held accountable
if they deviate from the approved drawing.
Check the ground loading of the segment stacks and improve the ground as needed to
prevent stacks from tipping. The yard crew must consistently follow the dunnage scheme
to prevent damage to segments. Usually, the dunnage is returned to the precast plant.
Discarding dunnage may result in extra charges from the precast plant.
 Transportation, handling and installation 595

10.3.4 Weather protection

Segments generally arrive complete with gaskets, but sometimes the gaskets are not applied
until the segments are on site. Wood or fiber packing (if used) is generally installed just
prior to sending the segments to the heading to avoid deterioration of the packer material in
the fabricator’s storage yard. The intent of the packing is to help distribute the ring-​to-​ring
thrust loads evenly.
In the event gaskets are installed on site, a tent may be required to keep the segments from
getting too hot or cold so that the adhesive can be applied and work as intended. Shelter
from rain may also be required.
If work is performed in extremely wintry conditions, segments may need to be stored in a
warming shed prior to being placed underground to make the gaskets pliable and keep the
annular grout from freezing around the installed liner. Cold segments may have ice in the
shear cones, pockets for lifting fixtures and bolt pockets. Since these segments are stored in
a stack, access to the holes to remove ice cannot happen until the segments are unloaded at
the heading. One may want to heat up segments to thaw the ice to the point where the ice
can be removed.

10.3.5 Segment lifting and hoisting

A precast segmental liner ring will have around five to nine pieces plus a key depending
on the tunnel diameter and the segment design. The key is usually a smaller piece but, in
some cases, it is almost as large at the other segments. Segments can weigh up to 16 tons,
have widths up to 2.2 m (7.2 feet) and can be from 230 to 700 mm (9 to 27 inches) thick.
Concrete strengths will be in the 35 to 55 MPa (5,000 to 8,000 psi) range, although higher
strengths are sometimes required. Internal reinforcement can be steel bar, steel fibers or a
combination. Despite the high concrete strength and massive size of the segments, the edges
on the gasket side chip easily and the gasket and sealing surface can be damaged.
Segments are usually delivered to the jobsite by truck and on occasion by rail. At the
segment plant a large forklift or gantry crane is used to load the truck. The truck usually
has a drop deck trailer to prevent loads from being over-​height. The best practice is to have
special cradles fabricated to carry the segment stacks on the trailer. A common and less
desirable practice is to use wood dunnage on the trailer decks that have a tendency to shift
during transportation. The dunnage between the segments must be compatible with the
method used to unload the segments at the site. If a forklift is used at the site, there must be
adequate clearance for the forks to slide under the segments. Usually there is a time limit in
the precast plant supply contract for offloading each segment truck. Efficiency is important
so the contractor does not incur additional costs from the shipper.

10.3.5.1 Forklifts
Forklifts are a popular way to unload and stack segments. They are a good solution if the
yard is large enough. Forklift operators must be trained and skilled, but they do not require
the certifications that are required for a crane operator. The forklift will be quite large to
handle a segment stack (Figure 10.5). A forklift can only access segment stacks on the out­
side edge of the storage area, so a left-​right-​parallel ring scheme is going to require a layout
for access to all three types of rings. The forklift can free up the shaft crane to service the
shaft, making the positioning of the segment delivery trucks less critical. The forklift should
have HDPE lined sleeves over the forks to prevent damage to segments and match up with
596 Handbook of Precast Segmental Tunnel Lining Systems

Figure 10.5 Forklift handling a segment stack.

Note: Large 25-​t on forklift required.

the curvature of the segment (Figure 10.6). Alternative options for the forklift forks are to
have custom forks fabricated or to use wood dunnage to match segment curvature placed
on top of the forks. The dunnage scheme must consider the placement of the forks to pre-
vent overstressing of the bottom segment in the stack. This scheme must be reviewed and
approved by the segment designer to avoid overstressing the segments.

10.3.5.2 Nylon slings

Other methods to move segments with cranes include nylon slings and dedicated segment
pickers. Nylon slings have flexibility to pick and move any number of segments but the
placement of the sling is slow, requires attention to sling placement and the segments’ sharp
edges will damage the sling unless softeners are used. Consistent correct placement of slings
is required to avoid segment damage. Considering that overhead loads should be moved as
safely as possible, slings are a last choice for moving segments.

10.3.5.3 Segment pickers

Dedicated segment lifting devices, commonly known as “pickers” come in several styles,
including the A-​frame, the side picker and the powered side picker. All require someone to
assist with placement of the picker to avoid striking and damaging the segment edges.
The A-​frame segment picker can lift a whole stack of segments. A compressed air tank is
integral with the picker and actuates two air cylinders to move the lifting arms in and out.
 Transportation, handling and installation 597

Figure 10.6 HDPE lined sleeves over the forks.

Note: Sleeves prevent damage to segments.

The system is simple and rugged but requires clearance at the segment ends for the place-
ment and removal of the picker. The operation of an A-​frame picker is shown in Figure 10.7.
The side lifting picker lifts a stack of segments by sliding in lifting fingers from the sides.
This eliminates the need for end clearance but requires additional side clearance. The picker
is operated by the motion of the crane on picker to actuate the lifting fingers and move them
in and out of position. Figure 10.7 shows such a picker in operation.
A variation of the side lifting picker is the powered side picker where the crane oper-
ator can operate the grab mechanism remotely by an electric cable suspended from a crane
overhead reel. The electrically actuated picker has the disadvantage that the picker must
be disconnected electrically every time the crane is used for another purpose. The discon-
nection process is at a height that requires a ladder and requires care to avoid damaging
the connector. The electrical power also adds another complication to the picker. Both the
A-​frame and the side lifting picker work equally well. Clearances at loading and unloading
points will decide which is best for the job.

10.4 TRANSPORTING SEGMENTS TO THE TUNNEL HEADING

10.4.1 Preparing the load for transport into the tunnel

Whether a tunnel has portal (drive-​in) entry or shaft entry determines how the segments get
loaded onto a transport vehicle. For portal entry there is more flexibility, and the segment
transporter can be loaded by a forklift or crane. Portal entry allows a forklift to bring the
segments from a remote location and does not require storage of segments immediately
adjacent to the portal. A crane is the only way to load a transporter at the shaft bottom
(Figure 10.8).
598 Handbook of Precast Segmental Tunnel Lining Systems

Figure 10.7 Mechanically operated side picker.

Note: Side lifting picker hoisting segments from a truck bed and placing in staging area.

10.4.2 Segment transportation on mobile equipment to the

tunnel heading
In the past, transportation of tunnel segments to the TBM tunnel heading was always done
by rail. Rubber tired transport vehicles also known as multi-​service vehicles are now mak-
ing inroads into tunnel operations. Many tunneling companies have legacy equipment that
includes locomotives and rolling stock and are reluctant to abandon it. Rail equipment is
still the best option for smaller diameter tunnels.
An important step for either mode of equipment is to check the available clearances on
the TBM trailing gear for the segment stack on the transporter. If the tunnel is of substan-
tial length, there will be a passing place in the tunnel and there will also be a need for a
passing place at the service shaft. The travel lanes should be laid out to allow the passage
of the widest loads on both lanes. Segment transporter loads may be the widest loads. Some
contractors like to save time by preinstalling the ring-​to-​ring connectors before sending the
transporters into the tunnel. It is important to makeake sure the extra width is accounted
for in the lane layout.
The segment stack must be accurately positioned on the transport vehicle. Off-​center
loading can create interference problems in the tunnel when passing conveyor booster drive,
slurry pumps or other trains at passing places. The segments (and the dowels if preinstalled)
must clear the trailing gear up to the unloading point. If a quick unloader is used, consist-
ency of placement on the transport vehicle assures that the segment stack can be unloaded
promptly with every delivery.
 Transportation, handling and installation 599

Figure 10.8 Loading a segment rail car.

Note: Rail car is at shaft bottom.

10.4.3 Rail transport

Rail borne set-​ups often use cars specially designed for transporting segments with a
drop deck between the wheels to provide segment height clearance into the trailing gear
(Figure 10.9).
The train is pushed into the tunnel with the locomotive located on the shaft or portal
end with the segment cars up front. Visibility is poor in the TBM direction, and the
operator usually relies on a video camera system to see the path ahead. These video
systems are a maintenance headache. The train usually has flat cars and grout cars
between the loco and the segment cars. A train usually brings in enough supplies for the
TBM advance equal to the width of one ring. Sections of track, utility pipe and greases
also make the trip with the segments. Derails of full segment cars can occur and are an
extreme hazard since the segments are just stacked on top of each other. A written pro-
cedure must be developed and tested ahead of time to rerail a segment car. The rerail
equipment must be ready and available with a written plan and trained personnel before
a derail occurs.
600 Handbook of Precast Segmental Tunnel Lining Systems

Figure 10.9 Segment rail cars

Note: This style rail car works better with side lifting segment picker.

10.4.4 Multi-​service vehicles

Multi-​service vehicles (MSV) are a newer technology with more flexibility than conven-
tional small gauge rail. As shown in Figures 10.10 and 10.11, rubber tired MSVs have an
operator’s cabin at each end of the vehicle providing superior visibility. The MSV can pull
multiple trailers over grades up to 20% and has a tight turning radius with multiple steer-
able driven wheels. The rubber tires along with good brakes mean the MSV can also handle
steep grades better than a rail system. Installing passing places inside the tunnel is less com-
plicated than with a rail system.
One still needs to pay attention to segment height clearance at the trailing gear. MSVs can
be supplied with rotating cabs that lay on their sides or folding cabs to add clearance for
segment unloading on the trailing gear. On the surface, a MSV can use conventional roads
to get to the loading zone or the maintenance garage. This mobility allows the MSV to move
to a segment staging area away from the tunnel portal. MSVs have higher air ventilation
requirements than trains carrying equivalent amounts of material.

10.4.5 Special conditions for inclines and declines

Standard small gauge locomotives are not suited for grades over 5%. Any operational
grades over 2% should have brakes on all the rail cars. Even operating at 5% can present an
extreme hazard if the operator is not well trained and highly experienced. Constant main-
tenance is required for all rolling stock. Train systems have been developed with hydraulic
power going to all the railcar wheels to provide both traction and braking, however, the
coefficient of friction between the steel wheel and rail is still the governing factor as to the
ability of the loco to safely handle the load.
MSVs can safely handle continuous grades up to 20% but segment stack stability should
be checked. Beyond 20%, one should start looking at specialized cable hoisting systems to
supply TBMs with segments and supplies.
 Transportation, handling and installation 601

Figure 10.10 Multi-​s ervice vehicle from the low-​p rofile operator’s cab end.
Note: Front end of MSV going into the tunnel, unloaded segments pass over the cab.
Photo courtesy of Techni-​M étal Systèmes division of Herrenknecht Tunneling Systems.

Figure 10.11 Multi-​s ervice vehicle from the power unit end.

Note: Tail end of the MSV going into the tunnel becomes the front on the way out.
Photo courtesy of Techni-​M étal Systèmes division of Herrenknecht Tunneling Systems.
602 Handbook of Precast Segmental Tunnel Lining Systems

10.5 SEGMENT HANDLING AND TRANSFER WITHIN THE TBM

10.5.1 Segment supply and unloading within the TBM trailing gear
For single shield TBMs, the tunnel is built by erecting a ring and then thrusting off the ring
to advance the TBM. Since the construction of the tunnel is usually on the project critical
path, any interruption of the ring erection delays the entire project. Every effort must be
made to have the segments immediately ready for ring assembly once the push is completed.
A double shield TBM does not have to wait on the ring to be assembled to continue mov-
ing but it consumes rings at a faster rate, so an efficient segment delivery system is just
as important. As previously discussed, the segments need to be unloaded and staged in a
defined order.

10.5.2 Dependency on TBM trailing gear configuration

When people think of a TBM, they think of the business end, the cutterhead and shield,
but there is a substantial amount of support equipment towed behind the TBM on movable
open frame type structures. These movable structures are called trailing gear, gantries or
back-​up trailers. The trailing gear carries everything required to support TBM operations
at the heading including the operator’s cab, muck removal, high voltage transformers,
ventilation, utility extensions, grease systems, ground conditioning and segment handling.
When the muck is removed by a tunnel conveyor, the trailing gear can be fairly short,
on the order of 100 m (330 feet), because no muck car trains need to be accommodated.
Slurry TBMs also tend to have short trailing gear as muck removal is via steel pipes with
pumps to get the excavated material to the surface. For a muck car system where the muck
trains are loaded and shuttled on the trailing gear, the trailing gear can be up to 200 m
(660 feet) long.
The trailing gear will generally be either open floor or closed floor. As the name implies,
open floor trailing gear has no floor, and the segment transporter stays on the tunnel invert
while the trailing gear continues to move past it. The trailing gear moves on wheel bogies,
steel skids or short sections of wide gauge track leapfrogged from the back of the trailing
gear to the front. This open floor trailing gear is popular for small diameter tunnels where
space is at a premium or in situations where segments stacked on a transporter would not
otherwise clear the trailing gear. MSVs transporting segments tend to have higher clearance
requirements than rail transport and may require an open deck while a rail system moving
the same segments could fit on a closed deck trailing gear.
When the TBM is advancing the trailing gear is also moving, as much as 50 to 75 mm (2
to 3 inches) per minute. Unloading using a moving trailing gear is manageable but requires
more attention than working on a closed floor trailing gear. A grout car connection to a
holding tank on the trailing gear could be put under stress once the connecting hose is fully
extended. Something resting on the moving trailing gear could come into contact with the
parked transporter and damage the equipment.
The closed floor system has a drive-​on deck above the tunnel invert. The drive-​on deck is
accessed by a ramp at the tail end of the trailing gear. When the segment transporter is on
this deck, the parked transporter goes along for the ride as the TBM pulls the trailing gear.
The trailing gear can be set up to move on track laid in the tunnel invert behind the TBM
as is normally the case for rail-​borne service to the TBM, or the trailing gear can move on
urethane clad wheel steerable bogies that run directly on the tunnel lining. The obvious
advantage to the closed deck trailing gear is that once the train or MSV is on the trailing
 Transportation, handling and installation 603

gear, the train or MSV is stationary relative to the trailing gear and unloading of segment
and supplies is simplified. Given a choice, most contractors prefer a closed floor system.
Once the segments arrive at the TBM trailing gear, the segments along with all the other
supplies need to be unloaded and delivered to the correct location. The quicker the trans-
port train or MSV can be unloaded, the sooner the train or MSV can go back to the service
shaft for another load of supplies. Eventually, if the tunnel is long enough, a second sup-
ply train or MSV, and perhaps even a third will need to be added to the supply chain. The
overall cycle time of the train to load at shaft-​travel to TBM-​unload at TBM-​travel to shaft
will dictate when another transporter must be added to the system. For intermediate length
tunnels in the 3,500 to 4,500 m (12,000 to 15,000 feet) range, a quick unloading system
may eliminate the need to run a second transporter and install a passing place in the tunnel.
For that reason, many clever schemes have been developed to speed up segment handling.

10.5.3 Segment lifting attachments

Tunnel liner segments do not come with handles. They are very smooth on the inside face
(intrados) and they must minimize exposure to long-​term corrosion so there is no exposed
bare metal that a crane or erector could grab. Despite that, segments need to be picked-​up
and manipulated. Devices designed to get a grip on a segment fall into two categories, mech-
anical and vacuum lifters.

10.5.3.1 Mechanical lifting

Mechanical lifters work by temporarily attaching a steel fixture to the segment. The segment
is cast with a screw pocket that can also function as a grout port. A steel fixture is screwed
into the grout port and that fixture provides a means for a crane or segment erector to grab
the segment. The steel fixtures are screwed into each segment just before they are unloaded
by the monorail crane located on the trailing gear of the TBM. The fixture stays in the seg-
ment until the ring is erected, then the fixture is removed and reused on the next set of seg-
ments. Figure 10.12 shows such a fixture in a segment and Figures 10.13 and 10.14 show
the details of the fixture.
The mechanical lifter is a simple system, but it has some disadvantages. First the threaded
hole must be cast in every segment. Second, a fixture must be inserted into every segment
and then the fixture must be removed from every segment. This exposes the worker to the
hazard of handling and installing a heavy fixture for every segment. The fixture must be
removed at height, which could result in dropping the fixture and injuring someone below.
The fixtures go through hundreds or thousands of cycles and must be periodically inspected
and replaced as necessary.
Mechanical fixtures with single point lifting have limitations when it comes to larger
diameter tunnels with heavier segments. Typical load limitation for a mechanical single
point lifting system is around 3 tons. Heavier segments may require additional reinforce-
ment for lifting stresses that can be eliminated with use of vacuum lifters described in the
next section.

10.5.3.2 Vacuum lifting

Most TBMs above 5 m (16.5 feet) diameter use vacuum lifting systems. A steel plate is
rolled to match the curvature of the segment intrados. The plate has a foam rubber gasket
604 Handbook of Precast Segmental Tunnel Lining Systems

Figure 10.12 Segment with lifting fixture installed.

Note: Fixture installed at TBM and removed after ring erection.

Figure 10.13 Lifting fixtures.

Note: Reused many times. Must be inspected daily.
 Transportation, handling and installation 605

Figure 10.14 Threaded plastic lifting insert in segment.

Note: Insert may also serve as a grout port.

around the perimeter and a port to pump out the air and produce a vacuum. The plate is
suspended from the monorail crane and is set down on top of the segment. A vacuum is
applied, and the segment is lifted with nothing more than atmospheric pressure exerted on
the segment. The lifting place is designed to hold a vacuum for 30 minutes in the event of a
power failure and the plate will not attempt to lift the segment until a specified vacuum is
reached. The physics are sound, but it is still amazing to see a segment weighing tons being
lifted by the application of a vacuum. Vacuum erectors have an excellent safety record.
Figure 10.15 shows a vacuum lifting plate arrangement under a monorail crane. The plate
has three vacuum zones. All three are used to lift a segment but only the center zone is used
to lift a key.
Segments made for vacuum lifting do not have a threaded insert for as a fixture, but they
do have to have holes cast in them for shear cones. Shear cones are not required for straight
vertical lifts but they are required for holding the segment in a subvertical or vertical plane.
The foam gasket on the plate seals off the segment for air evacuation but it is not capable
of transmitting shear. Holes cast in the segment line up with a matching plug. The plug is a
wear part with a steel core and a nylon outer shell. The shear cones also serve as an align-
ment tool to make sure the vacuum plate is correctly and consistently positioned over the
segment before the vacuum is applied.
The vacuum lifting plate is more complicated and expensive but has been proved to be
reliable. The plate does not require a worker to repeatedly handle a heavy fixture or retrieve
the fixture from an awkward position. There is no need to cast a threaded insert in the
segment unless a grout port is required. There is a need to cast shear cone pockets into the
segment, but the casting process is simpler compared to a threaded insert. One disadvan-
tage with a vacuum lifting plate occurs when handling a cracked segment. Occasionally a
cracked segment may have to be removed from the heading. The vacuum erector will not
work if the crack runs under a vacuum pickup zone. In those cases, the segment must be
lifted using chains or slings and the process is very slow. A whole shift can be lost moving a
damaged segment out of the heading.
In the past, precast concrete segments were reinforced exclusively with steel reinforcing
bar cages. The current trend is to use precast segments made with steel fibers. The steel
606 Handbook of Precast Segmental Tunnel Lining Systems

Figure 10.15 Vacuum lifting plate with gaskets.

Note: Plate has three lifting zones to allow for lifting of smaller key. Two shear cones visible in
center zone.

fiber segments are more susceptible to cracking if lifted from a single point. Vacuum lifting
distributes the load and is less likely to damage segments.

10.5.4 Staging segments and supplying the TBM erector

When the supply transporter (train or MSV) arrives at the trailing gear, the segments are
at the front of the transporter, so the transporter moves as far ahead on the trailing gear as
possible. The segments may be unloaded one by one using an overhead monorail crane or
they may be unloaded as a stack with a quick off-​loader followed by subsequent handling
by the monorail crane.
As previously shown in Figure 10.9, the segments travel through the tunnel with the nar­
row part of the segment stack set across the width of the transport trailer and the long part
of the segment stack set in the direction of travel. When the segments are unloaded, they
need to be rotated 90 degrees to be in the proper orientation for hand-​off to the ring erector.
In some cases, the segment cars have a turntable, and the entire stack is rotated upon arrival
at the trailing gear and before unloading. In other cases, the segments are unloaded one at
a time, transported to the unloading zone, and rotated before being released by the segment
crane located on the trailing gear of the TBM.

10.5.4.1 Individual segment unloading

If the segments are unloaded one at a time from the transporter (Figure 10.16), they can be
staged for efficient delivery to the segment erector using a segment feeder (Figure 10.17).
 Transportation, handling and installation 607

Figure 10.16 Segment moving from transporter to segment feeder.

Note: Vacuum lifting plate and monorail crane overhead.

Figure 10.17 Segments on a segment feeder.

Note: Segments are passing under the #1 trailing gear car and moving towards the ring erector
in foreground.
608 Handbook of Precast Segmental Tunnel Lining Systems

The segment feeder is essentially a heavy-​duty conveyor. The segments are unloaded, rotated
90°, and placed on the segment feeder in the order required by the segment erector. The
feeder brings the segments to the ring erector for pick-​up and placement. With a train, the
segment cars may be detached from the rest of the train so that the other supplies can be
unloaded simultaneously with segment unloading. Then the empty train is reassembled and
pulled back to the service shaft.

10.5.4.2 Segment stack unloading

The stack unloading method vertically lifts the entire stack of segments from the transporter
so that the transporter can pull back and be free of the load. Then the stack is lowered so
that a monorail overhead crane can access the stack and move the pieces one by one to the
tunnel heading for installation. The stacked segments serve as the staging area. Then they
are handed over to a delivery cradle or the segment feeder for the segment erector to pick
up as shown in Figures 10.18 and 10.19. The stack unloading method has the advantage
that multiple stacks of segments can be lifted and the train or MSV can be freed in less than
a minute. This advantage is only a benefit if the rest of the train can be unloaded and freed
at the same time. If, for example, the TBM must wait for muck cars to be loaded before the
train can leave, the advantage is lost.

Figure 10.18 Stack unloading station on trailing gear.

Note: Transporter delivers segment stack here and leaves as soon as the stack is lifted. Crane
has lifted top segment and will move to segment feeder.
 Transportation, handling and installation 609

Figure 10.19 Segment moved from stack unloader to segment feeder.

Note: The monorail lifting crane has rotated the segment 90° and set it on the hand-​o ff cradle
to the erector. Vacuum plate in the foreground at top.

10.6 RING ERECTION WITHIN THE TBM TAIL SHIELD

The ring building cycle in the simplest terms is as follows:

• Advance the TBM the length of one liner ring

• Select the ring position
• Erect the ring

In this section we will discuss how and why a particular ring erection position is selected
and the process to erect a ring. We will also discuss problems to avoid.

10.6.1 Steering the TBM and steering the segments

A shielded TBM with precast liner segments assembled in the tail requires the steering of the
TBM to follow the theoretical tunnel alignment and the steering of the tunnel liner to follow
the TBM. In a single shield TBM, the steering of the TBM is done by differential thrust off
the liner system and use of shield articulation (if the shield has an articulation). A double
shield TBM could be steered by a reaction off the grippers or off the liner ring. In either case,
the tunnel liner segments assembled in the tail of the TBM must follow the path of the TBM,
even if the TBM is off course.
Selection of the position of the next ring to be assembled can be done manually by the
field engineer or by a guidance system. Most TBMs use a guidance system to select ring
610 Handbook of Precast Segmental Tunnel Lining Systems

position but the principles are the same for manual selection. The selection of the next ring
position to be erected in TBM tail shield needs to consider:

• Concentricity of the tunnel ring inside the tail shield

• Direction of the TBM in the next few rings (theoretical and actual)
• Differential lead between the thrust (propel) cylinders
• Allowed ring positions

10.6.2 Importance of keeping the liner ring centered in the tail shield
For shielded TBMs, the ring is necessarily assembled inside the protection of the tailshield
and the ring outside diameter (extrados) is smaller than the tail shield inner diameter. If the
TBM maintains a pressurized face, the annular gap between the tail shield and the liner ring
must be sealed to maintain a pressurized zone around the TBM. The annular gap between
the two is sealed by up to four rows of wire brushes and a grease sealing system. There
are usually two or three rings of wire brushes (Figures 10.20 and 10.21) and the space in
between the rings is filled with a biodegradable grease that does the actual sealing. Part of
the brush system is the brush holder, which is a curved steel ring around the inside of the tail
shield to which the wire brushes are attached.
The sealing system can tolerate an eccentricity of about 50 mm (2 inches) between the
precast liner ring and the shield. The exact tolerance depends on the design of the TBM.
Once that is exceeded, a portion of the liner ring will come in hard contact with the brush

Figure 10.20 Three rows of tailshield brushes.

Note: The brushes are welded in place to form complete rings inside the tail shield.
 Transportation, handling and installation 611

Figure 10.21 Tail seal brushes in section.

Note: The spaces between the brushes are filled with grease and the brushes are flattened as
they press against the segments.

holder and damage to the segment is likely to occur as the liner ring is pushed-​out of the tail
shield, often described as being “iron-​bound”. Keeping the assembled ring in the center of
the tail shield reduces the possibility of segment damage.
As the liner ring is pushed out the back of the TBM, the gap between the ring extrados
and the tail shield is measured, usually at springline and at the top arch and invert. Each
tunnel liner ring has a known offset for the length of the ring to the tunnel longitudinal axis
and this is used to steer the ring back towards the shield center. The gap is usually meas-
ured manually but may be measured electronically if the guidance system has the necessary
equipment.

10.6.3 The tunnel alignment ahead

The direction of the TBM for the succeeding rings also must be considered. Is the TBM
going straight, making a little steering correction, or going into an extended curve? What
is the radius of the curve? The TBM guidance system will know the theoretical alignment
of the tunnel ahead and will also know the position of the TBM and where it is currently
headed. The guidance system can usually calculate ring positions three or four rings in
advance. Some systems can calculate up to eight rings in advance. For a universal ring, all
the rings are the same and the key may go anywhere. For a left-​right-​parallel ring system,
the correct order of rings delivered to the heading must be determined to keep the segment
key above the springline.
612 Handbook of Precast Segmental Tunnel Lining Systems

When the TBM is on a straight section (tangent) or in a circular curve, the selection of
rings soon becomes established and routine. Errors in TBM tunnel alignment and damage
to liner rings usually occur at the transitions between tangents and curves. If the tunnel is
designed with tangents and circular curves only, as most sewer tunnels are, the operator will
get in trouble if they blindly follow the guidance system. If they wait until the TBM is at the
end of a tangent before beginning a turn, they will start the turn too late and will usually
over-​correct to get the TBM back on course. This often causes damage to the liner rings in
the tail shield, which cannot follow an abrupt change in TBM heading. That same problem
can happen at the end of a circular curve.
If the TBM operator waits for the end of a circular curve to make the transition to a tan-
gent, the operator will again be too late and will oversteer the TBM to get back on track.
An experienced TBM operator will start and end a curve early thus putting in a transition
in and out of the curve. Rather than let the TBM operator freehand the alignment, the
situation can be improved by introducing spiral curve transitions at the exit and entry to a
circular curve within the guidance system design tunnel alignment. The TBM is less likely to
go off alignment following railroad and highway tunnel design alignments because they are
designed with spiral curves for trains and cars.

10.6.4 Differential thrust cylinder extension (lead)

The differential extension (lead) between the thrust cylinders must be minimized.
Theoretically, in a straight tunnel with parallel ring faces, the TBM thrust cylinders would
all extend the same amount as the TBM makes an advance. But since the TBM must make
corrections and go around curves using tapered segments, the extension of the thrust cyl-
inders can vary across the face of the ring. Minimizing the differential extension between
the thrust cylinders reduces the differential loading on the ring and reduces ring damage.
The cylinder extensions can be measured manually but most TBMs have electronic exten-
someters that can give the information directly to the TBM operator and the guidance
system.

10.6.5 Allowed ring positions

The precast tunnel liner system may be a universal ring or a left-​right-​parallel system. In
either case, every ring has a taper and can be made to follow the TBM by rotating the ring to
project the centerline of the ring in the desired direction. The rings are cast with connector
sockets for the ring-​to-​ring interfaces so that each ring connects to the previous ring in a pre-
defined manner (indexes). The connection may be with bolts or shear connectors. Typically,
the defined positions match the number of thrust cylinder pads, so a TBM with 19 thrust
cylinder pads will have a lining with 19 possible positions of rotation. Not all the positions
will be allowed for use because some of the positions will result in four corners (cross joints)
of segments meeting in a cruciform joint (two from the previously placed ring and two from
the newly placed ring). Cruciform joints (Figure 10.22) are discouraged because they result
in four corner gaskets being forced into a single point and the pressure could spall the con-
crete on the edges of the segments next to the gaskets.
Most guidance system programs can be adjusted to put different relative weights on the
importance of ring concentricity, alignment ahead and thrust cylinder lead for making a
ring selection. If the ring positions are being calculated manually, the field engineer needs to
make the same decisions.
 Transportation, handling and installation 613

Figure 10.22 Cruciform joint.

Note: This cruciform joint occurred when a ring was erected at a prohibited position.

10.6.6 Ring erection cycle

Once the ring position is selected, the ring erection begins. The ring building sequence is
always the same for every ring, just the starting position is different. If built starting with
the segment opposite the key, then the ring is built up on either side of the first piece placed
alternating on either side until only the key piece remains. If starting with the ring next to
the key, the pieces are built sequentially all the way around the tail shield until only the
key remains. To avoid confusion, we will call the person operating the ring erector the ring
builder.
Erecting and maintaining the liner ring in a circular configuration is important to prevent
damage and resist long term external forces. Erecting a ring properly (Figure 10.23) requires
practice and patience. Learning to build a quality ring takes time. Since one cannot prac-
tice building a ring before the TBM starts advancing, an experienced ring builder should be
present at the start of tunneling to train novice ring builders. Ring build times at the start
of tunneling will be long (up to an hour) but improve as the ring builder and their assistant
learn the routine. The one pass liner ring is a final product of the tunnel process and build-
ing poor-​quality rings can lead to cracking, spalling and gasket damage.
Most damage occurs to a liner ring during the erection and the push-​out from the tail shield
during TBM advance. The ring must be erected with the gaskets properly compressed to
allow the key piece (the last piece) to fit into the open slot. If the ring is not built correctly,
the slot will be too small and the key piece or the pieces on either side, most likely the gas-
kets, will be damaged. Because the assembled ring is made from segments and the segments
are placed individually, sloppy ring assembly can result in inconsistent gaps between the
segment edges which can result in spalling.
The best practice is to use a gap measuring tool and record the gaps at the edges of the
segments. Figure 10.24 shows a simple, rugged, gap measuring tool machined from steel
where each scribed line marks a 2 mm (roughly 1/​16 inch) increase in diameter from 4 to
614 Handbook of Precast Segmental Tunnel Lining Systems

Figure 10.23 Well built rings.

Note: A nice set of well-​b uilt rings with even gaps between segments and no lipping between rings.

Figure 10.24 Joint gap measuring tool.

Note: Measuring and recording gaps between rings informs the ring builder about the quality of
the constructed ring.

16 mm (1/​8 to 5/​8 inches). Consistent gaps all the way around the ring is an indication of a
circular ring. The ring builder can see how he is performing and improve accordingly.
A sample report is shown in Figure 10.25. A report should be prepared for every ring
describing whether any damage occurred during installation or from the advance afterward.
 Transportation, handling and installation 615

Figure 10.25 Sample segmental ring build report.

Note: Consistent reporting on ring build quality provides feedback to the ring erector operator.
616 Handbook of Precast Segmental Tunnel Lining Systems

Often this is required by the specifications, but it is also good practice to record the installa-
tion of every ring. If there is a rash of cracking, the reports can be used to determine whether
there are any systemic failures, which can be eliminated. The gaskets may be required to
resist water pressures up to 20 bar (290 psi), therefore good alignment between the gaskets
is necessary. The gaskets are designed to tolerate a certain amount of offset, which depends
on the design of the gasket, but the ring builder should not rely on that tolerance.

10.6.7 Ring assembly aids

The ring should be built with a minimum of offset (also called lipping) from ring-​to-​ring or
between segments. Various alignment guides for ring erection are covered in more detail in
other chapters. Segments usually connect ring-​to-​ring using bolts or snap-​together connec-
tors to keep the gaskets compressed. These connectors are plastic encased steel dowels with
circular ridges that interlock with the cast-​in-​place sockets. The connectors also provide
alignment with a cone shaped pocket cast in the segment with a matching cone built into
the plastic connector to ensure good alignment (Figures 10.26 and 10.27).
Bolted ring-​to-​ring connections usually get an alignment assist from a matching key
and pocket system cast into the segments. Segment-​to-​segment connections within a ring
are often assisted by an alignment rod with a matching pocket in the abutting segment
(Figures 10.28 and 10.29).

10.6.8 Advancing TBM and preparing for ring build

The typical cycle starts with the TBM beginning excavation for another ring. As the TBM is
being advanced, segments are being unloaded and staged for use by the ring erector. If the
segments have ring-​to-​ring connector dowels, these are installed. The ring builder and his
assistant make sure the shield invert is mucked out and the connecting hardware is ready. As
the TBM nears the end of shove, the field engineer takes measurements of the segment gap
at the arch, invert and both sides of springline. If a guidance system is used, they relay the
information to the TBM operator who inputs the information. The guidance system already
knows the lead on the thrust cylinders and the TBM position with regards to the tunnel
alignment and the alignment ahead.

Figure 10.26 Connecting dowel.

Note: Connecting dowels instead of bolts are often used for ring-​t o-​r ing connections.
 Transportation, handling and installation 617

Figure 10.27 Matching dowel hole.

Note: Plastic insert with recess.

Figure 10.28 Guide rod.

Note: Fits in longitudinal joint between segments within a ring. Usually glued in-​p lace at the
surface staging area.

The guidance system calculates the best ring position, cross-​checks this selection against a
table of forbidden positions. If that position is forbidden, then the program selects the closest
allowed position. The TBM operator or field engineer relays this information to the ring
erector and the ring assembly starts for the designated position. Immediately upon com-
pleting a ring, the ring erector notifies the TBM operator. The position of the completed ring
618 Handbook of Precast Segmental Tunnel Lining Systems

Figure 10.29 Guide rod pocket.

Note: Guide rod pockets are cast on each segment radial edge.

is put into the guidance system and another cycle begins. The field engineer, TBM operator
and ring builder should always check that the guidance system selection is reasonable. They
have the option to install the ring in a different position if necessary and input the selection
into the guidance system.

10.6.9 Erecting the segmental liner ring

At the start of a ring build, all the thrust cylinders in the shield circle are pushing against
the previous ring (Figure 10.30). As the segment is picked-​up and brought to position, the
thrust cylinders blocking only that segment needing to be installed are retracted. Once a
segment is placed, the thrust cylinders for that segment are extended to hold the segment
in place using the same pressure as the other thrust cylinders. A universal ring can be built
starting with the first segment in the arch using the thrust cylinders to hold the segment in-​
place. Pressurized face TBMs require that enough of the thrust cylinders be in contact with
the liner ring to prevent the TBM from being pushed back from the face. This accidental
movement can damage the sealing brushes and cause injuries.
The segment erector picks-​up each piece from the invert and rotates the segment up to the
correct orientation and then slides the segment into position against the last segment and the
previously built ring (Figure 10.31).
The erector operator has a control box that controls the segment erector. Early segment
erector control boxes were connected to the hydraulic controls with an electrical cable.
Modern segment erectors use a wireless chest mounted remote control unit that allows the
operator to move around to the best position for observing the segment without worrying
about a cord (Figure 10.32).
 Transportation, handling and installation 619

Figure 10.30 Thrust cylinders stroked out.

Note: The TBM has advanced one ring and a new ring erection cycle begins.

On a large diameter TBM where the top arch segments could be 6 m (20 feet) or more
off the invert, the operator needs to climb a purpose-​built platform to see up close how the
segment is being placed. The operator will have an assistant who can call out the position
of the segment on the far side of the operator. As the ring is erected, the assistant will install
the bolts (if bolts are used) and will tighten them with an air powered wrench. The ring
builder and his assistant are often not in clear sight of each other and must communicate
constantly to make sure the assistant is not pinched or crushed by the erector or segment
being installed.

10.6.10 Forces on the TBM segment erector

The segment erector will handle the segment using either a mechanical or vacuum oper-
ated gripping system, the same type of system that was used to move the segments from the
transporter to the segment erector. The difference here is that the segment will be hoisted
from the invert in the level position, it could be brought to the vertical or overhead position
and then positioned up against the previously placed ring. The ring erector plate may have
a laser mounted on it that aligns with a cast-​in-​place mark on the segment intrados that
620 Handbook of Precast Segmental Tunnel Lining Systems

Figure 10.31 Ring builder’s assistant.

Note: The assistant calls out the segment position where the ring builder is cannot see. The TBM
thrust cylinders are retracted for this segment only.

Figure 10.32 Wireless ring erector controls.

Note: Wireless controls allow ring builder to best position himself during ring erection.
 Transportation, handling and installation 621

provides a guide for the ring builder allowing him to quickly align the plate and shear cones
with the segment.
If the segments use an interlocking ring-​to-​ring dowel system, a certain amount of force
will be required to connect the dowels. Additionally, the segment must be forced against the
preceding segment to compress the gasket. The mechanical gripping system uses the same
fixture that was installed for unloading the segment from the transporter to transmit the
thrusting force of the erector arm into the segment. The vacuum plate requires the presence
of shear cones to transmit the thrust force because the foam gasket seal use to hold the
vacuum cannot handle the shear forces.

10.6.11 Ring damage during TBM advance

As the ring is pushed-​out the tail shield, the ring is compressed by the grout being injected
into the annular space around the ring. The ring gaskets compress and the ring changes
shape slightly and continuously until the entire segment is embedded in grout. With a pres-
surized face TBM, the compression is significant enough that the erection bolts will come
loose and need to be retightened. For many designs, the erection bolts are removed after the
grout sets. Some liner ring designs do not use any bolts at all. This is also a time when cracks
can develop from uneven loading of the thrust cylinder pads on the segments.
If the previous ring has an offset between segments, the offset will function as a high point
and cause a crack to form between thrust cylinder pads. Rings that were erected with the
intrados edges touching can also spall (Figure 10.33). Rings not centered in the tailshield
may strike the tail brush holders and become cracked. Gasket damage also becomes appar-
ent when the ring is exposed to pressurized groundwater (Figure 10.34). At this point the
ring cannot be removed and the damage must be repaired by other means such as urethane
grout injection. Damage to liner rings must be avoided as this is the final product for one-​
pass linings.

Figure 10.33 Spalling segment edges and corners.

Note: Spalling can occur when liner rings squat or the segments have an offset.
622 Handbook of Precast Segmental Tunnel Lining Systems

Figure 10.34 Leaking gasketed joint.

Note: Leaks can be caused by damaged gaskets or by concrete cracks in the gasket seat area.

10.7 IMPORTANCE OF ANNULAR GROUT SUPPORT AROUND

SEGMENTAL LINER RINGS
Just as important as the ring erection is the annular grout injection that takes place as the
assembled ring exits the tailshield. Annular grout is covered in detail in Chapter 11. Because
the TBM is necessarily a larger diameter than the tunnel liner ring that is assembled inside,
there will be an annular gap in the 150 mm (6 inches) range between the ground cut by the
TBM and the tunnel liner ring extrados. This gap needs to be filled immediately to prevent
ground settlement and to evenly support the ring. The most effective measure is to inject
grout through the tail shield trailing edge at multiple ports to ensure that the annular gap is
always filled. The gap can also be filled by injecting grout through ports cast into the liner
rings. When the TBM is operating with a pressurized face, the annular grout is injected with
a pressure of 1 to 2 bar (15 to 30 psi) above the face pressure. This prevents ground water
flow from following the tunnel lining and washing out the grout.
The TBM trailing gear runs over the lining with track for roller bogies. The first set of
wheels runs over the lining within about five rings of the tunnel heading so the grout needs
to provide adequate support by then. Pressurized face TBMs use a grout with sand and pea
gravel or a two-​part grout. The sand and pea gravel cementitious grout relies on the water
being squeezed out of the grout matrix to provide immediate support. The two-​part grout
uses a cement and bentonite mixture that is catalyzed at the point of injection with a sodium
silicate-​based accelerator.
The two-​part grouting system may have the cement/​bentonite mix pump through a 50 or
75 mm (2 or 3 inches) diameter pipe from a grout plant on the surface to the TBM. The line
diameter is kept small to maintain full flow through the entire cross-​section of the pipe. The
accelerator may be brought down to the TBM in 1.0 m3 (1.3 CY) containers. Usually there
is a holding tank on the TBM, and the tank should be full prior to each ring advance. The
field engineer should visually check the amount of grout pumped during the advance and
document the information. Verifying the amount of grout placed is a useful confirmation
of the grout pumping instrumentation and provides additional confidence that the grouting
has been correctly done.
 Transportation, handling and installation 623

When the TBM is an open face shield, the problem is that a fluid grout will run towards
the front of the TBM. TBMs can be outfitted with spring steel plates at the tail shield to
prevent forward movement of the grout but these plates wear out and will not work if
rock in the arch or shoulder fall out at the face. In these cases, the solution has been to
inject pea gravel through the segments using shotcrete pots. The gravel provides immediate
support however lipping between rings and squatting does occur. The newer method is to
inject a two-​part grout behind the tail shield that gels within seconds after injection and has
adequate bearing for the first liner ring loaded by the trailing gear.

10.8 SPECIAL CHALLENGES DURING TBM LAUNCH

The TBM must be launched with the first rings built as circular as possible. TBMs launched
from a shaft or portal usually start from a temporary but massive thrust frame. Liner rings
are built inside the tail shield and are pushed out into open air instead of into a confined
space with pressurized annular grout (Figure 10.35). As the liner ring is made up of heavy

Figure 10.35 Segment damaged during TBM launch.

Note: A temporary ring in the launch shaft. Launching is always a tricky process with
abnormal loads.
624 Handbook of Precast Segmental Tunnel Lining Systems

Figure 10.36 Rigid launch frame.

Note: Liner launch rings wedged against the frame at springline to keep the rings circular.

discontinuous segments, the ring tends to squat upon exiting the tail shield. Every ring is
built and connected next to the previous ring so if the first ring squats, then the ring builder
will have to fight the tendency of every subsequent ring to squat.
Different contractors use different schemes at TBM launch to prevent ring squat from
building and external frame to post-​tensioning cables around the rings to provide compres-
sion forces. Figure 10.36 shows a rigid frame with wedges to help round-​up the temporary
liner rings inside the launch shaft. Figure 10.37 shows another TBM launch scheme using
post-​tension cables. Once the TBM is inside the tunnel and grout pressure can be built
up, the rings tend to round up, but it can still take 20 rings to get the squat out. Another
consideration is that the gaskets between the rings are like powerful springs. Although the
ring-​to-​ring connectors should be able to resist the gaskets pushing apart, contractors often
bolt tie bars across the first five rings from the portal to make sure the rings do not move.

10.9 SAFETY CONSIDERATIONS FOR SEGMENT HANDLING AND

INSTALLATION
Handling and installing heavy precast segments present many safety concerns. The equip-
ment used to handle the segments is powerful and has the power to crush, pinch or ampu-
tate limbs. The segments are heavy and overhead lifting always has the potential for a
sudden release of energy.
The minimum number of people required to unload, move and erect the segments should
be used. These people must be properly trained in the use of the equipment and must work
 Transportation, handling and installation 625

Figure 10.37 Post tension of launch rings.

Note: Launch rings are kept circular by installing post tension cables as the TBM advances
through the headwall.

together as a team. Any substitute workers must be trained and watched until they know
the routine. Workers must not stand under overhead loads or in areas where a segment
erector failure could cause injury. Because the segment area has a wide range of motion and
the operator has poor visibility of many areas, visitors should not be allowed in the imme-
diate area during ring erection. The segment handling equipment must be inspected by a
mechanic following a checklist to ensure that all parts of the equipment have been checked.
Inspections should be done at the start of every shift. The segment erection area must have
work platforms that allow the ring builder and their assistant to reach any part of the seg-
ment needed to install bolts and check for alignment of the segment.

10.10 TRENDS IN AUTOMATION OF SEGMENT HANDLING AND

ERECTION
The supply and installation of segment rings is “the” frequent operation in mechanized
tunneling using precast segmental lining. Therefore, the approach towards automation
of individual steps along the entire supply and installation process has moved into the
focus for economical as well as for quality and safety reasons. Particuarly for long tunnels
where thousands of rings need to be installed, an investment in fully automated or operator
assisted systems can become interesting.
Whereas automation in segment production (precast plant) and storage yards has been
introduced for some time, segment handling operations within the TBM has only moved
into the focus in recent years.
626 Handbook of Precast Segmental Tunnel Lining Systems

For the operations within the TBM and gantry area two trends are present:

• Fully automated subsystems

• Operator-​assist systems

The fully automated subsystems can include all operations for segment handling starting
with unloading from the transport vehicle up to the point where the erector picks up the
segment.
However, when introducing automated handling systems, the segments as well as the seg-
ment stack on the supply vehicle must be suitable for automation. Whereas the segments
are high-​precision precast elements that may already incorporate means like bar codes or
similar for automatic identification, increased requirements on geometrical accuracy of the
segment stacks, including the dunnage placed between segments and the stack position on
the supply vehicle are necessary.
If the human operator is eliminated from segment handling within the gantry area, the
possibility for a final human visual check of the individual segments before being supplied
to the erector is no longer available. The visual check and subsequent correction measures
typically includes the detection of damages caused during transport in the tunnel, correct
position of the segment seals, and contamination of the segment surface by dirt, snow or ice
that could cause a malfunction of the vacuum lifter. Instead, an automated inspection done
by artificially intelligent cameras is required. Also, the integration of an automated cleaning
device for the segment intrados is advisable to eliminate the risk of vacuum plate malfunc-
tions caused by surface contamination.

Figure 10.38 Robotic arm at segment unloading station.

Note: The robot arm is removing dunnage when the segments are unstacked at the heading.
Work area is closed-​o ff to workers. Photo courtesy of Herrenknecht Tunneling Systems.
 Transportation, handling and installation 627

Whenever automated systems are installed physical barriers with interlocked access doors
need to be installed to prevent personnel from entering the area where automated hand-
ling operations are in progress. Consequently, such areas within the gantry would only be
human accessible during manual operation mode or maintenance.
Even a fully automated system needs to provide the option for safe manual operation including
the possibility to reverse the transport direction and bring damaged segments back out from the
installed position in the tail shield to the transport vehicle and finally out of the tunnel.
Automated systems require a larger amount of sensor technology and more computer-
ized control systems. The harsh environment in a tunnel heading must be addressed for
component selection and design as well as appropriate qualification and training of on-​site
maintenance personnel. Besides the possibility for fully automated systems, modern sen-
sor technologies provide the possibility to display additional information on the erector
operator’s control panel for the ring erection itself. Such information can include ring build
clearance, or the relative position of the segment currently being placed segment to its neigh-
boring segments.
Measuring and providing such geometrical information to the erector operator electron-
ically can eliminate the need for an assistant ring builder climbing up and down the erector
platforms and performing manual measurements. This combined with the use of boltless
rings would result in faster ring erection and the elimination of a person who is often out of
sight of the ring builder leading to a safer environment.

BIBLIOGRAPHY
For preparation of this chapter, the references were used and are grouped into four categories of
authority published documents, tunnel research and industry association documents, govern-
ment agency documents and authored documents as listed below.

AUTHORITY PUBLISHED DOCUMENTS

AMERICAN CONCRETE INSTITUTE (ACI)

38800 Country Club Drive, Farmington Mills, MI 48331, USA

Telephone: +​1-​248-​848-​3800; Facsimile: +​1-​248-​848-​3800
Website: www.concr​ete.org
• 533.5R-​20 Guide for Precast Concrete Tunnel Segments

TUNNEL RESEARCH AND INDUSTRY ASSOCIATIONS

ASSOCIATION FRANÇAISE DES TUNNELS ET DE L’ESPACE SOUTERRAIN (AFTES)

One Great George Street, SWIP 3AA, London, United Kingdom.

Telephone: 0207 665 2238
Website: www.britis​htun​nell​ing.com
• Recommendations for the Design, Production, and Installation of Segmental Ring Recommendation
for the Design, Sizing and Construction of Precast Concrete Segments Installed at the Rear of a
Tunnel Boring Machine (TBM)s, 2005.

BRITISH TUNNELING SOCIETY

42 Rue Boissiere, 75116 Paris, France

Telephone: +​33 (0)1 85 34 33 20
628 Handbook of Precast Segmental Tunnel Lining Systems

Website: www.aftes.fr
• Tunnel Lining Design Guide, March 2004

DEUTSCHER AUSSCHUSS FUR UNTERIRDISCHES BAUEN E.V. (DAUB)

Mathias-​Bruggen-​Strasse 41, D-​50827, Koln, Deutschland

Telephone: +​49 (221) 59750
Website: www.daub-​ita.de
• Lining Segment Design: Recommendations for the Design, Production, and Installation of
Segmental Rings, 2013.

INTERNATIONAL TUNNELING ASSOCIATION (ITA/​A ITES)

Chemin de Balexert 9, 1219 Chatelaine, Switzerland
Telephone: +​41 (0) 22 547 74 41
Website: www.ita-​aites.org
• ITA Report No.22Guidelines for the Design of Segmental Tunnel Linings.

OSTERREICHISCHE BAUTECHNIK VEREINIGUNG (OBV)

Karlgasse 5, 1040 Wien, Osterreich

Telephone: +​43 (1) 504 15 95
Website: www.bau​tech​nik.pro
• Guideline for Concrete Segmental Lining Systems, 2011.

GOVERNMENT AGENCY DOCUMENTS

Precast Concrete Segmental Liners for Large Diameter Road Tunnels Literature Survey and
Synthesis, FHWA-​ HIF-​20-​
035. U.S. Department of Transportation, Federal Highway
Administration (FHA).
Standard 1926.251–​ Rigging Equipment for Material Handling, U.S. Department of Labor
Occupational Safety and Health Administration (OSHA).
Standard 1926.32–​ Definitions, U.S. Department of Labor Occupational Safety and Health
Administration (OSHA).

AUTHORED DOCUMENTS
Duhme, R.; Rahm, T.; Thewes, M.; and Scheffer, M., 2015, “A Review of Planning Methods for
Logistics in TBM Tunneling”, Proceedings of ITA WTC 2015 Congress and 41st General
Assembly, Lacroma Valamar Congress Center, Dubrovnik, Croatia.
Scientific American, August 9, 1890, “The Great Railway Tunnel Under the St. Clair River, Between
the United States and Canada”, Scientific American Magazine Vol. 63 No. 6, pp. 87–​88.
Chapter 11

Annular backfill grouting of precast

segmental tunnel linings
Lead authors: Shane Yanagisawa, Nestor Garavelli, and Joel Joaquin
Contributing authors: Jeffrey Champa, Richard McLane,
Mirko Martini, Bradley Krumel, Christophe Bragard, and
Dennis Arbour

11.1 INTRODUCTION
This chapter will discuss why annular grout is needed, what the desired characteristics are
for annular grout, and the different types of grouts. Some sample grout mixes are provided
for each type of grout as examples only. Placement methods and example projects are also
discussed.
Making sure the correct mix design is selected prior to tunneling and documenting/​veri-
fying the grout quantities injected during tunneling operations is key for proper execution,
ground loss minimization, cost savings and eliminating rework. Low volume quantities that
are not addressed immediately in the field will result in having to do secondary grouting,
which entails more labor and additional cost. Knowing the job specifications and making
sure the grouting requirements are clear to all parties is extremely important prior to start-
ing any TBM operation.

11.2 NECESSITY FOR ANNULAR GROUT

When advancing a tunnel by means of a shielded TBM and precast tunnel liner system, the
liner is assembled inside the protection of the TBM’s tailshield. The outside diameter of
the assembled liner ring (known as the extrados) is smaller than the diameter of the tunnel
excavated by the TBM cutterhead. As the liner is pushed out the tailshield, a gap (tail void)
develops between the native ground and the extrados of the assembled liner. This gap will
be in the range of 100 to 180 mm (4 to 7 inches) and varies depending on the overcut, the
tail shield thickness, placement of grout injection pipes, and the sealing brush arrangement.
Figure 11.1 shows the sources of the annular gap. This annular gap goes all the way around
the liner ring and must be filled with grout. There are several methods available to fill this
gap and they depend on many factors, which will be discussed in this chapter. We generally
refer to the gap filling material as annular grout but there are many different types of filler
material.
Backfilling of the annular gap is integral and inseparable from the advancement of the
TBM. Although the details of how the TBM cuts the ground at the face may be the initial
focus of many young practitioners, they will soon find that the TBM cannot advance far
without the annular grout. The proper backfilling of the annulus is essential to reducing the
convergence of the excavated ground and stabilizing the segmental liner ring. Injection of
grout into the annular space in soft ground TBMs has a direct influence on the surface settle-
ment behavior. In the short term, modern TBMs operating at high face pressures exert high

DOI: 10.1201/9781032675541-11
630 Handbook of Precast Segmental Tunnel Lining Systems

Figure 11.1 Sources of annular gap surrounding the TBM and segmental lining.
Note: Gap starts with overcut at cutterhead and increases as the mid-​s hield and tail shield typ-
ically have slightly smaller diameters. Largest source of annular gap is between the liner ring and
the tail shield extrados.

thrust loads on the erected lining and TBMs going around curves exert differential pressures
on the segmental lining.

11.3 OBJECTIVES OF ANNULAR GROUT

The annular grout is the interface between the surrounding ground (or rock) and the precast
concrete liner. The objectives of the annular grout should not be confused with the desired
properties of the grout. Objectives of a successful annular grouting system are as follows:

• Ensure uniform continuous contact between the grout and the precast segments.
• Prevent or reduce the movement of the ground, which could result in surface settle-
ment above the tunnel.
• Serve as a sound foundation or bedding for the precast segmental line ring.
• Ensure that the annular gap is filled and that the grout stiffens in a timely manner to
provide ring stability, especially in non-​cohesive soils.
• Avoid movement of the segments generated by the thrust force and shield steering.
• Prevent heave or flotation of the lining.
• Provide adequate support for loads from the advancing TBM trailing gear.
• Protect the precast segments from direct contact with chemically aggressive soil
or rock.
• Provide a barrier against water moving along the tunnel lining extrados.
• Provide a secondary barrier to water and fines moving through damaged or mis-
aligned segment gaskets.
• Distribute internal tunnel pressures that are greater than the ground water pressure in
operating water and sewer tunnels.

The grout must provide uniform continuous contact so that the liner ring can function as a
compression ring. If the void is not completely filled all around the ring, the native ground
 Annular backfill grouting of precast segmental tunnel linings 631

will move towards the ring in soft ground tunnels or leave a void at the crown in hard rock
tunnels and will not load the ring uniformly causing deformation of the ring. Ground loos-
ening above the tunnel and moving into the annular gap will eventually result in surface
settlement. The settlement can be immediate if there are non-​cohesive soils overhead but
will eventually show up even with cohesive soils.
The TBM makes an overcut with the cutterhead that is in the range of 10 to 20 mm (3/​
8 to 3/​4 inches) larger in radius than the front shield. This overcut is required to steer the
TBM and reduce skin friction. The distance from the cutterhead to the tail end of the shield
where the liner ring is pushed out of the tail shield is typically in the range of 8 to 11 m (26
to 33 feet). TBMs with a mid-​shield articulation and a tail shield typically step down or
tapered about 10 mm (0.375 inch) in diameter with each section make the TBM easier to
move after a shutdown when the ground may squeeze in slightly towards the TBM along
the length of the TBM. The ground is held back by the slurry pressure in the case of a slurry
TBM and by the conditioned soil packed around the shield in the case of an earth pres-
sure balance (EPB) machine. Some specifications require the injection of bentonite slurry
around the EPB TBM shield to keep the annulus open. At the end of the TBM tail shield,
the annulus widens to around 130 to 180 mm (5 to 7 inch). This is the gap that needs to be
filled promptly and completely with grout. The potential for further tunnel induced surface
settlement once the TBM tail shield passes a given point is nil provided the annular gap is
filled and the grout is adequately stiff.
Fully embedding the precast liner ring in pressurized grout also keeps the ring circular.
The liner ring has a natural tendency to squat from its own weight if there is not uniform
support from the grout. Modern precast liner ring design allows for a small amount of squat
by relieving the interior and exterior edges so that the segment edges within a ring do not
spall and expose the gasket. Excessive squat will cause spalling at the edges of the opposing
segments and expose the gasket. Uniform loading of the ring will help keep the ring circular
and will assist with the construction of the next ring in the tail shield since the rings are con-
nected to each other via connector pins or bolts.
For hard rock tunnels the problem is the fall-​out of rock blocks created by intersecting
joints or the failure in the quarter arch created by swelling rock. Rock tunnels can produce
water that can wash out the grout. When tunneling through water-​bearing rock formations,
incomplete grouting can lead to the movement along the tunnel liner extrados that increases
as the tunnel progresses. Inadequate grout liner support will lead to ring squat and offsets
between liner rings that will result in damage as the TBM trailing gear advances.
The complete TBM advance cycle for a single shield to excavate and build one ring can
be one hour or less for the typical 6.0 m (20 feet) diameter machine. Ring widths typically
range from 1.2 to 1.8 m (4 to 6 feet) although some have been made 2.2 m (7.2 feet) wide.
As soon as the ring is built, the TBM resumes advancing, which means that the next to last
ring already embedded in the grout is experiencing thrust and steering forces about one
hour after the grout has been placed.
TBMs pull a long train of equipment behind them that is about 100 to 150 m (330 to 500
feet) long known as gantries, back-​up or trailing gear. That train of equipment has substan-
tial weight and usually advances on some type of wheeled system. Often there is a bridge-​
like structure that spans from the end of the TBM to the front of the first wheel supported
trailing gear. The idea is for the bridge to span around five to seven rings before the first
wheel loads are imposed on the liner. These wheel loads can be quite high for a large diam-
eter TBM. The grout needs to be stiff enough so that when the wheels roll over the invert
of the rings, the liner rings do not move up and down like piano keys. The first wheel loads
632 Handbook of Precast Segmental Tunnel Lining Systems

must be resisted by the liner ring and grout combination about 2 to 5 hours after grout
placement depending on the rate of TBM advance.

11.4 GROUT TYPES

Attempts have been made to classify annular grout by the degree of influence of the cementi-
tious materials and by how the physical components are combined. Like many classification
systems, there are always exceptions.

11.4.1 Classification by amount of cement

The EFNARC (European Federation of National Associations Representing for Concrete)
in its “Specification and Guidelines for the Use of Specialist Products for Mechanized
Tunneling (TBM), 2005” subdivided backfill grout into three categories, based on the pres-
ence and quantity of active cementitious materials:

• Active –​full hydration of a cement component that acts as a binder and provides the
stiffness and hardness properties of the final product
• Semi-​inert –​a lesser amount of cement and a slower strength gain and lower ultimate
strength
• Inert –​grout without cement, but perhaps with lime or clay

The EFNARC division between active and semi-​inert grout are defined by the amount of
cement in the grout but are somewhat arbitrary. For this chapter, we will use Thewes and
Budach’s classification that more than 200 kg/​m3 (344 lbs/​CY) of cement constitutes an
active grout while 200 to 50 kg/​m3 (344 to 86 lbs/​CY) of cement constitutes a semi-​inert
grout. Semi-​inert grouts are also called semi-​active or reduced active grouts. Active mortar
grouts have the highest long-​term strength because of the amount of cement used. Active
non-​mortar grouts also have the possibility of being chemically accelerated at the point of
injection. Mortar in this context means that the grout contains a fine aggregate.
Semi-​inert mortar grouts are mainly composed of the same elements, as found in the
inert type, but there is stronger reliance on the properties of the non-​cement ingredients to
give the proper rheological properties. A smaller percentage of cement or other hydrating
material will cause a degree of hydration to occur. The grout may harden slowly although
stiffening will occur more quickly.
Inert mortar grouts contain no Portland cement but may contain other hydrating ingredi-
ents, such as hydraulic lime. Inert grout ingredients are a mixture of water, various grades of
natural and crushed sand, and supplementary fine materials such as fly ash, metakaolin, slag
cement, lime, silica fume, crushed stone filler and bentonite. The compressive strength is very
low, and the setting time will be delayed or sometimes will never set. Inert mortar grouts are
economical backfill grouts. Inert mortar grouts have been used in several projects, mainly
by French contractors and engineers. The fine aggregates must be carefully selected and pro-
portioned to get the right mix of flowability during placement and stiffness after injection.
The classification system is not perfect. For example, some projects use slag cement as the
only active ingredient. This could be classified as an inert grout since it contains no Portland
cement, but the grout does hydrate and gets hard so it could be considered active grout. Pea
gravel is an inert ingredient used in some grouts, but the gravel is later flooded with cemen-
titious grout, which is clearly active.
 Annular backfill grouting of precast segmental tunnel linings 633

11.4.2 Classification by grout components

Grouts can also be classified by how the physical components are used:

• Single component –​all ingredients are mixed outside the tunnel and delivered to the
tunnel heading for injection into the annular space
• Two component –​one component is the cementitious product, and the other is an
accelerator, which are only mixed at the last moment to provide a grout that becomes
a gel within seconds of injection.
• Pea gravel –​blown into the annular gap and then later flooded with cementitious grout

Single component grout can be active, semi-​inert, or inert. Cementitious grout is composed
of cement, bentonite, water, admixtures and a fine aggregate. Fly ash may also be substi-
tuted for part of the cement. One component grouts are batched outside the tunnel and
transported to the tunnel heading via rail or rubber-​tired transport. Once at the heading,
the grout is transferred to a holding tank for injection into the annulus. The injection system
must maintain the backfill mortar pressure at or above the face pressure (as a rule of thumb
one to two bars above face pressure). Injecting the grout with too high a pressure can force
the grout to go into the cutter head chamber or into the wire brushes. Grout contaminated
brushes may require replacement. The stability of the ring occurs when there is adequate
cohesion of the mix and adequate friction at the grout/​ring and grout/​soil interfaces.
Chemical admixtures play an important role in active and semi-​inert single component
grouts since the mix should have enough fluidity to be transported or pump from the
grout plant to the TBM; and the mix should be able to set within 4 to 5 hours of batching.
Admixtures could be hydration controllers, viscosity modifiers or a superplasticizer.
Advantages of the one component grout are that the ingredients are less expensive than a
two-​component grout mix and visually checking grout volumes injected from holding tanks

Figure 11.2 One-​c omponent cementitious grout.

Note: Grout must remain flowable, with no segregation or bleeding, until injected.
634 Handbook of Precast Segmental Tunnel Lining Systems

is easy. Over the last 10 years, two-​component grouts have become popular but one com-
ponent grouts are still favored by some contractors and engineers.

11.4.3 Example grout mixes

Example grout mixes are provided below for the major grout types. Some of these are fur-
ther discussed in Sections 11.9, 11.10 and 11.11 where examples of grout application are
given. These are examples only. As is discussed in Section 11.9 on grout mix development,
the mixes must be tailored to the available ingredients and wide variations in grout proper-
ties can be expected with different materials.

11.4.3.1 One-​component active grout

The grout mix shown in Table 11.1 was used for an EPB TBM at the Riachuelo tunnel
outfall project in Buenos Aires, Argentina. The TBM advanced in clayey sands under 4 bar
(60 psi) of water pressure. Dimensions of the tunnel are 5.2 m (17 foot) excavated diameter
with a length of 12 km (7 miles) and most of the tunnel is located under the Rio de la Plata
estuary. The owner’s engineer requested a one-​component grout because of prior difficulties
using a two-​component accelerated grout. Initial mix designs used fly ash, but the fly ash
supply was not reliable. The 0 to 6 mm (0 to 0.25 inches) aggregate was crushed and diffi-
cult to pump. No other aggregates were economically viable. The 400 kg/​m3 (675 lbs/​CY)
of cement was used to overcome these difficulties. This is an excellent example of how local
preferences and materials can govern the selection and design of the grout. More details are
in Section 11.9.

11.4.3.2 One-​component semi-​inert

This semi-​inert grout mix shown in Table 11.2 was used for the Metro Doha Red Line
South tunnel project. The twin tunnel project is 13 km (8 mile) long with an excavated

Table 11.1 Mix design for one-​c omponent active grout.

This mix used more cement to work with locally available materials

Component quantities
Metric Imperial
Item Mix design components Quantity UoM Quantity UoM
A Primary materials for 1.0 m 3 for 1.0 CY
1 Cement 400 kg/​m 3 689 lbs/​C Y
2 Pre-​h ydrated bentonite slurry at 5% 147 kg/​m 3
253 lbs/​C Y
3 Aggregates 0 to 6mm (0 to 0.25 inches) 757 kg/​m 3
1,304 lbs/​C Y
4 Aggregates fine 606 kg/​m 3
1,044 lbs/​C Y
5 Water 185 kg/​m 3
38.24 gal/​C Y
B Admixtures
1 Retarder/​s uperplasticizer combination 2.28 l/​m 3 60 oz/​C Y
 Annular backfill grouting of precast segmental tunnel linings 635

Table 11.2 Mix design for one-​c omponent semi-​inert grout.

An economical mix for 26 km (16 miles) of tunnel for one-​c omponent
semi-​i nert grout

Component quantities
Metric Imperial
Item Mix design components Quantity UoM Quantity UoM
A Primary materials for 1.0 m 3 for 1.0 CY
1 Cement 40 kg/​m 3 69 lbs/​C Y
2 Bentonite 20 kg/​m 3
34 lbs/​C Y
3 Aggregates (crushed and powdered 1,150 kg/​m 3
1,981 lbs/​C Y
limestone)
4 Fly ash 200 kg/​m 3 344 lbs/​C Y
5 Water 350 kg/​m 3
72.24 gal/​C Y
B Admixtures
1 Micro air entrainment 0.75 l/​m 3 20 oz/​C Y
2 Long chain polymer superplasticizer 3.0 l/​m 3
79 oz/​C Y

diameter of 7 m (23 foot) and was driven in shale and limestone rock containing ground-
water with high amounts of sulphates and chlorides.

11.4.3.3 One-​component inert grout

The grout mix shown in Table 11.3 was used for the Pannerdensch Kanal Tunnel in The
Netherlands. The 9.8 m (32 foot) diameter, 1.6 km (1 mile) long twin tunnels were both
driven by a slurry shield TBM, primarily in sand formations with only 5 m (16 feet) of cover.
The French contractor had extensive experience with inert grouts. The careful selection of
non-​cementitious binders and fine aggregates resulted in an economical grout where no
admixtures were required.
The grout injection pressure was at least 0.5 bar (7.5 psi) over the groundwater pressure.
The short-​term stiffness required to support the liner ring came from the excess water being
squeezed out of the mortar after injection. This consolidation was essential to the successful
application of the grout. Since the compressive strength was low in the hours after injection,
the modulus of elasticity (E) was an important criterion in the grout design and testing. The
grout was designed to have a minimum E of 1.5 MPa (145 psi) after 30 minutes and a min-
imum E of 3.5 MPa (508 psi) after 90 minutes. Days later, the lime binder started setting
and the grout stiffness exceeded that of the surrounding ground. Compressive strength was
0.38 MPa (55 psi) in 28 days and 2.0 MPa (290 psi) in 90 days. Grout workability was at
least 24 hours.

11.4.3.4 Two-​c omponent active

The grout mix shown in Table 11.4 was used for the Anacostia River Tunnel in Washington
D.C (United States). The tunnel was excavated in heavy clay and clayey sands with up to 3
bar (45 psi) of water pressure. This mix is typical of a two-​component grout placed through
636 Handbook of Precast Segmental Tunnel Lining Systems

Table 11.3 Mix design for one-​c omponent inert grout.

Carefully selected binders and fine aggregates and no admixtures for
one-​c omponent grout

Component quantities
Metric Imperial
Item Mix design components Quantity UoM Quantity UoM
A Primary materials
1 Hydraulic lime 72 kg/​m 3 124 lbs/​C Y
2 Fly ash 251 kg/​m 3
432 lbs/​C Y
3 Limestone fillers 152 kg/​m 3
262 lbs/​C Y
4 Natural sand 0 to 1.4 mm 527 kg/​m 3
908 lbs/​C Y
(0 to 0.055 inch)
5 Natural sand 1.4 to 3.4 mm (0.055 to 263 kg/​m 3 453 lbs/​C Y
0.134 inches)
6 Natural sand 3.4 to 5.6 mm (0.134 to 198 kg/​m 3 341 lbs/​C Y
0.220 inches)
7 Natural sand 5.6 to 8 mm (0.220 to 329 kg/​m 3 567 lbs/​C Y
0.315 inches)
8 Water 295 kg/​m 3 60.89 gal/​C Y

Table 11.4 Mix design for two-​c omponent active grout.

Grout pumped through up to 3,750 m (12,300 feet) of pipeline to the TBM for two
component active grout

Component quantities
Metric Imperial
Item Mix design components Quantity UoM Quantity UoM
A Primary materials for 1.0 m 3 for 1.0 CY
1 Cement 280 kg/​m 3 482 lbs/​C Y
2 Bentonite 30 kg/​m 3
52 lbs/​C Y
3 Water 838 kg/​m 3
173 gal/​C Y
B Admixtures
1 Hydration stabilizer–​r etarder   4 l/​m 3 106 oz/​C Y
2 Sodium silicate based accelerator 75 l/​m 3
15.5 gal/​C Y
added at tail shield injection mixer
 Annular backfill grouting of precast segmental tunnel linings 637

Table 11.5 Mix design for pea gravel with active grout.

This fluid grout used for both flood and contact grout using pes gravel and
active grout

Component quantities
Metric Imperial
Mix design components
Item for 1.0 m 3 (and 1.0 CY) Quantity UoM Quantity UoM
A Primary materials for 1.0 m 3 for 1.0 CY
1 Pea gravel –​ clean with none passing ¼ placed in advance placed in advance
inch sieve and 100% passing ½ sieve
2 Cement 897 kg/​m 3 1,545 lbs/​C Y
3 Fly ash 98 kg/​m 3
169.0 lbs/​C Y
4 Water 427 kg/​m 3
88.13 gal/​C Y
5 Pre-​h ydrated bentonite slurry 6% 232 kg/​m 3
47.88 gal/​C Y
solution
B Admixtures
1 Superplasticizer 12 l/​m 3 317 oz/​C Y
2 Hydration stabilizer (retarder) 0.5 to 1.0 l/​m 3
13 to 26 oz/​C Y

the tail shield in a soft ground tunnel where required ultimate strengths are low. More
details are in Section 11.10.

11.4.3.5 Pea gravel and flood grout

This pea gravel and flood grout system shown in Table 11.5 was used on a 6.6 m (22
foot) diameter rock tunnel in Canada. First, pea gravel was injected into the annular
space. The fluid grout below was used to fill the gravel interstices and as a contact (proof)
grout to ensure that all the annulus was filled. The gravel met the usual requirements for
concrete aggregates including abrasion and alkali silica reactivity. More details are in
Section 11.11.

11.5 GROUT INJECTION METHODS

Before moving on to grout design, a discussion of the methods to inject annular grout and
the equipment used will be helpful. Grouting the annular space behind the TBM tail shield
can be done in two different ways. Traditionally, grout was injected through the precast seg-
ments, which was adequate when TBMs advanced under little or no hydrostatic pressure, or
above the water table. Today, when TBMs operate at higher face pressures, injecting grout
from the trailing edge of the TBM tail shield is the preferred method.
Pressurized face TBMs have been operated with face pressure ranges anywhere from 1
to 20 bar (15 to 300 psi) although most operate in the 2 to 5 bar (30 to 75 psi) range. The
pressure at the face is necessary to counteract the ground pressure, which is a combination
of porewater pressure and active soil pressure. This pressure exists not only at the face but
also along the length of the shield and at the annular space.
638 Handbook of Precast Segmental Tunnel Lining Systems

Grout injection quantity limits are usually preset at the theoretical quantity plus 15 to
20%. Grout pressure limits are usually set a 1 to 2 bar (15 to 30 psi) above the face pres-
sure. Grout injection should stop when one of these limits is exceeded. Slurry TBMs will
have the same grout pressure at the face, along the shield, and in the annulus because any
vacant space is filled with fluid. EPB TBMs tend to lose pressure along the shield and the
larger annulus behind the tail shield may be at a lower pressure. For TBMs operating at
atmospheric pressure, quantity is usually the controlling factor. Injecting too much grout
risks forcing the grout around the TBM shield and possibly grouting in the TBM. Grout
quantities and pressures should be tracked to look for worrying trends.
Grout injection pressure readings can often be a source of disagreement. Grout pres-
sures sensed at the pump are a combination of the pressure from the face +​static pressure
of the grout head +​the dynamic pumping pressure. If a two-​component grout is used, the
viscosity changes quickly over time as the grout gels. The best indication of grout pressure
is the grout pressure reading from a top port sensor at the end of the TBM advance when
the annular gap is filled, and no more grout is being injected. A good practice is to pump
grout through the circuit before the TBM is launched at various flow rates to determine the
dynamic pumping pressures. This information will assist with interpreting pumping pres-
sures during normal operations.
A grout that is difficult to pump can result in grout over-​pressures around the precast
segmental liner. For example, a one-​component grout with poorly selected aggregates will
require more pump pressure to move the grout through the lines. If the injection pump
cut-​off pressure is set to 5 bar (72.5 psi) above face pressure to overcome dynamic pres-
sures in the grout line, the entire liner ring can be over pressurized. When the TBM gets
to the end of the advance and the last bit of fluid grout is injected into the annular space
reaches the cut-​off pressure, all the fluid annular grout in the space will be at the cut-​off
pressure.
Both pressurized and non-​pressurized face TBMs may have an excluder ring (Figure 11.3)
around the outer perimeter of the tailshield, typically overlapping thin spring steel plates, to
prevent grout from moving forward around the TBM shield. Once launched, the excluder
shield is not accessible for repair or inspection and can be easily damaged or worn away as
the TBM advances. The excluder shield is not a sophisticated device and will not seal off
areas where the ground has fallen out prior to grouting. Annular grout best practices should
not rely on the presence of the excluder ring but should rely on closely monitoring our grout
quantities and pressures. Excessive grouting pressures will force grout around the TBM or
past the tailshield wire brushes and grease seals.
The tail seal system and the grouting system must work together. Without a properly
operating tail seal system, there is no way for the annular grout to achieve the desired
pressure as the grout will leak past the tail seal brushes and contaminate the tail seal area
with grout. The tail seal grease must be injected between the wire brushes with a quantity
and pressure adequate to seal off the tail shield and protect the wire brushes. Operating
the annular grout system at excessive pressures will force grout past the tail seal brushes,
damage the tail seal system and create a clean-​up problem inside the TBM tail shield. The
grout injection rate should be commensurate with TBM advance rate to maintain the pres-
sure in the annulus. Grout flow in the pipes leading to the injection port should maintain
a minimum flow rate to prevent clogging. About 0.67 m/​sec (2.2 feet/​sec) is a good rule of
thumb. The ring erector operator has the best view of the segment/​tail shield interface and
must alert the grout operator when excessive grout leakage is observed. A slight increase in
 Annular backfill grouting of precast segmental tunnel linings 639

Figure 11.3 TBM grout ports and excluder plates.

Note: Grout ports are circled in yellow. Tail seal brushes on interior. Excluder plates on outside
perimeter.

tail seal grease pressure may be enough to stop the leak. Continuous leakage of grout will
require inspection of the tail seal system.
The decision for either method must be made early in the project as the tail shield method
requires the installation of grout pipes and discharge ports (Figure 11.4) in the tail shield.
Grouting through the segments requires the installation of grout ports in the segment. If
through-​segment grouting is selected as the production method, provisions should be made
in the segments to perform remedial grouting, also known as proof grouting or secondary
grouting. The TBM trailing gear may carry the equipment to support the grouting operation
and must be compatible with the method used.

11.5.1 Grouting through the precast segments

Grouting through the precast segments cannot be done until the grout port has cleared the
tail shield. Consequently, unstable ground can collapse on the liner ring before the grout is
placed. Grouting after the ground has collapsed will not be successful in evenly loading the
rings or preventing surface settlement. Once the grout is injected and the grout has an initial
set, injecting additional grout will disturb the grout already placed and may unevenly load
the previously installed ring. Completely filling the annulus at the 12 o’clock position can be
difficult because the grout tends to slump and flow towards the shield. Air can be trapped
at the arch and resist complete grout filling at the top of the ring.
Some segments are cast with grout ports that either have a removable cap or are
drilled out in the tunnel to allow injection. A grout hose is connected to the grout
640 Handbook of Precast Segmental Tunnel Lining Systems

Figure 11.4 Close-​u p of grout ports.

Note: One grout port is a spare that stays filled with tail seal grease until required.

port and the grout is injected. The quantity and pressure of the grout are monitored,
and injection is stopped when preset limits are achieved. The grout hoses are moved
up as the next liner ring is pushed out of the tail shield. The grout can either be a
mortar or a two-​component grout. Reinjection through a grout port is not possible if a
two-​component grout is used. Mortar grouts are less susceptible to line plugging when
injected through the segments.
Typically, the contractor can select between mechanical lifting or vacuum lifting of
the segments. For mechanical lifting, a threaded plastic insert is cast into the center of
gravity of each segment. This threaded plastic insert can also be used for grouting, either
primary or secondary. A hollow adapter that screws into the insert (Figure 11.5) with a
grout hose fitting on the end projecting inside the tunnel is used to connect to the grout
pump hose.
The procedures below describe the steps for grouting through segments in various
conditions.

• Dry tunnel
• Drill with a 25 mm (1 inch) diameter roto-​hammer bit through the back of the
segment
• Screw in adapter
• Attach grout hoses and pump to refusal
• Dry tunnel–​digger shield
• Same procedure as above
• Drilling should be done 10 m (30 feet) back from the tail of the shield This is to
prevent grout traveling on top of the machine
• Wet or pressurized tunnel
• Install adapter with a 50 mm (2 inch) diameter NPT full port ball valve
 Annular backfill grouting of precast segmental tunnel linings 641

Figure 11.5 Adapter for through-​s egment grouting.

Note: Rope thread on lower half matches cast-​in-​p lace insert and machine thread connects to
grout hose.

• With the valve open, drill with a 25 mm (1 inch) diameter roto-​hammer bit through
the back of the segment until contact with water
• Quickly remove the drill bit and close the 50 mm (2 inch) diameter valve
• Attach grout hoses and pump to refusal
One-​component or two-​component grouts can be used with through-​the-​segment injection.
A recent development has been to inject two-​component grouts through the segments when
advancing in rock tunnels. The grout is injected farther back from the tail shield to prevent
grout going around the shield and being ingested by the cutterhead.

11.5.2 Grouting through TBM tail shield –​general

For pressurized tunnels, grouting is best done through pipes that are embedded in the tail
shield. This arrangement (Figures 11.4 and 11.6) allows for grout to be injected simultan­­
eously with the advance of the TBM. Tail shield grouting requires that the TBM be designed
and manufactured for the task. Tail shield grouting can be done with either one-​part mortar
grouts or two-​part accelerated cement grouts. The grout piping is embedded in the tail
shield and discharges right at the trailing edge of the shield. Embedding grout pipes in
the tail shield increases the thickness by about 50 mm (2 inch) and therefore increases the
annular gap by the same amount.
Tail shield grouting has the advantage that the annulus behind the TBM is immediately
filled providing excellent support to the ground and the segments. Because the injection
system is integral to the TBM, the grout injection can occur in semi-​or completely automatic
modes with the quantities and pressures continuously recorded. The grout pressure can be
maintained automatically while the TBM is at a standstill during ring erection. Modern
computer controlled grouting systems can follow a set pattern of injection using some or
all of the grout ports. A TBM can advance with only some of the ports being operational if
desired quantities and desired pressures are being obtained. Slowing down the advance rate
of the TBM may be required if all grout ports are not operational and the injection pump
cannot inject enough grout.
642 Handbook of Precast Segmental Tunnel Lining Systems

Figure 11.6 Injecting grout through TBM tail shield.

Note: Injection port structures increase annular gap but provide more complete filling of the gap.

Tail shield grouting has the disadvantage that the pipes carrying the grout through the
shield are relatively small and can clog easily with one-​part grouts if the grout is not prop-
erly designed. Even two-​part grouts can set up and clog the lines if the lines are not kept
clean. Grout lines are usually installed in four quadrants of the tail shield. Spare grout lines
may be embedded in the tail shield for rapid change over when a line is plugged. These
are usually filled with tail seal grease to keep them from filling up with grout when not in
use. Cleaning the lines is usually done by rodding with a cable or by jetting with high pres-
sure water.

11.5.3 Two-​component grouting through the tail shield

With two-​part grouts, additional piping is required to mix in the accelerator and have a
flushing circuit as shown in Figure 11.7.
The two-​component pipes meet at a pipe junction within the shield and where the grout
and accelerator are mixed and injected into the annular gap between the extrados of the
ring and the native ground.
A proper set-​up for TBM in-​tail shield mixing of the two components will have the mixer
as close to the tail as possible to minimize clean-​up and clogging. The TBM should have a
spare grout line at each grout location to reduce TBM down time in the event a grout line
becomes clogged. A line should be provided for backflushing the lines with water to prevent
clogging. Clean-​out water will fall to the shield invert where a small pump (air diaphragm
or submersible) can move the water into the discharge water tank.
There is a specific sequence recommended for two-​component grouting connected to the
TBM advance cycle. Grouting injection through the tail shield should resume only after the
TBM has advanced a small amount to the point where the grout pump can operate continu-
ously. Grout injection should begin with the grout only followed by accelerator injection
starting a few seconds later. At the end of a stroke the accelerator injection should stop
shortly before grout injection stops. These measures all help prevent grout plugs inside the
tail shield injection ports. Grout injection should continue past the end of the TBM excava-
tion stroke until the cut-​out pressure is achieved. This makes sure the annular crown space
is filled.
 Annular backfill grouting of precast segmental tunnel linings 643

Figure 11.7 Injecting two-​c omponent grout through TBM tail shield.

Note: A component is the grout and B component is the accelerator. Image courtesy of
Herrenknecht Tunneling Systems.

11.6 SELECTION OF THE GROUTING METHOD

11.6.1 Development of different grouting methods

The various grout classifications provide no guidance to the young practitioner as to how
to think about all the different types of grout. A little history on the development of the
grouts will help. One-​component grouts were first used with early western pressurized face
EPB and slurry tunnel boring machines starting in the 1970s. These traditional grouts were
composed of fine aggregate, water, cement, bentonite and some admixtures to control set
and bleed. The ingredients were mixed in a concrete style batch plant and transported to the
tunnel heading for delivery to a temporary holding tank until they could be injected through
the tunnel liner. The initial support depended on the change in state from a flowable grout
to one with some stiffness caused by the excess water being squeezed out by the grouting
pressures. Admixtures kept the grout fluid until it could be injected.
Although one-​component grouts started-​out with high cement contents, engineers real-
ized that the grout only needs to be as strong as the surrounding ground. Providing initial
support needed for the lining to resist TBM thrust forces and trailing gear wheel loads were
the immediate requirements. This led to reductions in cement content and finally to the
development of one-​part grouts that rely more on the careful selection of the fine aggregates
and a binder to provide the necessary grout properties.
Grouts have been developed that use kaolinite clay or lime as the binder. While inert
grouts may sound strange, consider that a slurry TBM advancing in sands below the water
table erects a precast ring that will be surrounded by sand and the 150 mm (6 inches) layer
of grout filling. The annulus does not need to be any stronger than the surrounding ground
644 Handbook of Precast Segmental Tunnel Lining Systems

if it can provide the immediate support needed by the ring and fill the annulus before the
native ground collapses into the annular void. One-​part grouts can be active, semi-​active or
inert. One-​part grouts are still used although they have been largely replaced by two com-
ponent grouts.
Two-​component (i.e. bi-​component) grouts are also called A+​B grouts where A is the fluid
cementitious material and B is the accelerator. They were invented by the Japanese to pro-
vide more immediate support and started being used in western countries around 2000. The
grouts are made by combining a cementitious component (generally cement, bentonite, and
a retarder) with an accelerator (generally sodium silicate based). Fly ash and slag cement
may also be used.
The two components are mixed just before or just after injection into the annular space
and the grout forms a gel withing 5 to 15 seconds that continues to gain strength by the
hour. The grout is very fluid at first but stays in place once gelling starts and provides nearly
immediate support to the ground and liner ring. Because the cementitious component con-
tains no sand, it can be pumped from the surface to the TBM using 50 or 75 mm (2 or
3 inch) diameter steel pipes, eliminating the need for rail or rubber-​tired transport to the
heading. The grout mixing plant on the surface is also simplified as the grout can be mixed
using colloidal mixers instead of a concrete-​style batch plant. Two component grouts can
provide more consistent properties as the primary ingredients are made in factories with
high-​quality control. Two-​component grouts are classified as active grouts.
Shielded TBMs that advance through rock and assemble a finished precast liner in the tail
shield were first used in Europe. These shielded rock TBMs operated at atmospheric pressure
so there was not an immediate need to support the ground to prevent surface subsidence,
however the annular void needs to be filled to prevent the loosening of the rock and uneven
loading of the liner ring. There was also the need to support the assembled rings so that they
maintain a circular profile and carry the wheel loads of the trailing gear immediately behind
the TBM. Clean pea gravel with a maximum diameter of 10 mm (0.375-​inch) was used to fill
the annular space around the liner ring. Because the TBM was not operating in a pressurized
face mode, there was nothing to prevent a fluid grout from moving towards the TBM cutter-
head once injected. Pea gravel was injected as the initial supporting medium. The pea gravel
was blown in through the segments to fill the annulus as much as possible and then later the
same injection points were used to inject a fluid cementitious grout to fill the interstitial spaces
in the pea gravel and prevent the migration of ground water. The pea gravel filler was inert,
but the flood grout was an active grout. Two-​component accelerated grouts are beginning to
replace pea gravel/​flood grout as contractors become more familiar with the technology.

11.6.2 Selection of a grouting method and grout type

The selection of the grouting system and grout will depend on many interacting factors
including the following:

• Sensitivity of the ground to settlement

• Presence of structures along the alignment
• Owner preference
• Contractor preference
• Type of ground being excavated
• Availability of materials
• Cost
 Annular backfill grouting of precast segmental tunnel linings 645

If the tunnel alignment passes under buildings, bridges and critical utilities, such as old
water mains, there will be more concern with settlement than if the tunnel is passing under
open fields or under bodies of water. Soft ground is more susceptible to settlement than
rock. Some types of soft ground, clean pressurized sands, for example, may be more sensi-
tive than others, such as heavy consolidated clays. The owner and contractor may have a
liability to third parties if settlement is not controlled. Owners or contractors may have to
make settlement performance promises to obtain the necessary permits. For sensitive condi-
tions, grouting through the tail shield is the preferred method.
The type of grout to be used may depend on the owner’s past tunneling experience. If the
owner has had bad experiences with two-​component accelerated grout, they may request
that a one-​component grout be used. In a traditional design/​bid/​build project the owner and
designer may specify the grouting method and the type of grout to be used before the job
is bid. Most western contractors are moving towards grout injection through the tail shield
using two component accelerated grouts for soft ground and hard rock applications. The
ingredients themselves are more expensive than one-​part mortar grouts but the system sim-
plifies the surface grout mixing plant, and the grout can be delivered via a pipeline to the
TBM. One-​component style mortar grouts require a batch plant, like a concrete plant with
all the attendant material handling equipment.
If the contractor already has the equipment invested in the batch plant and rail delivery
cars, he may make the calculation that a one-​part grout is still more economical. Familiarity
with the technology plays a big part in the selection process. One-​part grouts do clog tail
shield grout pipes more easily than two-​part accelerated grouts. The crews must be trained
to properly operate the grout system and clean it out at the first sign of clogging.
Availability of grout ingredients is always a consideration. In developed countries, the
prompt delivery of quality ingredients such as cement, fly ash, bentonite and accelerator
may be a given. In remote locations or under-​developed countries, there may be limitations
on availability, timeliness of deliveries or reasonable pricing. In these cases, looking at one-​
part semi-​inert or inert grouts where local aggregates can make up most of the grout and
admixtures can be minimized may be appropriate.
Large quantities of grout are required for tunnels of any length. For example, a 10 km (6
miles) tunnel with an excavated diameter of 8 m (26 feet) and an annulus gap of 150 mm
(6 inches) requires 37,000 m3 (47,360 CY) of grout. If the sample one-​component mix in
Table 11.1 is used, 14,800 tons of cement and 50,431 tons of aggregate are required. If the
same space were fill using the accelerated two-​component grout in Table 11.4, no aggregate
would be required but 10,360 tons of cement would be required. The grout should never
limit the advance of the TBM. One needs to be sure that the requisite materials can be deliv-
ered consistently to allow maximum production.
Table 11.6 below summarizes the different grouts and grouting systems and where they
may be used.

11.7 MIX DESIGN DEVELOPMENT

Grout mix design development starts with deciding what properties are desired. Commonly
used definitions and terminology for the various grout properties are described in this
section. Actual test procedures used to verify the grout properties are covered in Section
11.12. Annular grout is batched using many components that have common names such
as “cement”, “fly ash” and “sand”, but these materials have characteristics that can vary
widely from one source to another. The materials variability requires an iterative mix design
646 Handbook of Precast Segmental Tunnel Lining Systems

Table 11.6 Grout application by type

Application Backfilling Required

range system equipment Specific remarks
Material
Item description Rock Soil A B a b c d
1 Active      Conventional mortar,
mortar system stiffness behavior depends
(one-c​ omponent) on aggregate selection
and admixture effects on
cement.
2 Semi-​i nert     Mortar mix where stiffness
mortar system behavior depends on
(one-c​ omponent) primarily on aggregate
selection, but admixture still
has effect on cement.
3 Inert    Stiffness behavior depends
mortar system on aggregate selection
(one-c​ omponent) although non-​c ement
powers such a hydraulic lime
influence late age stiffness.
4 Two        Cement gels 8 to 15 seconds
component after mixing w. strength gain
active grout from cement, no aggregate
system present
(A+​B or
bi-​c omponent)
5 Pea gravel    Used in hard rock tunnels.
followed by Increased bedding by using
active flood mortar at the bottom,
grout normally lower modulus
of deformation and lower
properties of embedment
than for an Active Mortar
Systems.
Source: Modified from a table by Thewes and Budach [10].
Table Legend and Notes:

Applicable a
Piston pump
X
Limited applicability b
Peristaltic pump
A
Backfilling through holes in the segments C
Progressive cavity pump
B
Backfilling through the tail skin d
Compressed air

process. Multiple trial batches are prepared and tested, using ingredients available to the
tunnel project, until the right combination of properties is achieved.

11.7.1 Performance requirements

Now that the annular grout objectives, grout types and grout injection methods have been
discussed, we can move onto what makes a good grout. For the long term (days to years),
the grout should have strength properties that at least match the surrounding soil, have a
lower permeability and be volumetrically stable. Design specifications tend to concentrate
on the long-​term properties while the short-​term properties are left to the contractor. Tunnel
 Annular backfill grouting of precast segmental tunnel linings 647

grouts do not have as background the body of specifications that have accumulated for cast-​
in-​place concrete.
There are some guidelines but there is still lots of room for experimentation.

11.7.2 Quantifiable properties of fresh and hardened grout

For mix development and testing purposes, grouts are defined by the following properties:

• Flowability and viscosity

• Bleed and segregation
• Stabilization time (pot life)
• Gel time (for two-​component grouts)
• Compressive strength

11.7.2.1 Flowability and viscosity

Flowability is the capability of the grout to move by flow while viscosity is resistance of a
fluid to flow. The methods used measure these properties depend on the consistency of the
annular grout. Grouts with low mobility are typically tested using a slump measurement.
Variations of the test method can be used, depending on the consistency of the annular
grout. The spread of the sample can be measured along with the slump. The slump cone
can also be inverted, and the spread of the grout can be measured. With very high mobility
annular grouts, the consistency can be measured using a flow cone. There are various types
including ASTM flow cone and a marsh funnel. The time of efflux of a volume of the grout
is measured in these procedures.

11.7.2.2 Bleed and segregation

Bleed is the separation of water from a grout mix due to gravitational settlement of the
solid materials and is a form of segregation. The excessive occurrence of bleed leads to plug-
ging in pump lines and settlement of the particulate ingredients in a tank. Bleed is a time-​
dependent property. When grout is at rest in a pipe, the heavier sand and cement particles
will move to the pipe invert and the water will rise to the crown decreasing the flowability of
the grout. If the bleeding is excessive, the grout will refuse to move. Increased line pressure
will only squeeze more water out of the line making the blockage worse. Bleed is measured
by putting a grout sample in a graduated cylinder and measuring the amount of water that
rises to the top with time. Two-​component mixtures transported through pipelines have the
most concern with this performance requirement although any grout that must be pumped
or stored should have low bleed.

11.7.2.3 Stabilization time

Stabilization time is generally measured by loss of mobility. For highly mobile annular
grout, this can be accomplished using a flow cone method. For lower mobility grouts, slump
and or spread loss measurements can be used. The requirements can be up to 72 hours of
stabilization. This is due to the long transfer of the grout from the surface to the TBM via
grout lines, or potential other delays such as a planned shutdown or a weekend stoppage.
Typically, a longer shutdown would require that the lines be drained and cleaned.
648 Handbook of Precast Segmental Tunnel Lining Systems

11.7.2.4 Gel time

Gel time is measured for high mobility two-​component grouts. This is the time required for
the grout to lose fluid mobility so that it will no longer move under gravity. Gel time is in
the range of 5 to 15 seconds. In non-​pressurized face rock TBMs, grout that is not gelled is
prone to washout if moving water is present.

11.7.2.5 Unconfined compressive strength

Unconfined compressive strength (UCS) can be measured using a variety of methods. The
age at which the grout sample will be tested will vary depending upon the requirements of
the annular grout specification. UCS development can be measured at all ages using cube
or cylindrical specimens. Test specimen size, shape and test ages should be outlined in the
requirements of the annular grout specification.
Some specifications can require methods measuring penetration resistance to indicate
early strength gain. This strength testing is used to measure strength gain in less than one
hour. The early UCS requirements are driven by the grout’s ability to hold the segment ring
in-​place. UCS in the following hours are driven by the added weights, such as loading from
the gantries. UCS after 24 hours show the strength development and quality of the grout.
A higher early UCS requirement usually results in a higher ultimate UCS.
The compressive strength requirements are generally specified in the contract documents
or given by the designer of the tunnel liner. Minimum values are typically set for three
stages. The first stage is after 1 hour, when the ring has been fully pushed out of the tail-​can
of the TBM. The next stage is after 3 to 8 hours, when the first gantry of the TBM trailing
gear rolls onto the ring. After 28 days, there are desired final values for which the grout
strength should plateau.

11.7.2.6 Fluid and installed properties

Desired placing and installed properties should be as listed in Table 11.7. As highlighted by
the table, the fluid properties of the grout prior to and during injection are very different
from the required properties the grout is placed. The high advance rates of current TBMs
magnify these contradictions.

Table 11.7 Desired properties for fluid and installed grout.

Properties desired before and immediately after placement are opposites

Item Fluid properties Installed properties

1 Pumpable to the point of injection Stiffness as soon as possible after placement
2 Fluid enough to fill annulus rapidly Predictable change from fluid to stiff grout
3 Extended working time No slumping once placed
4 Minimal bleeding during transport Resist washout if moving water is present
5 Uses available materials
6 Economical
7 Environmentally acceptable Inert –​ will not contaminate water table
 Annular backfill grouting of precast segmental tunnel linings 649

Table 11.8 EFNARC recommendations for one-​c omponent grout properties.

European and ASTM test methods for various one-​c omponent grout properties

Test method reference

Item Characteristic European ASTM Requirement
1 Bleeding EN 480-​4 :1997 C-​9 40 Less than 1% up to
the initial set or initial
stiffening time in the
case of Active and Semi-​
Active Grouts
72 hours in the case of an
Inert Grout
2 Setting time for active Vicat Tests per EN C-​1 91 Initial setting time within +​
grout method will 413: 1995 3 hours and 15 minutes of
depend on maximum the stated setting time.
aggregate size: Penetration test similar
to setting time method Final setting time not later
• 0 to 2 mm (0 to per EN 413:1995 than 4 hours after initial
0.08 inches) setting time.
• > 2.0 mm Temperature of
(>0.08 inches) to be ±1°C of the
temperature used in the
proving trials
3 Setting time for semi-​ As above Initial setting time within +​
inert grouts 4 hours and 1 hour of the
stated setting time.
Final setting time not later
than 12 hours after initial
setting times
4 Workability
4.1 Slump tests EN 413-​2 :1995 C-​1 43 ±25mm (1 inch) of stated
slump value
4.2 Spread table C-​1 427 ±10% of stated spread
value

EFNARC Specifications and Guidelines includes the following guidance for one-​part
grouts (Table 11.8).
For two-​component grouts, the requirements are slightly different. Bleeding is still an
important property as the A component needs to stay mobile for hours or days before being
mixed with the accelerator. As with one-​part grouts, excessive bleeding indicates settlement
of the cementitious particles, whether in a pipeline or in a tank. Bleeding and settlement can
have bad consequences when the A part is being pumped through a pipeline that is many
kilometers long. The liquid consistency of the A component is an important characteristic if
it is to be pumped to the heading.
Gelling time is generally in the range of 5 to 15 seconds and depends on the application.
After gelling, the grout has the consistency of a plastic clay. After one hour, the compressive
strength can be between 0.05 and 0.5 MPa (7 and 70 psi). After 24 hours, the compressive
strength can be between 1.5 and 2 MPa (218 to 290 psi). The 28-​day strength can vary
between 2 and 4 MPa (290 and 580 psi). The required 28-​day strength is generally higher
650 Handbook of Precast Segmental Tunnel Lining Systems

for tunnels excavated in rock. The strength achieved depends on the ratios of cementitious
materials, water, bentonite and accelerator. Cement contents can vary a lot for this type
of grout.
Two-​ component grouts are less susceptible to wash-​ out once the grout has gelled.
Pressurized face TBMs have less problems with washout since the TBM is operated at a
pressure that prevents ground water movement. Rock TBMs that operate at atmospheric
pressure have more problems with washout because the water can be moving toward the
excavated face.
The desire time of set depends on the grout deliver method to the heading and the advance
rate of the TBM. If grout cars are used to transport grout from the batch plant to the TBM,
there will probably be a storage tank at the shaft, the transport in the railcar, and the stor-
age tank on the trailing gear before the grout is finally injected into the annulus. Delays may
occur during TBM advance and that may require dumping the grout because of the rela-
tively short set time commonly used in one-​component grouts. If the grout is transported
along the tunnel by pipeline, the grout may rest in the pipe for extended periods without
agitation.

11.7.3 Mix design development

The grout mix design should be developed and tested prior to the start of tunneling using
the materials that are economically viable for the grout. Proper mix design and selection
of the grout components and testing is the key to minimizing cost, while ensuring a safe,
fast-​paced operation. While the project specifications may contain performance require-
ments for the grout, they may not entirely specify the behavior or materials desired by the
contractor.

11.7.3.1 Mix design iterative process

Grout mix design development is an iterative process that takes time. Generally, 28-​day
compressive strength results are required by the specifications and that defines the testing
cycle. Sometimes a mix design may be available from a nearby project that uses the same
materials, but the markets change all the time and what may have been economically feas-
ible for a previous project may no longer be possible. The contractor may have the option to
decide what type of grout (discussed in Section 11.6) should be used and may decide to run
grout mixes for several types of grout. The multiple ingredients interact with each other thus
requiring multiple tests to hold determine optimum levels of each ingredient. Figure 11.8
shows the iterative process for a two-​component grout, which is simplified because aggre-
gate is not present as an additional variable.
When developing two-​component grouts, the mix should be designed and documented to
meet the required strength and shrinkage properties within an allowable range of acceler-
ator additions. The allowable accelerator range will allow the field to change the accelerator
addition to adjust the desired gel times based on actual conditions.

11.7.3.2 Cementitious products

Cement types do not all perform in the same manner due to the chemical composition,
phase composition, sulfate content and Blaine fineness. Table 11.9 shows test results for a
two-​component grout. Identical mix designs were tested using two sources of local type I/​II
 Annular backfill grouting of precast segmental tunnel linings 651

Figure 11.8 Iterative process for two-​c omponent grout mix design.

Note: Development time is less for one-​c omponent grouts.

cements. The flow and the compressive strength results varied significantly from one cement
to the other. Always test the different available materials to optimize the mix design and
have an alternative mix design for cases where critical materials become unavailable or too
costly to use.
Fly ash is widely used in annular grout. Fly ash is a very fine powder product of coal com-
bustion in electric power generating plants. Class C is usually a reactive fly ash while F is a
pozzolanic fly ash. Type F fly ash tends to produce more consistent results in grout mixes.
The ASTM specifications for fly ash are not tight and the characteristics of the fly ash vary
from powerplant to powerplant depending on the coal used and how the power plant is
operated.
ASTM C476 and C270 permit the use of cement meeting ASTM C595, “Standard
Specification for Blended Hydraulic Cements”. ASTM C595 allows 15% to 40% fly ash
by weight of Portland cement as a cement replacement. Fly ash is used in two-​component
annular grout mainly because it cost less than cement. Fly ash increases durability and
enhances pumpability
Table 11.10 provides an example of two different Class F fly ashes with different perform­
ance. Using the same mix design as in Table 11.9 and same components Fly Ash 1 increased
the flow duration after 24 and 48 hours. This case is not typical but demonstrates the vari-
ability of two fly ashes that both meet the ASTM Type F classification.

11.7.3.3 Bentonite
Bentonite is an absorbent aluminum phyllosilicate clay used in two-​component annular
grout as an economical way to control bleeding. Bentonite clays absorb several times its
mass in water enhancing the rheological properties of the grout mix. Bentonite performance
depends on the source of the bentonite and how it reacts with the other components in
652 Handbook of Precast Segmental Tunnel Lining Systems

Table 11.9 Flow and strength comparisons for the same mix with different cements.
Never assume that a grout mix from one project can be used on another without
testing

Cement 1 Cement 2
Metric Imperial Metric Imperial
ASTM
Item Description code Qty UoM Qty UoM Qty UoM Qty UoM
A Viscosity D6910
retention (no
accelerator)
1 Initial 42.0 sec 43.0 sec
2 24 hours 45.0 sec 65.0 sec
3 48 hours 46.0 sec min n/​a
flow
4 72 hours 53.8 sec no flow n/​a

B Bleed (%) C940 1.75% 1.47%

C Accelerant AGA 41S AGA 41S
D Compressive 75 × 3” × 75 × 3” ×
strength 150 mm 6” 150mm 6”
1 1 hour 0.3 MPa 43 psi 0.4 MPa 59 psi
2 4 hours 1.3 MPa 187 psi 1.1 MPa 156 psi
3 8 hours 1.8 MPa 266 psi 1.8 MPa 264 psi
4 24 hours 2.9 MPa 426 psi 2.8 MPa 409 psi
5 3 days 3.4 MPa 497 psi 3.5 MPa 507 psi
6 7 days 3.9 MPa 571 psi 4.0 MPa 581 psi
7 28 days 5.0 MPa 718 psi 4.3 MPa 626 psi

Table 11.10 Variability of type F fly ash with regards to viscosity retention.

Fly ash is a byproduct of coal combustion, and the product specifications are
loosely written

Flyash 1 Flyash 2
Metric Imperial Metric Imperial
ASTM
Item Description code Qty UoM Qty UoM Qty UoM Qty UoM
A Viscosity retention D6910
(no accelerator)
1 Initial 38.1 sec 36.6 sec
2 24 hours 56.3 sec 36.4 sec
3 48 hours 73.0 sec 36.7 sec
 Annular backfill grouting of precast segmental tunnel linings 653

Table 11.11 Mix design used to test bentonite variability.

Mix design used to test various bentonites shown on Figure 11.9

Component quantities
Metric Imperial
Item Mix design components Quantity UoM Quantity UoM
A Primary materials for 1.0 m 3 for 1.0 CY
1 Cement 372 kg/​m 3 641 lbs/​C Y
2 Bentonite 35 kg/​m 3
60 lbs/​C Y
3 Water 867 kg/​m 3
178.95 gal/​C Y
B Admixtures
1 Hydration stabilizer–​r etarder 5 L/​m 3 132 oz/​C Y

Figure 11.9 Variability of bentonite by source and manufacturing plant.

Note: Have mix designs ready for different sources if the primary supplier is not reliable.

the grout mix. As an example, various bentonite sources and one bentonite from the same
source but different manufacturing plants were tested using the same grout design shown
in Table 11.11.
As shown in Figure 11.9, no two bentonites had the same effect on bleed and even ben­
tonite sourced from the same quarry performed differently depending on the manufac-
turing plant.
654 Handbook of Precast Segmental Tunnel Lining Systems

11.7.3.4 Fine aggregates

Fine aggregates (commonly called sands) will vary by mineralogy, gradation and particle
shape. Gradation and particle shape will affect the flow, pumping properties and strength
properties of the grout. A gap graded sand may flow better than a uniformly graded sand.
The uniformly graded sand may pump better than a gap graded sand. Angular aggregates
usually flow less than rounded aggregate but can provide a higher strength. Cementitious
grouts should have the fine aggregate tested for alkali silica reactivity.

11.7.3.5 Admixtures
Use admixtures meeting the requirements of ASTM C494, the Standard Specification for
Chemical Admixtures for Concrete, and ASTM C1017, the Standard Specification for
Chemical Admixtures for Use in Producing Flowing Concrete, as required to improve
pumpability, control time of set, to hold sand in suspension, and to reduce segregation and
bleeding. Do not use foam or admixtures that promote steel corrosion. Admixtures have
various chemistries that may not be compatible with other ingredients in the mix and must
be tested for compatibility and mix stability.
Accelerators used can vary depending on the type of grouting. The most common accel-
erator used in annulus grouting is sodium silicate. Sodium silicates vary by density (solids
content). As density increases, so does viscosity, and this will affect the pumping and dosing
of the additive. There are many sources of sodium silicate. Testing is required to determine
the performance of the grout mixture. The use of other types of accelerators is not typical.
Retarding additives can be grouped into two categories. These are set retarding and
hydration control additives. Set retarders typically are sugar-​based products. The sugar
source is variable. These can be efficient in the grout mixture design. They are less efficient
in lower water to cement ratio grouts as dosages can cause issues in mixture performance.
In two-​component grouting, they are effective and can extend the workability of the unac-
celerated grout mixture for many days.
Hydration control additives retard the setting by controlling the hydration of Portland
cement and other cementitious materials. They are more efficient and predictable at lower
water to cement ratios. Testing is required to determine the effect they have on the grout
design. A hydration controlling admixture will stop cement hydration for a defined time
interval by forming a protective barrier around cementitious particles. This barrier effect-
ively delays the hydration process and initial set and helps to maintain the flow time
of grout
Plasticizers can be various chemistries including lignosulfonates, gluconates, sulfonated
naphthalene formaldehyde condensates (BNS), sulfonated melamine formaldehyde conden-
sates and polycarboxylates.
Lignosulfonates are products of the wood pulp industry. In cementitious systems, they
act as a plasticizer and water reducer and can have a retarding effect on the system. Many
companies provide lignosulfonate-​based additives that are formulated to provide various
performances. As an example, they can be designed to keep the retarding effect or to provide
a normal set time development. Sodium gluconate is a retarding additive that also provides
a plasticizing effect. Testing is required to determine the effect they have on the grout design.
Sulfonated naphthalene formaldehyde condensates (BNS) and sulfonated melamine for-
maldehyde condensates are commonly referred to as high-​range water reducers (HRWR).
They can provide much lower water contents for the same consistency and/​ or higher
slumps and flows at the same or slightly lower water contents. They do not greatly retard
 Annular backfill grouting of precast segmental tunnel linings 655

cementitious mixtures like the lignosulfonate products. If overdosed in the grout, it can
create problems such as bleeding, segregation and retardation. HRWRs are also provided
by many companies and are formulated to provide variable performance. This can include
dose efficiency/​flexibility, strength enhancement, workability retention and rapid disper-
sion. Testing is required to determine the effect they have on the grout design.
Polycarboxylate superplasticizers are highly efficient HRWR additives that typically
allow high water reduction at a low dosage. They can provide much lower water contents
for the same consistency and/​or higher slumps and flows at the same or slightly lower water
contents. They do not greatly retard cementitious mixtures like the lignosulfonate products.
If overdosed in the grout, it can create problems such as bleeding, segregation and retard-
ation. These are also provided by many companies and are formulated to provide variable
performance. This can include dose efficiency/​flexibility, strength enhancement, workability
retention and rapid dispersion. Testing is required to determine the effect they have on the
grout design.
Anti-​washout admixture can be used for grout placed in wet conditions such as between
the precast segmental tunnel lining and the TBM excavated rock surface. However, the use
of anti-​washout additives in two component systems can cause problems with reactivity.
The accelerator may not blend easily with the grout, effecting performance.

11.8 GROUT EQUIPMENT

The equipment required to perform annular grout can be broken down into three phases

• Batching
• Transport through the tunnel
• Grout injection into the annulus

The equipment can vary greatly depending on whether the grout is a mortar type one-​
component grout or a two-​component grout made with only cement and bentonite. Mortar
grouts require concrete style batching equipment at the portal (or shaft) and must be trans-
ported to the heading in grout cars. Two-​component grouts can use a simpler colloidal mix-
ing system and pump the grout to the heading. Regardless of grout type, the entire system
must be sized to provide grout at the maximum rate that the TBM can advance. This can
vary widely depending on whether the TBM is an EPB, slurry or hard rock machine and the
type of ground that the TBM is advancing through. The batch plant is part of the supply
chain feeding the TBM and every part of the chain must have enough capacity to keep the
TBM advancing at its maximum speed.

11.8.1 Batching and mixing equipment

Regardless of grout type, the grout batch plant must store enough ingredients to make
grout continuously regardless of typical delivery interruptions. This requires determining
the reliability of every supplier by looking at their stockpile locations and methods of jobsite
delivery. For example, a batch plant in a metropolitan area may only require a three-​day
supply of cement and admixture but may require additional onsite storage for bentonite
that comes from another country. An extra-​large stockpile of sand may be needed because
the sand quarry may be nearby but perhaps only daytime deliveries at specified hours are
allowed, and the quarry is closed during heavy rain.
656 Handbook of Precast Segmental Tunnel Lining Systems

Different powdered products are delivered by trucks with the same size and type fittings.
The connection ports should be clearly marked. Ports for different ingredients should have
a physical separation to make connecting to the wrong silo less likely to occur.

11.8.1.2 One-​component grout

One-​component grouts are mortars with a powder component that requires silos and a fine
aggregate component that requires storage bins. Admixtures are usually required. All these
items typically get delivered by rubber-​tired vehicles in the form of trucks, front end loaders
and forklifts. There may be a railhead nearby but seldom are materials directly delivered by
rail. The grout discharged from the plant usually goes into a holding tank set-​up for rapid
discharge into a tank car of some type for transport to the tunnel heading. Those inputs and
output dictate the plant location and yard set-​up. The grout batch plant placement must be
coordinated with all the other support equipment such as ventilation fans, segment staging
and muck removal. Shaft jobs are the worst for access because all the support items are
clustered around the shaft collar.
The batched grout usually goes into a holding tank with an agitator to prevent segrega-
tion. The holding tank capacity is at least equal to the amount that will be shuttled in by
the transporter. Movement of the grout from batching to holding tank to transporter is best
done by gravity flow, which is quick and dependable for the fluid grout. For tunnels termin-
ating in a shaft, the batch plant will be erected next to the shaft to allow gravity flow to a
holding tank mounted in the shaft. The tank will be high enough above the tunnel floor to
allow gravity discharge into the transport vehicle holding tank. For portal entry tunnels, the
batch plant must be next to the railroad tracks, or if using a multi-​service vehicle next to the
staging point for loading segments.
The batch plant requires silos to handle the ingredients such as cement, fly ash, bentonite
or other fine powders. These silos will be immediately adjacent to the grout mixer and will
be supplied by delivery trucks that blow the powder into the silos. The truck should be
parked within 5 m (16 feet) of the silo for efficient delivery of the product. Tanker trucks
delivering powder products are usually at the legal maximum length limit and must be
accommodated by the site layout.
Fine aggregates (sand) must be delivered to the batch plant. The sand delivery trucks
require space to turn and should not conflict with muck hauling and segment delivery. If the
grout mix design has several different aggregates, storage bins must be set up for each one.
Considering the proximity of the batch plant to the shaft or portal, the aggregates are often
stockpiled on site but are continuously shuttled to the batch plant hopper using a front-​end
loader.
The batch plant must quickly combine and thoroughly mix ingredients without benefit
of coarse aggregate. Planetary concrete mixers, commonly known as pan type mixers
(Figure 11.10) are favored as they are compact and quickly mix ingredients. The ability to
quickly prepare a batch allows for a smaller mixer that can make multiple batches and keep
up with the rest of the tunnel supply chain.
For consistency, all ingredients except admixtures and water should be weighed.
Admixtures and water are metered. When ingredients are weighed, the primary variable
with the biggest effect on consistency is the fine aggregate water content. A microwave
type measuring device inside the pan mixer can accurately determine the water content and
make adjustments to the water addition. Water content can be determined manually using
ASTM C566 but the method is slow and the moisture content will vary within the aggre-
gate stockpile.
 Annular backfill grouting of precast segmental tunnel linings 657

Figure 11.10 Pan type planetary mixer.

Note: Fast and efficient.

Because the batch plant is a link in the supply chain that keeps the TBM going, consist-
ency and reliability are key. Modern batch plants still require an operator to monitor the
operation and maintain the plant, but the batching process should be computer controlled.
Except for the absence of coarse aggregate, the batch plant is no different than any other
concrete batch plant and should be set up and operated with the same standards. After ini-
tial set up the batch plant should be certified to National Ready Mix Concrete Association
(NRMCA) or equivalent standards.
A storage tank with mild agitation is required so that the supply train or transporter can
be instantly filled and sent back into the tunnel. Although a good mix design will resist
bleeding and segregation, a paddle or ribbon mixer will ensure that a consistent mix is
delivered to the TBM. The storage tank should be at least equal in volume to the trans-
porter tank.

11.8.1.2 Two-​component grout

Two-​component grouts only require bentonite, cement, water and admixture(s). They are
much simpler to set up and have a much smaller footprint overall. Because the grout is so
fluid and is sometimes pumped directly to the heading, the plant does not have to be located
immediately by the shaft or portal. The plant can be set up at a location where access is
better for the delivery of the powdered ingredients. The silos need to be larger or more
numerous since the volume of powdered ingredients used is greater. If grout is sent to the
heading using a pipeline and the storage tank is on the TBM, the holding tank at the batch
plant need only be large enough to maintain a steady flow through the pipeline.
A colloidal grout mixer (Figure 11.11) is used to combine the ingredients. A colloidal
mixer has a tank with a recirculating volute (centrifugal) pump. The rotor in the pump is
specially designed to resist the abrasion from the cement and to provide excellent mixing of
the powdered products with water. The water is added to the tank first. As the powdered
658 Handbook of Precast Segmental Tunnel Lining Systems

Figure 11.11 Containerized colloidal mixing unit and holding tank.

Note: Colloidal mixer on left with mixing pump, holding tank has paddle mixer.

ingredients are added, the shearing action of the pump on the water and powder creates a
uniform, well blended mix in just a few minutes. Colloidal grout mixers come in various
sizes and can produce from 2 to 100 m3/​hr (3 to 130 CY/​hr). The same pump mixing the
grout can discharge the grout to the holding tank very quickly. The powdered ingredients
should still be weighed to ensure a consistent grout, but the handling of fine aggregates is
eliminated along with the need to determine the moisture content.
As with the one-​component batch plant, consistent output and reliability are key. The
batch plant can be operated by computer with one operator to monitor and maintain the
plant. Since the number of ingredients is reduced and the powdered products have good-​
quality control, the opportunities for problems are reduced. Although the plant is not a
concrete-​style batch plant, the critical items such as the powder scales, the water meter and
the admixture meters should be calibrated according to manufacturer's specifications and
checked regularly.
Occasionally, the two-​component grout will be delivered to the heading via rail or rubber-​
tired vehicle. In these cases, a holding tank with mild agitation is required to load up the
transport vehicle as soon as it shows up at the shaft or portal.

11.8.2 Transporting backfill grout by rail or multi-​service

vehicle (MSV)
One-​component grouts are difficult to pump long distances reliably, so they are transported
by narrow gauge rail car or multi-​service vehicle (MSV) from the shaft or portal to the
heading. As described in Chapter 10, each service train going to the heading carries the
supplies required to advance the TBM one ring and that includes segments and grout. The
grout required can usually be carried in one open top grout car. The grout car has an agi-
tator to keep the grout from segregating, which can happen on a bumpy ride to the heading.
 Annular backfill grouting of precast segmental tunnel linings 659

The agitator is normally ribbon mixer on a horizontal shaft powered by the locomotive’s
hydraulic circuit. Sometimes the mixer is powered by an electrical motor. Once a valve at
the tank car bottom is connected to a pump and the car is emptied with the ribbon mixer
assisting by pushing the grout towards the discharge point. Because the ribbon mixer shaft
is below the top of grout, the bearing seals are exposed to the grout and a failure of the
seal leads to a failure of the shaft bearings. Make sure the bearings have automatic greasing
systems.

11.8.3 Transporting grout by pipeline

The fluidity of two component grout lends itself to pipeline delivery of the grout to the
TBM. The pipeline is usually a 50 or 75 mm (2 to 3 inches) steel pipe. Smaller line diam-
eters are preferred to keep the fluid velocity up and cut down on solids build-​up in the pipe.
A minimum grout velocity of 1.2 m/​sec (4 feet/​sec) is recommended. Because the pipeline
may extend thousands of meters and the pumping may be interrupted for hours at a time,
the grout usually contains a hydration controlling admixture and an admixture to prevent
bleeding and segregation. The bentonite in the admixture reduces bleeding but too much
bentonite will reduce the grout strength. The pipeline will discharge into a holding tank on
the TBM trailing gear, so the pump does not need to match the TBM grout consumption
rate during advance. Instead, the pump must be sized to match the entire TBM advance
cycle, which usually includes 15 to 45 minutes to erect a precast liner ring. The pump must
be of the positive displacement type and could be of the peristaltic, progressing cavity or
piston type.
The accelerator for the two-​component grout can be sent to the TBM via a dedicated
pipeline or it can be delivered to the TBM in 1.0 m3 (1.3 CY) containers (totes). Because the
amount of accelerator required is a fraction of the grout, bringing a tote to the TBM on the
supply vehicle is the normal method.

11.8.3.1 Peristaltic pump

The peristaltic pump (Figure 11.12) operates by squeezing a rubber tube to move the grout
along like toothpaste being squeezed out of a tube. The squeeze hose is U-​shaped in the
pump housing. A multi-​arm rotor with a roller at the tip of each arm presses against the
hose. When the roller rotates, the roller pushes over and against the hose pushing the grout
ahead and creating a pressure. After the roller passes over the hose, the deformed hose snaps
back to its original shape, which produces a low-​pressure zone that draws more grout into
the pump. Pump output is varied by changing the rotor revolution rate.
Peristaltic pumps have the advantage that no grout contacts the mechanical drive. The
hose is the only wear item. Peristaltic pumps are available in capacities up to 50 m3/​hr (65
CY/​hr) at up to 25 bar (375 psi) although high-​pressure operation reduces the life of the
pump hose. For maximum pump life, an operating range in the 10 to 13 bar (150 to 200
psi) range provides reasonable hose life. If higher continuous pressures are desired, oper-
ating a larger pump at lower RPM will provide better pump hose life. A larger pump will
run slower, and the hose will last longer bur the pump initial cost is higher. Keeping the hose
lubricated with silicon grease extends hose life.
Because peristaltic pumps are self-​priming and can be run dry without damage, they can
be put in series when boosters are needed without holding tanks or complicated controls.
A typical application for a tunnel pipeline would use a 65 to 76 mm (2 to 3 inches) hose
660 Handbook of Precast Segmental Tunnel Lining Systems

Figure 11.12 Peristaltic pump and controls.

Note: Simple and reliable.

size with a rotor speed of 40 to 50 rpm and outputs in the 15 to 25 m3/​hr (20 to 33 CY/​hr).
A foam ball or bullet type pig can be pumped through peristaltic pumps arranged in series.
Other advantages of peristaltic pumps are that there are no seals or valves to wear out
or replace. The pump can accurately measure grout flow within 1%. The pump can move
sanded grouts, but wear is accelerated. The hose must be checked daily for delamination or
bulging. Tests pumping 20% slaked lime show a hose life of 800 to 1,000 hours and a min-
imum life of 400 hours is anticipated. The peristaltic hose can be replaced in 15 minutes.
The pump does produce a pulsating flow, but this is not a problem when pumping into a
holding tank on the TBM.

11.8.3.2 Progressive cavity pump

The progressive cavity pump (PCP) has a solid helical corkscrew-​like rotor that rotates
inside a stator (Figure 11.13). These pumps are also known as eccentric screw or moyno
pumps. The stator has a double helix cavity. The rotor seals tightly against the stator as it
rotates creating a set of fixed-​sized cavities that move along the length of the rotor/​stator
assembly. The pumped material is moved along inside these progressing cavities. The stator
is made from an elastomeric material dimensioned to have a zero clearance or slight inter-
ference fit against the stator.
At the point where the rotor and stator contract, there is a small amount of sliding move-
ment. This area is lubricated by the fluid being pumped. At higher pressures, the sliding seals
will begin to leak some fluid so when pumping higher pressures, more stages are added.
 Annular backfill grouting of precast segmental tunnel linings 661

Figure 11.13 Progressive cavity pump cut-​a way view.

Note: Good for accurate metering of viscous fluids at wide head variations. Image courtesy of
Herrenknecht Tunneling Systems.

Each seal only handles the pressure difference between the adjacent cavities and the pressure
difference is limited to somewhere between 1 and 6 bar (15 to 90 psi). When more abrasive
fluids are pumped, the pressures towards the lower end are used.
PCP have at least two stages (cavities) and the number of stages increases as the pressure
goes up. PCPs get longer and more expensive as pressures go up and stages are added. PCPs
run at low speeds to keep rotor/​stator friction heat low and this leads to physically large
pumps for high-​pressure, high-​volume pumping.
Because the fluid being pumped provides the sliding lubrication between the rotor and the
stator, the pump cannot be allowed to run dry. If allowed to run dry, the stator will heat up,
expand and be damaged in just a few minutes. Running the pump dry is the biggest reason
for PCP failures. PCPs should be set up to stop if dry running is detected.
PCPs excel at handling viscous fluids, accurate metering and providing a smooth non-​
pulsating flow. The stator rotor assembly naturally functions as a valve with a seal between
each progressing cavity, so no check valves are required. The pump will accurately meter
flows over a wide range of head pressures. The volume pumped is directly proportional to
the rotor rotational speed. PCPs have a short life when pumping sanded grouts. PCPs have
some good characteristics and they have been used to pump two-​component grouts through
tunnel pipelines, but they are better suited to be used as metering pumps on the TBM. PCPs
are more expensive than peristaltic pumps with the same pumping capacity.

11.8.3.3 Hydraulic operated piston pump

High-​pressure piston pumps (Figure 11.14) are also used to move grout from the colloidal
mixer to the TBM. These pumps have been used to move grout up to 3,658 m (12,000
feet) at pressures up to 40 bar (600 psi). These dual acting piston pumps are powered by a
hydraulic power pack. The range of pressures and volume available are adjusted by chan-
ging the size of the grout pumping pistons. For example, a 37 kW (50 HP) power pack is
capable of pumping 24 m3/​hr (31 CY/​hr) at 30 bar (450 psi) using 150 mm (6 inches) pis-
tons or 15 m3/​hr (20 CY/​hr) at 55 bar (800 psi). These piston pumps have a long flat pump
curve before dropping off sharply and they should be operated in the flat region of the curve.
These pumps can be connected in parallel to achieve the necessary flow at higher pressures.
662 Handbook of Precast Segmental Tunnel Lining Systems

Figure 11.14 Hydraulic powered piston pump and controls.

Note: Able to pump at high pressures. Photo courtesy of Tecniwell S.r.l.

Typical grout piston pumps used for tunnel supply lines range from 37 to 45 kW (50 to
60 HP) although smaller pumps are available. These pumps are a good alternative to the
peristaltic pump if higher pressures are required and no pumps can be permitted in the tun-
nel. These pumps are reliable although the pistons and cylinders wear and need to be peri-
odically replaced. They are more expensive than a similar sized peristaltic pump.

11.8.3.4 Tunnel grout pipeline

The tunnel grout supply line will typically be assembled from seamless steel pipe and
grooved pipe couplings with ball valves installed at regular intervals. The pipe will have a
groove cut around each end to work with the coupling (Figure 11.15). Pipe is usually in the
50 to 80 mm (2 to 3 inch) diameter range, meets Schedule 40 wall thickness requirements,
and comes in standard lengths of 6.4 m (21 feet). The pipe should be cut groove Schedule
40 and not lightweight Schedule 10 or 20 with a roll groove. Cut groove pipe has the same
internal diameter throughout. Roll groove pipe has a diameter reduction at the end of each
pipe, which is bad for flow resistance and hard on the cleaning pig.
Grooved pipe couplings come in rigid and flexible styles. Rigid couplings are for arrow
straight runs of pipe. Flexible couplings allow deflection at the joints and are best suited
to tunnels, which often have curves. For a 65 mm (2.5 inches) diameter Schedule 40
pipe, flexible couplings are available with pressure ratings from 35 to 70 bar (500 to
1,000 psi).
 Annular backfill grouting of precast segmental tunnel linings 663

Figure 11.15 Grooved pipe coupling assembly.

Note: Drawing courtesy of TBM Supply–​D ivision of BPS Supply Group.

There should be a ball valve in the grout line every 100 to 150 m (330 to 500 feet).
A ball valve should be used instead of a cheaper butterfly valve to allow passage of cleaning
pigs. Gate valves can be used but they are more expensive and have lower-​ pressure
ratings. A valve is required because the grout line must be disconnected every time a new
piece of pipe is added. When the line is broken all the grout in the line back to the last valve
will run out. A bathtub style catch pan may be part of the tunneling scheme to catch and
reuse the drained grout but this pan has a finite capacity. Often the grout is just dumped
on the ground for the discharge water system to clean up. The muck extracted from the
water treatment plant then has a high Ph, which can cause disposal problems. Collecting
large quantities of spilled grout during final tunnel cleanup may also cause disposal issues.
Developing and using methods to minimize grout spillage during pipe extensions and clean-
ing will save time and money.
The valve, coupling and pipe should all have pressure ratings that exceed the pressure that
can occur when the grout line pump is pushing against a closed valve or plug. Otherwise,
the first time the grout plant operator turns the grout pump on, and someone forgets to
open the ball valve, the line will fail explosively somewhere in the tunnel and the other
connections will have been overstressed. A pressure relief valve should be installed just
downstream of the pump outlet, but one should not count on that valve to save the system.
The entire grout line should be composed of quality components from reputable suppliers.
When the grout line is in use, a cleaning pig (Figure 11.16) should be pushed through
the pipeline as often as every shift to wipe the pipeline walls clean of sludge. If the interval
between pig runs is too long, the pig runs the risk of becoming stuck when there is too much
sludge in the line. Running the pig from the surface batch plant to the TBM requires plan-
ning to catch the grout in a grout car or holding tank. Otherwise, the spilled grout in the
heading makes a mess and pumping it back to the surface through the waste discharge line
causes issues to the water treatment plant increase plant operating costs. Use a non-​brush
type pig if using a peristaltic booster pump.
664 Handbook of Precast Segmental Tunnel Lining Systems

Figure 11.16 Pipeline cleaning pig.

Note: Urethane skin with foam core and spiral wire brush.

Running a cleaning pig through a grout line requires a written plan and training of the
workers to prevent injuries. The sudden release of energy occurring when a pig pushing a
plug of material comes out of the pipeline can be unexpected and very forceful!

11.8.4 Grout injection equipment on the TBM and trailing gear

When the one-​component grout is delivered to the TBM on the transporter, the grout is
transferred from the transporter to the holding tank. Unloading usually occurs concurrent
with the unloading of segments. A piston-​type transfer pump moves the grout from the
transporter to the holding tank. Progressive cavity pumps and peristaltic pumps wear out
rapidly when used to pump sanded grouts. Once in the holding tank, piston style grout
pumps are used to force the grout into the annular gap as the TBM advances.
When two-​component grout is delivered to the TBM, it could arrive either by transporter
or by pipeline. In either case, the grout will go into a surge hopper before injection into the
annular gap. Grout injection is done with positive displacement pumps (piston, progres-
sive cavity or peristaltic). The choice of pumps depends on the TBM manufacturer and the
contractor’s preference.
Progressive cavity pumps (PCPs) are preferred by many for two-​ component grouts
because they deliver the grout and accelerator to the mixing chamber with a steady non-​
pulsating flow. As previously described in Section 11.8.3.2, they have many characteristics
that make them ideal metering pumps. On a large TBM there will be several of them on the
trailing gear to split the grout flows and keep the pump sizes manageable (Figure 11.17).
These pumps should be mounted in accessible locations where they can be taken apart
and serviced. PCPs should always be flushed out with water when grouting operations are
shut down to avoid having grout harden inside the stator/​rotor chamber. The pumps have
no rotor to stator clearance and will be damaged if started with hardened grout inside the
progressing cavity areas.
The grout circuit must also have a method to measure grout volume pumped and grout
pressure. Grout volume can be measured by counting the strokes of a piston pump or by
using a magnetic flowmeter with peristaltic or progressive cavity pumps. The flow meters
 Annular backfill grouting of precast segmental tunnel linings 665

Figure 11.17 Progressive cavity pumps.

Note: Pumps on TBM for grout and accelerator metering.

should be mounted vertically to make sure a full flow of grout is read by the meter. Pressure
is measure by an electronic transducer. The flow meter can be anywhere between the pump
and the injection point. For tail shield grouting, the pressure sensor should be at the point
where the grout pipe enters the tail shield. Even then, the pressure sensor will still be about
4 m (13 feet) from injection point.

11.9 PROJECT EXAMPLE OF ONE-​COMPONENT GROUT

The Riachuelo Project in Buenos Aires used a one-​component grout in challenging condi-
tions. Tunnel was driven by a 5.2 m (17.1 feet) excavated diameter EPB TBM advanced in
a dense sand formation under the Rio de la Plata estuary for 12 km (7 miles) at a pressure
of 4 bar (60 psi). A five-​piece segmental liner ring was installed, 1.4 m (4.6 feet) wide and
had a 4.3 m (14.1 feet) inside finish diameter. The annular gap to be filled with grout was
125 mm (5 inch) wide. Main access to the tunnel started in a shaft only 40 m (130 feet)
from the shoreline. The contractor initially intended to use a two-​component (A+​B grouts),
but the owner’s engineer requested that a one-​component grout be used because of previous
bad experience with the A+​B grouts.

11.9.1 Batch plant

The batch plant (Figure 11.18) had stockpiles for fine aggregate and a 0 to 6 mm (0
to 0.25 inch) gravel that were charged into a weigh hopper. There were three silos for
666 Handbook of Precast Segmental Tunnel Lining Systems

Figure 11.18 Riachuelo Shaft Site and Grout Batch Plant.

Note: Batch plant with aggregate bins and mixer trucks at right, four contiguous shafts at center.

cement storage and two for bentonite. The bentonite was mixed in a colloidal mixer and
pre-​hydrated in a tank prior to combining the ingredients in an 8 m3 (10 CY) concrete
mixer truck.
The mixer truck moved to the shaft and discharged the grout into a hopper at the top
of the shaft (Figure 11.20). Grout then traveled down a 250 mm (10 inch) drop pipe. At
the bottom of the shaft, an energy absorbing chamber received the grout and discharged
through a pipe to a grout tank on the supply train. The 8 m3 (10 CY) delivered to the train
was enough to grout the annulus of two rings.

11.9.2 Delivery to heading

The TBM and trailing gear were serviced by a narrow-​gauge rail system. Muck removal was
by conveyor belts. Each supply train brought to the heading grout, segments and supplies to
the heading for the advance of two liner rings. The grout transport tank (Figure 11.19) had
a horizontal axis ribbon mixer to keep the grout mixed. The mixer was also used to force
the grout towards the discharge pipe end of the tank.

11.9.3 Trailing gear equipment

The TBM had the following equipment on the trailing gear (see Figure 11.21):

• Transfer pump from rail tank to mortar tank: 1 unit

• Pumps for grout injection through tail shield: 2 units
• Injection pump maximum capacity: 167 l/​min (44 gal/​min)
 Annular backfill grouting of precast segmental tunnel linings 667

Figure 11.19 Mixer truck discharges grout.

Note: Grout is dropped to shaft bottom via a 250 mm (10 inch) line.

Figure 11.20 Grout tank loading zone.

Note: Grout transport tank is quickly filled with 8 m 3 (10 CY) of grout.
668 Handbook of Precast Segmental Tunnel Lining Systems

Figure 11.21 Unloading the grout on the trailing gear.

Note: Piston pump moves the one-​c omponent grout to the holding tank.

• Maximum injection flow: 167 × 2 pumps: 354 l/​min (88 gal/​min)

• Mortar tank capacity: 8 m3 (10.2 CY)

All the grout pumps were the piston style positive displacement similar to a concrete pump.
Initially there was no waiting time for supply trains since a fresh resupply train could reach the
TBM before the ring build was complete. Once the TBM began to wait on trains, a passing place
(rail switch) was installed in the tunnel to allow a resupply train to wait closer to the TBM.

11.9.4 Grout injection

The required system capacity was calculated as follows:

• Maximum excavation advance: 100mm/​min (4 inch/​

minutes)
• Minimum excavation period per ring: 14.1 minutes
• Theoretical grout volume per ring: 3.66 m3 (4.68 CY)
• Maximum injection volume considered (115% of theoretical): 4.21 m3 (5.39 CY)
• Required injection flow rate: 4,210 /​14.1 =​299 l/​min (79 gal/​min) < 354 l/​min
(88 gal/​min)

The TBM had four pairs of grout ports set up for grouting through the tail shield. The
grout ports were located at about 10, 2, 4 and 8 o’clock positions. At each pair of grout
ports, one was in active use and the other was a spare. Each grout port was connected to a
grout pipe embedded in the tail shield that had a connection at the front of the tail shield.
The spare grout ports and pipe were normally filled with grease to keep them clean and
 Annular backfill grouting of precast segmental tunnel linings 669

Figure 11.22 Grout injection control panel screen shot from a Herrenknecht TBM.
Note: The injection system in manual mode. Only top two ports are active.

empty of grout until needed. The pressure sensor for each grout port was located at the
front of the tail shield where the grout pipe enters the tail shield, so the pressure sensor
measured the dynamic head and static head during a pump stroke. The static grout pressure
in the annulus 3.5 m (11.5 feet) away from the injection point was measurable only when
the grout was not moving. The maximum injection pressure was initially set at 6 bar (90 psi)
based on the formula below and using 4 bar (60 psi) as the reference pressure.

Pmax (bar) =​(Ptarget +​0.5 bar) × 1.35 (11.1)

The grout could be injected using either the manual or automatic mode. In the manual
mode, the injection pumps were activated individually from the control panel and the
pumping speed could be adjusted. Start-​up and cut-​off pressures were preset but could be
adjusted if required. The pump had to be restarted manually when the injection pressure
value fell below the preset limits. Figure 11.22 shows one page of the touch screen grout
control panel.
In the automatic mode, the grout injection pumps started and stopped by operation from
the control panel. The start-​up and cut-​off pressures were preset, and the injection process
stopped when the cutoff pressure was reached in any of the injection lines. Process would
restart when the pressure fell below the cut-​off value.
The computer calculated the theoretical volume of mortar required based on the advance
rate of the TBM. Relationship between the actual volume injected and the theoretical was
constantly monitored and reviewed. Grout volume parameters were set as follows:
670 Handbook of Precast Segmental Tunnel Lining Systems

• Low volume alert:   95% of the theoretical injection volume

• Low volume alarm: 90% of the theoretical injection volume
• Upper volume alert: 110% of the theoretical injection volume
• Upper volume alarm: 115% of the theoretical injection volume

The automatic system also monitored grout injection pressures.

• Low-​pressure warning: 85% of the theoretical injection pressure

• Low-​pressure alarm: 65% of the theoretical injection pressure
• High-​pressure warning: 115% of the theoretical injection pressure
• High-​pressure alarm: 135% of the theoretical injection pressure

The computer stopped grout injection when

(a) The maximum injection pressure was reached and did not drop or
(b) The maximum injection volume was reached

If (a) or (b) occurred, the TBM advance automatically stopped. Then the field supervision
checked the excavation (tons measured by the scales) and grout injection (injected m3) data.
The individual ring was checked but previous three rings were also checked to determine the
cause of the alarm and look for trends in the data.

11.9.5 Grout issues and resolution

The tunnel was successfully completed in 2019. Mechanical systems selected for mixing,
grout delivery and injection worked well. The primary problems with the grout lay with the
initial grout mix design, which was subsequently redesigned in several steps.
The grout mix developed at the start of the project (Table 11.12) was intended to be a
semi-​inert mortar following design principals espoused in Linger, Cayrol and Boutillon’s
paper TBM’s backfill mortars–​Overview–​Introduction to Rheological Index [2]‌. Ingredients

Table 11.12 Initial annular grout mix design for the Riachuelo Tunnel.
Mix was considered unpumpable by the field supervision

Component quantities
Metric Imperial
Item Mix design components Quantity UoM Quantity UoM
A Primary materials for 1.0 m 3 for 1.0 CY
1 Cement 80 kg/​m 3 138 lbs/​C Y
2 Fly ash 250 kg/​m 3
431 lbs/​C Y
3 Aggregates 0 to 6mm (0 to 0.25-​inch) 918 kg/​m 3
1,581 lbs/​C Y
4 Aggregates fine 795 kg/​m 3
1,369 lbs/​C Y
5 Water 230 kg/​m 3
47 gal/​C Y
B Admixtures
1 Retarder/​s uperplasticizer combination   1 l/​m 3 26 oz/​C Y
 Annular backfill grouting of precast segmental tunnel linings 671

Table 11.13 Grout mix design modified by Riachuelo Tunnel field supervision.

One 23 kg (50 lb) bag of bentonite added to a 4 m 3 (5.2 CY) batch of grout

Component quantities
Metric Imperial
Item Mix design components Quantity UoM Quantity UoM
A Primary materials for 1.0 m 3 for 1.0 CY
1 Cement 130 kg/​m 3 224 lbs/​C Y
2 Fly ash 250 kg/​m 3
431 lbs/​C Y
3 Bentonite 5.75 kg/​m 3
10 lbs/​C Y
4 Aggregates 0 to 6mm (0 to 0.25 inch) 728 kg/​m 3
1,254 lbs/​C Y
5 Aggregates fine 826 kg/​m 3
1,423 lbs/​C Y
6 Water 235 kg/​m 3
54.7 gal/​C Y
B Admixtures
1 Retarder/​s uperplasticizer combination 1.0 l/​m 3 26 oz/​C Y

in the proposed mix were crushed gravel and natural river sand in combination with cemen-
titious material to form a paste with a low initial compressive strength and high plasticity.
The mortar was intended to move from a fluid and thixotropic behavior to stiff mortar
almost as soon as it was injected into the annulus. Injecting the mortar under pressure was
expected to expel part of the free water to help stiffen the grout. The mix was intended to
have an unconfined compressive strength of 0.2 MPa (29 psi) at 24 hours and 1 MPa (145
psi) at 28 days.
Supervision modified the mix by adding 5.75 kg/​m3 (10 lbs/​CY) of powdered bentonite
[one 23 kg (50 lb) sack in a 4 m3 (5.2 CY) grout batch]. Because the manufactured gravel
was jagged and elongated and did not have the round shape of natural pea gravel. More
natural fine sand was substituted for the 0 to 6 mm (0 to 0.25 inch) manufactured gravel.
Also, more cement was added resulting in the mix shown in Table 11.13.
Using bentonite provided lubrication and prevented bleeding, but the mix was still diffi-
cult to pump and slow to set. High pump pressures were required to move the grout through
the lines. Natural pea stone to replace the 0 to 6 mm (0 to 0.25 inches) manufactured gravel
was not commercially available. Considering the unreliable fly ash supply and the use of
manufactured gravel, the final modified mix design was as listed below in Table 11.14.
Pre-​hydrated bentonite slurry provided a grout with more consistent properties over
time. The 5% bentonite slurry resulted in a total water content of 323 kg/​m3 (67 gal/​CY).
Bentonite addition was 7.25 kg/​m3 (12.2 lbs/​CY). Fly ash was eliminated and the 400 kg/​m3
(675 lbs/​CY) of cement compensated for the rough shape of the 0 to 6 mm (0 to 0.25 inch)
gravel. This one-​component active mix was successfully used for the rest of the tunnel drive.

11.10 PROJECT EXAMPLE OF TWO-​COMPONENT GROUT INJECTED

THROUGH THE TBM TAIL SHIELD
The Anacostia River CSO project in Washington D.C. (United States) is a good example of
a TBM that used a two-​component (A+​B) accelerated grout while advancing in heavy clay
and clayey sand conditions. The 7.94 m (26 feet) excavated diameter EPB TBM advanced
672 Handbook of Precast Segmental Tunnel Lining Systems

Table 11.14 Final one-​c omponent active grout mix design for the Riachuelo Tunnel.
This mix used more cement to work with locally available materials

Component quantities
Metric Imperial
Item Mix design components Quantity UoM Quantity UoM
A Primary materials for 1.0 m 3 for 1.0 CY
1 Cement 400 kg/​m 3 689 lbs/​C Y
2 Fly ash
3 Pre-​h ydrated bentonite slurry at 5% 147 kg/​m 3 253 lbs/​C Y
4 Aggregates 0 to 6 mm (0 to 0.25 inch) 757 kg/​m 3
1,304 lbs/​C Y
5 Aggregates fine 606 kg/​m 3
1,044 lbs/​C Y
6 Water 185 kg/​m 3
38.15 gal/​C Y
B Admixtures
1 Retarder/​s uperplasticizer combination 2.28 l/​m 3 60 oz/​C Y

Table 11.15 Two-​c omponent grout properties specified and test results.

This mix used more cement to work with locally available materials

Test results
Specifications Test results
Project specification for
Item fresh grout Metric Imperial Metric Imperial
A Fresh grout
1 Viscosity 48 sec 46 sec
2 Bleeding (1 hour) <5% 0
3 Gel Time 8-​1 2 sec 10 sec
B Strengths
1 1 hour >0.1 MPa >14.5 psi 0.2 MPa 29 psi
2 1 day 0.95 MPa 138 psi
3 7 days 1.65 MPa 239 psi
4 28 days >1.72 MPa >250 psi 1.95 MPa 282 psi

for 3.75 km (2.25 miles) at a pressure of 3.5 bar (50 psi). Six-​piece segmental liner ring was
1.8 m (6 feet) wide and had a 7.01 m (23 feet) inside finish diameter. The annular gap to be
filled with grout was 133 mm (5.25 inches) wide. Tunnel started in a shaft and passed under
a river for 460 m (1,500 feet). The grout was batched at the TBM starting shaft and was
delivered to the TBM via a 75mm (3 inch) pipeline.
Grout was a combination of cement, bentonite and water. There was enough retarder in
the mix to keep the grout fluid in the pipeline for 72 hours. The grout proportions are listed
in Table 11.4. Contract requirements and the test results of the laboratory trial batch are
given in Table 11.15.
 Annular backfill grouting of precast segmental tunnel linings 673

11.10.1 Batch plant

The batch plant (Figure 11.23) was located near, but not immediately at the shaft collar,
providing more room for segment deliveries and much hauling. The shaft was serviced by a
rail type gantry crane, so the elimination of the batch plant next to the shaft helped provide
a clear trackway for the crane operations.
The batch plant had a 2 m3 (528 gallon) colloidal mixer (Figure 11.24) capable of auto­
matically batching and mixing 30 m3/​hr (40 CY/​hr). Powder ingredients and water were
weighed while the stabilizer/​retarder was metered. The mixer discharged to a 4 m3 (1,057
gallons) holding tank with mechanical agitation. For winter operations, the mixer, holding
tank and booster pumps were under an enclosed heated tent. Other equipment at the batch
plant included the following:

• Cement silo: 100 ton

• Cement silo: 75 ton
• Bentonite silo: 100 ton
• Sodium silicate tank: 2 each × 20 m3 (26 CY)
• Stabilizer/​retarder tank: 10 m3 (13 CY)

Batch sequence is shown in Figure 11.25.

Figure 11.23 General site layout with weatherized batch plant and silos.
Note: Because the grout can be pumped, the batch plant does not have to be next to the shaft
allowing flexibility in layout.
674 Handbook of Precast Segmental Tunnel Lining Systems

Figure 11.24 Colloidal grout batch plant.

Note: Shipped in a container for quick job set-​u p. Photo courtesy of Tecniwell S.r.l.

Figure 11.25 Grout batching and delivery sequence.

Note: Only one type of cement was used on Anacostia River Tunnel

11.10.2 Delivery to tunnel heading

A pump at the grout plant was capable of pumping the grout 3,658 m (12,000 feet) to the
trailing gear holding tank through a 75 mm (3 inch) pipeline. The pump was a piston style
positive displacement pump with the following characteristics:
 Annular backfill grouting of precast segmental tunnel linings 675

• Max Flow 250 l/​min (66 gpm)

• Max Press 50 bar (725 psi)
• Rated Power 37 kW (50 HP)

Accelerator was brought to heading by rail flatcar in 1 m3 (273 gal) containers filled from
the 20 m3 (5,300 gal) tank on the surface.

11.10.3 Trailing gear equipment

The TBM had the following equipment on the trailing gear:

• Pumps for grout injection through tail shield: 4 each

• Grout injection pump maximum output: 115 l/​min (40 gal/​min)
• Accelerator injection pump maximum output: 17 l/​min (4.5 gal/​min)
• Maximum injection flow: (115+​17) × 4 pumps: 528 l/​min (139 gal/​min)
• Grout tank capacity: 8.5 m3 (10.9 CY)

All the grout and accelerator pumps were the progressive cavity type.

11.10.4 Grout injection

The required system capacity was calculated as follows:

• Maximum excavation advance: 100mm/​min =​> 18.29 min/​

ring
• Theoretical grout volume per ring: 7.15 m3 (4.68 CY)
• Max injection volume required (120% of theoretical): 8.58 m3 /​ring (5.39 CY/​ring)
• Required injection flow rate:    8,581/​18.29 469 l/​min <
528 l/​min (139 gpm)

Like the Riachuelo TBM, the TBM had four pairs of grout ports set-​up for grouting
through the tailshield located at about 10, 2, 4 and 8 o’clock. At each pair of grout ports,
one was in active use and the other was a spare.
The grout control panel was a touchscreen separate from the TBM operator’s controls.
This allowed the grout operator to monitor the grouting and the ground conditioning.
Figure 11.26 shows a page from the touch screen control panel.
The TBM PLC calculated the required grout volume based on the TBM advance. After
the target volume was reached, the PLC continued to pump grout until the target pressure
was reached as indicated by the pressure transducer on the discharge side of the pump.
At the start of an excavation stroke, the TBM operator would signal the grout plant
operator to commence the grouting operation. The grout was pumped first, followed, after
a short delay, by the accelerator. Mixing of the grout and accelerator occurred before the
grout exited the ports at the end of the tail shield.
The grout injection flow rate was regulated by the grout plant operator during excavation
to maintain a pressure range. When the TBM excavation stroke neared completion, the
TBM operator informed the grout operator who stopped the accelerator pump followed by
the grout pump. The accelerator pump was turned-​off about 100 mm (4 inches) before the
end of the excavation stroke to prevent plugging the tail shield line. Immediately after, the
676 Handbook of Precast Segmental Tunnel Lining Systems

Figure 11.26 Grout injection control panel screen shot from the Herrenknecht TBM.
Note: Another page allowed the operator to set the grouting parameters

grout operator backflushed the system with water to prevent the grout lines from getting
blocked.
Grout pressures were measured continuously by pressure sensors located near the pump
on the TBM trailing gear. They were connected to the PLC, automatically stopped the
grout pumps if the pressures reached the maximum limit. The grout pressures were regu-
larly reviewed based on the actual quantity of grout being placed, monitoring of ground
movement and segmental lining movement. The volume and pressure of the grout and
accelerator was recorded automatically during advance. For immediate feedback, the
grout operator kept manual records of grout and accelerator consumption for every ring.
Grout pressures and quantities were set with percentage limits similar to those set for the
Riachuelo system.

11.10.5 Keeping the system clean

Over weekends, holidays or extended stoppages the grout lines were cleaned and flushed
with water. To prevent the grout in the lines from hardening, the pipes and hoses of the
components on the TBM had to be checked and wiped cleaned at regular intervals with
foam balls. The storage tank also had to be cleaned-​out.
The grout lines in the tail shield were cleaned-​out with a power washer and a sewer style
cleaning nozzle at the end of the hose. A thorough cleaning is essential as minor build-​ups
of grout can accumulate over time and block the entire line causing delays during TBM
production operations.
Although the grout was designed to have a 72-​hour life, grout standing still in the tunnel
supply line could bleed with the heavier cement settling on the pipe invert leaving a sludge
layer in the line. This could lead to plugs in the grout lines, so the lines were cleaned out
with a pig at the end of the last working day of the week, or any time the TBM was due to
 Annular backfill grouting of precast segmental tunnel linings 677

be shut down for extended periods. Ideally end of the last mining shift was planned so that
pig chased out the grout need to complete the TBM advance so that only small amounts
were wasted. This was an important consideration as every 1,000 m (3,300 feet) of 75 mm
(3 inch) line held 4.4 m3 (5.6 CY) of grout and the final length of the pipeline was 3,658 m
(12,000 feet).

11.10.6 Grout issues and resolutions

The two-​component mix design was very successful and there were no issues with the grout
mix itself. Figure 11.27 shows how well the annulus crown space was filled by the grout.
The batching system with the colloidal mixer was very reliable as was the high-​pressure
piston pump that pushed the grout through the tunnel pipeline. Keeping buildup off the
pipeline walls was done by running a bullet style cleaning pig through the line. The foam
balls commonly used for short concrete pump lines were inadequate. Cleaning was done on
the last shift of the week and before any major TBM shutdown.
The grout holding tank and the associated injection pumps were about 65 m (213 feet)
away from the tail shield connection point. The four grout lines were hung in a vertical con-
figuration, which made access to any single line for disassembly and clean out difficult and
time consuming. Most of the grout plugs occurred between the injection grout pumps and
the tail shield grout injection ports. Having the grout pumps closer to the heading would
have made finding the plugs and cleaning the lines easier.
The biggest problem was the grout spilled in the invert every time the tunnel grout supply
line was disconnected to extend the line by an additional 6 m (20 feet). There were cut-​off
valves mounted every 100 m (330 feet) to prevent the entire line from draining and a tank
to catch the spilled grout but this was not enough. Occasional clean outs of the entire system

Figure 11.27 Complete annular grout filling at the crown.

Note: The TBM advanced through a shaft and the tunnel segments were later removed revealing
the grout.
678 Handbook of Precast Segmental Tunnel Lining Systems

also added to the mess. Spilled grout was washed-​down the invert and pumped out through
the waste discharge line to the water treatment plant. The sludge in the water treatment
plant would reach a pH over 12 and required the use of sulfuric acid to adjust the pH.
The cement and bentonite sludge were difficult to handle and expensive to dispose. For the
follow-​on tunnel contract which was 8.2 km (5.1 miles) long, the tunnel grout pipeline was
eliminated in favor of grout cars.

11.11 PROJECT EXAMPLE –​ PEA GRAVEL WITH FLOOD GROUT

This tunnel is a good example of a precast segmental liner placed in a rock tunnel back-
filled with pea gravel and then flooded with grout. Located in Canada, the tunnel was a 7.6
km (4.6 miles) long, 6.56 m (21.5 feet) excavated diameter tunnel driven in rock using a
shielded TBM. The precast concrete lining has an inside diameter of 5.72 m (18.75 feet) and
a thickness of 255 mm (10 inches). The annulus required 3.30 m3 (4.3 CY) of pea gravel and
grout combined per meter of tunnel or about 25,000 m3 (32,650 CY) for the whole tunnel.
The pea gravel and grout mix ingredients are listed in Table 11.5.

11.11.1 Pea gravel placement

11.11.1.1 Pea gravel delivery system
The pea gravel was stockpiled in covered storage locations to prevent the introduction of
contaminants. A steam jenny and steam lines installed within the base of the piles prevented
freezing and clumping in cold temperatures. Each tunnel liner supply train had two pea
gravel rail car chassis. One chassis carried a full pea gravel hopper and each hopper had two
compartments (Figure 11.28) to allow direct feed into two pneumatic shotcrete pots that
acted as pea gravel injectors.
Each supply brought to the heading segments for one liner ring and a rail hopper car of
full of pea gravel to fill the annulus. The supply train entered the tunnel with one chassis
empty to receive the empty hopper and a second chassis carrying a full hopper for exchange
at the heading. The train positioned itself on the trailing gear to allow the empty pea gravel
hopper to be lowered from its position overhead in the pea gravel gantry onto the empty
chassis. The train is then moved ahead, and the full hopper was raised into position while
the TBM advanced commenced, and the first muck car was filled. Pea gravel injection
started after the full pea gravel hopper was secured.

Figure 11.28 Removable hopper on rail car chassis.

Note: Full hopper switched out for empty one at the heading.
 Annular backfill grouting of precast segmental tunnel linings 679

Figure 11.29 Shotcrete pot under hopper.

Note: Pot blows gravel to injection point.

11.11.1.2 Pea gravel placing system

Two shotcrete pots rated at 13 m3/​hr (17 CY/​hr) were used to blow the pea gravel from
the hopper to the injection point (Figure 11.29). Pea gravel flow to the shotcrete pots was
regulated by a guillotine-​controlled chute. One shotcrete pot alone was able to keep up
with the TBM advance. If one pot failed, the second pot was used to support the advance,
by switching from one side of the crown to the other (to avoid unbalanced support) with
its hose, until the second pot was returned to service. The pea gravel was a natural stone so
that it could easily be blown in place.

11.11.1.3 Location and sequence of gravel injection

The liner segments had a threaded port cast into the lining for connection to the pea gravel
and grout injection hoses. When a grout port was selected for injection, the bottom of the
port was drilled out to remove the concrete plug. Custom-​made adapters connected the dis-
charge hose to the injection ports. The injection ports in the segments did not have check
valves installed because the check valves would have been destroyed by the fast-​flowing pea
gravel.
The pea gravel hoses were connected (Figures 11.30 and 11.31) from the first trailing gear
platform, three, four or five rings away from tail shield, at the 11 and 1 o’clock positions so
that the natural angle of repose of the pea gravel tied into the tail shield at the springline.
The pea gravel was blown in and then the hoses were moved ahead and reconnected after
the TBM advanced one ring. The goal was to fully support the ring leaving the tail shield
up to spring line with no signs of distress or deflection. The hose was periodically connected
680 Handbook of Precast Segmental Tunnel Lining Systems

Figure 11.30 Pea gravel connector adapter.

Note: Connects gravel injection hose to segment injection ports.

Figure 11.31 Pea gravel injection hose connected to segment.

Note: Using the port closest to the crown.
 Annular backfill grouting of precast segmental tunnel linings 681

six to eight rings behind the tail shield at the 12 o’clock position to fill the crown with pea
gravel. An open face of pea gravel fill was kept at injection locations to avoid pea gravel
locking up in the hose. Because a void in the crown could develop at previously injected
rings as pea gravel settled, pea gravel was re-​injected at previously injected ports to ensure
complete filling.

11.11.1.4 Volume measurements

All pea gravel cars were filled to capacity at the portal. Once the pea gravel was blown
behind the segments, any remaining pea gravel in the tank was deducted from the full known
volume. The amount consumed was recorded for each ring injected to verify adequate sup-
port for the liner ring. Pea gravel volumes were monitored and compared to the theoretical
volume. A running average of gravel consumption per ring advance was kept, checking for
trends that would signal a problem.

11.11.1.5 Pea gravel issues and resolutions

The placement of pea gravel within the annulus required additional field testing and adjust-
ments to ensure the process was performed correctly. A drawing of the process is shown in
Figure 11.32. Some issues and the required adjustments were as follows:

• Getting the gravel to flow from the injection port to fill the annulus required consist-
ency during injection.
• Water could be injected at adjacent ports to provide lubrication and carry the pea
gravel to the invert, preventing the material from bridging.
• Directional inserts were welded to the injection port adapters to direct the gravel.
• Hoses from the hard line to the connection ports were steel reinforced and the bends
limited.
• Hose couplings were lengthened to prevent the hose from pulling-​out at the segmental
liner and hardline connections.

Figure 11.32 Pea gravel angle of repose (shown by green line).

Note: Pea gravel provides support for newly installed liner rings up to springline. Background
TBM drawing provided by Herrenknecht Tunneling Systems.
682 Handbook of Precast Segmental Tunnel Lining Systems

• Hose blow outs and line wear would occur. Weak points in the system were addressed
and the workers protected at these locations. Hardened piping was used to transport
the gravel as close as possible to the injection locations.

11.11.2 Flood and contact grout injection

11.11.2.1 Grout mix design development
A simple grout was developed using cement, fly ash, bentonite and water. See Table 11.5
in the sample grout section for the mix design. The mix was tested and adjusted to achieve
the desired viscosity and set time. A pot life of 4 to 6 hours was achieved principally by
replacing much of the cement with fly ash. Viscosity was adjusted using a superplasticizer
and pot life was lengthened using a hydration stabilizing retarder. Target strength for the
grout was 15 MPa (2,175 psi). Agitation within the grout cars and holding tanks was done
to prevent segregation and slump loss. The same grout mix design was used for both flood
grout and contact grout. Bentonite was included to assist in control of bleed and retard
the mix.

11.11.2.2 Grout delivery system to heading

Grout was transported by a single grout car on a supply train from the portal to the TBM
trailing gear. The grout was then pumped into the storage tank on the TBM. From the stor-
age tank, grout was pumped as needed to trailing gear gantries for flood and contact grout
injection.

11.11.2.3 Grout pumping system

Pumps connected to the grout tank delivered grout to the grouting stations through 75 mm
(3 inch) hoses (Figure 11.33). A grout pressure transducer style sensor on each grout line
measured pressures, regulating the pump output pressure to pre-​set limits to prevent over-​
pressure. These same sensors sent information back to a data logging system.
The trailing gear was designed in such a way that train deliveries were fully integrated
with TBM excavation cycle activities. Once the train arrived at the heading, the pea gravel
activities were completely separated from the grout activities.

11.11.2.4 Location and sequence of flood grout injection

The objective of flood grouting was to fill the interstices in the pea gravel that was already
placed in the annulus. Grout was typically pumped through two separate lines from the
holding tank to the injection points at 11 and 1 o’clock about 75 rings (114 m, 375 feet)
back from the tail shield (Figure 11.34). The grout was batched at a consistency that resulted
in a sloping forward toe or “wave” of about 45° within the pea gravel moving towards the
heading. Grouting from Port No.3 only was typical. Grouting from Ports 1 up to 3 was
done in high water flow areas. See Figure 11.34 for port designations.
Grout flow was verified through open injection ports ahead of the active injection points.
The flood grouting was completed back far enough to ensure it did not migrate into the
overcut between the rock bore and the tail skin extrados. Pressure for the flood grout was
adjusted based on performance and was not allowed to exceed the five-​bar design pressure
of the tunnel lining. Initial flood grouting did not achieve a refusal pressure.
 Annular backfill grouting of precast segmental tunnel linings 683

Figure 11.33 Flood grout injection hose connected to segment.

Note: Using the ports closest to the crown.

Figure 11.34 Section view of flood grouting advancing wave.

Note: Grout advance estimated from open grout holes. Background TBM trailing gear drawing
provided by Herrenknecht Tunneling Systems.
684 Handbook of Precast Segmental Tunnel Lining Systems

Segment grout injection ports not previously used for pea gravel injection had to be
drilled out through the last 50 mm (2 inches) of concrete. A plastic check valve was threaded
down into the embedded injection port leaving enough thread to connect to the grout hose
adapter fitting. The check valve allowed the 75 mm grout hose to be disconnected after
injection was completed.

11.11.2.5 Crown check (contact) grout

Further back on the trailing gear a grout injection hose was periodically connected at the
12 o’clock position to ensure pea gravel voids at the crown were filled with contact grout.
Contact grout was injected until a refusal pressure of 2 bar (29 psi) over hydrostatic pres-
sure was achieved. Refusal volume was considered less than one liter per minute over five
minutes. To control the back-​grouting pressure, the computer was set with at the refusal
pressure at an upper limit of 5 bar. The data logging system recorded pressure and volume.
Sufficient set times were required prior to contact grouting as grouting with pressure too
early would push the original grout and give an inaccurate grout take quantity.
The grout injection system was not of the recirculating type but instead the pumps were
controlled and monitored by pressure transducer sensors and flow meters. The pressure
sensors were calibrated, and a test performed to account for any head loss (pressure drop)
between the point of injection and the location along the grout line where the pressure sen-
sor was positioned.

11.11.2.6 Grout volume measurements

Each grouting line was equipped with a pressure transducer grout sensor to monitor and
report to the data acquisition system the injection pressure. A magnetic flow meter on each
line monitored the injection flow rate and volume. The amount of filling, a suitable running
average of volume injected behind the segments (pea gravel and grout) was compared to the
corresponding running average of the theoretical volume confirmed by the volume of tunnel
muck removed.

11.11.2.7 Flood and contact grout issues and resolutions

During flood grouting no significant pressure was typically observed due to the advancing
free face of the grout wave. However, during sections of tunnel in which water inflows were
observed, a pressure was required to displace the water behind the lining. Additional verifi-
cation holes were used along the lining to allow trapped water to be displaced. These holes
also provide visual confirmation the grout wave was achieving the correct angle of repose.
Connecting to all available ports at the 12 o’clock as the TBM advances during pea gravel
injection was necessary to minimize the amount of flood grouting required. The lining was
easily filled from 1 to 11 o’clock leaving a void in the crown section. Without the pea gravel
back fill at the crown, the flood grout flowed forward leaving a void at the crown. This led
to high volumes of grout take during crown (contact) grouting.

11.12 TESTING PROCEDURES

This section will give a general idea of the testing procedures used to develop the grout
mix and later to test for quality. The ASTM procedures are briefly described here but the
 Annular backfill grouting of precast segmental tunnel linings 685

relevant ATSM document should be followed. Some of the tests are not ASTM prescribed
but are helpful to define a grout mix characteristics.

11.12.1 Testing equipment and apparatus

• 50 mm (2 inch) cube molds
• Penetrometer
• Marsh funnel
• Stopwatch
• Compressive strength testing device
• Heavy duty 12 mm (0.50 inch) corded drill
• Ribbon type mortar mixer
• Weigh scale –​1,000 g (2 lb) capacity
• Weigh scale 10 kg (22 lb) capacity
• Thermometer
• Set of plastic graduate cylinders
• Set of 20 liter (5 gallon) buckets with tight fitting lids

All equipment should be properly calibrated prior to any annular grout testing.

11.12.2 Preparing test batches

Use a trial batch grout mix form (Figure 11.35) to calculate the ingredient proportions
required for the test. The form should have spaces for recording the test results as the test
proceeds. If using aggregates, take samples for moisture content determination and run the
moisture test first. Store the rest of the aggregates in sealed 20 liter (5 gallon) buckets until
the trail batch to maintain moisture content.
Procedures are slightly different for one-​component and two-​component grouts. For
two-​component grouts begin by weighing out the ingredients (bentonite, stabilizer, fly ash,
cement) per the trail batch sheet. Start by adding the water to a 20 liter (5 gallon) bucket.
Add the bentonite to the bucket and mix for 60 seconds. Next, add the stabilizer and mix
for 30 seconds. After adding fly ash to the bucket, mix for another 30 seconds. Finish by
adding the cement and mixing for a final 60 seconds. These mixing times are only as a
guideline; however, the times chosen shall be consistent with the ones programmed into the
batch plant.
For one-​component grouts, use the moisture determination to calculate the weights of the
aggregates adjusted for moisture. Put the aggregates water in first followed by the aggre-
gates. Then add the fine powders and admixtures as described in the previous paragraphs.
Do not skip the step of determining the moistures of the aggregates and making the batch
adjustment.

11.12.3 Density
The grout density will be calculated when the trial batch sheet is prepared. Density is a good
quick check to see if the batched grout has the right amount of water. This is especially true
for one-​component mortar grouts where the aggregate moisture content is determined sep-
arately. A mud balance is a good way to check density quickly (see Figure 11.36).
686 Handbook of Precast Segmental Tunnel Lining Systems

Figure 11.35 Trial batch worksheet.

Note: Example trial batch for a one-​c omponent grout mix.

To use the balance, fill the balance cup with the sample to be tested. Tap the side of the
balance cup several times to break up any entrained air or gases. Put the lid onto the balance
cup by pushing it downward with a slow rotating motion until it is firmly seated.
Make sure that some of the test sample is forced out through the vent hole in the lid. Clean
any sample from the outside of the balance cup and lid. Fit the knife edge of the balance
arm into the fulcrum and balance the assembly by moving the rider along the arm. The
mud balance is horizontal when the level bubble fluctuates an equal distance to either side
of the center line. Take the reading from the side of the rider nearest the balance cup. The
 Annular backfill grouting of precast segmental tunnel linings 687

Figure 11.36 Mud balance.

Note: Simple method to check grout density.

measurement reading should be recorded to the nearest 0.01 g/​cm3 (which is equivalent to
specific gravity).

11.12.4 Flowability, viscosity and stabilization time

The methods used depend on the consistency of the annular grout. One-​component grouts
with low mobility are typically tested using a slump measurement. ASTM C143/​C143M
the Standard Test Method for Slump of Hydraulic-​ Cement Concrete represents such
a test. Variations of the test method can be used, depending on the consistency of the
annular grout.
For more fluid one-​component grouts, the spread of the sample can be measured. The
slump cone is inverted, and the spread (diameter) of the grout can be measured. The German
DIN 18555 flow table test combines slump and flow table test. See Figure 11.37 for an
example of measuring grout spread.
A slump is taken and then the table is dropped, and a spread measurement is taken. Be
sure to look at the spread grout to see if the ingredients are uniformly carried-​out to the
edge of the grout.
With very high-​mobility two-​component annular grouts, the consistency can be meas-
ured using ASTM C939 the Standard Test Method for Flow of Grout for Preplaced-​
Aggregate Concrete (Flow Cone Method) or ASTM D6910/​D6910M the Standard Test
Method for Marsh Funnel Viscosity of Clay Construction Slurries. The time of efflux
of a volume of the grout is measured in these procedures. The two methods vary by the
cone orifice and volume of grout tested. ASTM C939 cone (Figure 11.38) uses a 12.5 mm
(0.50 inch) orifice and 1,725 ml (58 fluid ounce) of grout volume for testing. ASTM
D6910 Marsh funnel (Figure 11.39) uses a 6 mm (0.25 inch) orifice and looks at the
length of time to move a quart 946 ml (0.25 gallon) of grout through the funnel.
To perform the ATSM D6910 test, make sure the funnel is clean and dry. Fill the Marsh
funnel’s dedicated cup up to the 1 liter (1 quart) marking. While sealing the end of the fun-
nel, pour the grout through the screen at the top until the full quart has been poured. Obtain
688 Handbook of Precast Segmental Tunnel Lining Systems

Figure 11.37 Measuring grout spread.

Note: A slump cone is inverted and filled with grout, then lifted.

Figure 11.38 ASTM C939 cone.

Note: Use with one component mortar grouts.
 Annular backfill grouting of precast segmental tunnel linings 689

Figure 11.39 ASTM D6910 cone.

Note: Marsh cone for two-​c omponent fluid grouts.

a stopwatch and measure the time from when you release the seal at the bottom and the
grout begins to flow until the full quart has passed through the funnel.
Stabilization time is typically measured using the same methods as for flowability and
viscosity. The loss of these properties is measured over time.

11.12.5 Bleed test

The ASTM C940 bleed test has been modified to use 1,000 ml (34 fluid ounces) or grout
instead of 800 ml (27 fluid ounces) to provide a direct measurement of bleed. To run the
test, fill a 1,000 ml (34 fluid ounces) graduated cylinder with un-​accelerated grout (see
Figure 11.40). Cover the top with either tape, or plastic wrap secured by a rubber-​band
to prevent evaporation. Record the time. After an hour has passed, measure and record
how many milliliters the bleed water has separated from the grout and accumulated at
the top of the grout column. Each 10 ml (0.34 fluid ounces) marking of the graduate
cylinder represents 1% bleed. Recording should be at hourly intervals after the first hour
has passed.

11.12.6 Time of set

Two-​component accelerated grouts require a different technique than much slower setting
one-​component grouts. To test an accelerated two-​part grout, prepare a penetrometer for
use and make sure it is sitting level.

• The weight of rod and cone should be 100 grams (0.22 lbs)
• Measure the amount of accelerator to be used and pour into a 2 liter graduated
beaker (plastic). In a separate beaker, measure 1 liter of the grout minus the amount
690 Handbook of Precast Segmental Tunnel Lining Systems

Figure 11.40 Bleed test.

Note: 1000 ml graduate allows direct reading of bleed.

Figure 11.41 Penetrometer.
Note: Use for time of set test.

of accelerator being used [example, if there is 80 ml (2.7 fluid ounces) of accelerator,

then add 920 ml (31 fluid ounces) of grout].
• Next, start a stopwatch then quickly pour the accelerator into the grout then back
into the empty beaker and again into the other empty beaker. Because it will be hard-
ening very rapidly, immediately pour the accelerated grout into a metal cake pan
(Figure 11.41).
 Annular backfill grouting of precast segmental tunnel linings 691

Figure 11.42 Penetrometer devices for ASTM C403 & C953 testing.

Note: (L) Electronic force meter (R) Vicat test apparatus.

• Keep stopwatch running and position the metal pan under the penetrometer.
• Adjust the height so the tip of the weight sits just on top of the grout. When the stop-
watch reaches a time of 1 minute, 30 seconds, drop the weight into the grout and
measure the distance. Reposition to a clean portion of the grout.
• Repeat this each minute (stopwatch: 2:30, 3:30, 4:30, etc.…).

The test is complete when three tests have a reading of less than 14 mm (0.55 inches).
Record the time at the final reading as the “setting time”.
Time of set for one-​component mortar grouts can be determined using methods such as
ASTM C403/​C403M the Standard Test Method for Time of Setting of Concrete Mixtures
by Penetration Resistance or ASTM C953 the Standard Test Method for Time of Setting of
Grouts for Preplaced-​Aggregate Concrete.
A modified Vicat needle test (ASTM C807) is used to measure the time of setting of
hydraulic cement mortar (Figure 11.42). The elapsed tie after initial mixing of water and
cement required for the mortar to reach a penetration resistance of 500 psi is defined as ini-
tial setting time and the elapsed time to read a penetration resistance of 27.6 MPa (4,000
psi) is defined as final setting time. Penetration resistance is not the same as compressive
strength. A penetration resistance of 27.6 MPa (4,000 psi) corresponds to approximately
0.70 MPa (100 psi). Time of set is not commonly requested for two-​component grouts.

11.12.7 Gel time

Gel time is generally not tested in low mobility grouts. High-​mobility two-​component grout
annular grouts can be measured for gel time. The general test for this property uses two con-
tainers of the same volume (Figure 11.43). Grout is in one container and the proper amount
of accelerator for that amount of grout is in the second container. The two containers are
692 Handbook of Precast Segmental Tunnel Lining Systems

Figure 11.43 ASTM C 403 gel time.

Note: Use for two-​c omponent grouts.

mixed by pouring the grout sample into the accelerator and then back and forth between
the containers until the sample will no longer pour out of the container. The time from the
start of this procedure until its end is measured as gel time.
To perform the test, measure the amount of accelerator to be used and pour it into a 2
liter (67.6 fluid ounces) graduated plastic beaker. In a separate beaker, measure 1 liter of
the grout minus the amount of accelerator being used [example, if there is 80 ml (2.7 fluid
ounces) of accelerator, then add 920 ml (31 fluid ounces) of grout]. Next, start a stopwatch
then quickly pour the accelerator into the grout then back and forth between the beakers.
Do this until the grout “gels” and no longer flows.

11.12.8 Pot life

Fill up a beaker with a few hundred milli-​liters of un-​accelerated grout, and then pour into
a bag. Seal bag and allow it to hang for at least a week. Check the grout regularly for signs
of hardening. Record the time when the grout begins to harden.

11.12.9 Compressive strength

Using a graduated cylinder, measure the amount of accelerator to be used and pour into
a 2 liter graduated plastic beaker. In a separate beaker, measure 1 liter of the grout minus
the amount of accelerator being used [example, if there is 80 ml (2.7 fluid ounces) of accel-
erator, then add 920 ml (31 fluid ounces) of grout]. Make a 50 mm (2 inches) three-​gang
cube mold ready and lubricate the inside with WD-​40 or equivalent. Next, quickly pour
the accelerator into the grout then back into the empty beaker and again into the other
empty beaker. Because it will be hardening very rapidly, immediately pour the accelerated
grout into each cube of the mold (Figure 11.44). Be sure to overfill and then scrape away
 Annular backfill grouting of precast segmental tunnel linings 693

Figure 11.44 Three gang cube mold.

Note: Use to prepare 50 mm (2 inch) mortar cubes. Plastic versions also available.

Figure 11.45 Unconfined compressive strength testing apparatus.

Note: Use to test 50 mm (2 in) mortar cubes.

the excess grout to make a flat top. Cover with plastic wrap and a heavy flat object. Record
the time at pouring.
After the grout has been in the molds 60 minutes, carefully remove the test cubes and
submerge them in a bucket of water kept at 20°C (68°F). Select a cube and place on the
compressive strength testing device (Figure 11.45). Position it so the entire cube is within
the extents of the upper and lower crushing plates. Set the meter to measure the peak force
with the desired unit. Compress the cube and record the reading on the meter. Repeat this
for a second and a third cube. Being sure to have consistent units, convert the force into
pressure by dividing the reading by the area of one side of the cube. Average and record the
three pressure readings. After another 5 hours (6 hours since cubes poured), repeat this test
for a second set of three cubes. Retain the other three cubes in water to be tested 28 days
later by a certified lab.
694 Handbook of Precast Segmental Tunnel Lining Systems

11.13 ANNULAR GROUT QUALITY CONTROL

Annular grout quality control has the objective of ensuring that the annular space is filled
with the correctly batched grout that behaves as designed in the laboratory. Grout that fills
the annular space and stiffens as planned results in precast liner rings maintaining their
circular shape and minimizes ground settlement. The approved grout needs to have field
performance close to what was measured in the laboratory.

11.13.1 Quality control at the batch plant

The first step is to ensure that the ingredients are the same as those used to develop the
mix. Powdered ingredients such as cement, type F fly ash, and bentonite are usually quite
consistent. Mill certifications should be obtained for every delivery, examined, and kept for
record purposes.
One-​component mortar grouts use aggregates, which should have a sieve analysis done
weekly to verify consistency at a minimum. Tests for alkali silica reactivity, abrasion, clean-
liness, flatness, density, soundness are done on sample aggregates before mix develop-
ment work begins. These tests are typically repeated by the supplier on a six-​month basis.
Usually, the largest consumers of aggregates in the area, such as highway departments,
require the tests on a regular basis and the jobsite can obtain copies of the tests. If the aggre-
gates are being produced exclusively for the tunnel project, then a testing schedule must be
established.
Like a concrete batch plant, a one-​component grout plant should produce a printed ticket
for every batch of grout that records the total quantities by weight of powder ingredi-
ents and aggregates, the moisture adjustment for the aggregates, the amount of water and
admixtures added to the batch and the time when the batch left the grout plant. The tunnel
or grout engineer should review these tickets every day to check for any worrisome trends.
Weigh scales and fluid meters should be tested during commissioning and every six months
thereafter. The plant can be inspected and commissioned per the guidelines of the National
Ready Mixed Concrete Association’s Plant Inspector’s Guide.
One-​component mortar grouts are easily sampled at the batch plant, and if no water is
added to the mix during transport, it will be the same as on the TBM. Tests that should be
done daily at the batch plant include the following:

• Density
• Slump
• Set time
• Bleed
• Temperature
• Grout cubes
• Air content

Density measurements made with a mud balance provide a quick way to check the amount
of water in the mix since all other ingredients are weighed and the moisture in the aggregate
is the only variable. Air content should be checked if the grout density is low.
For two-​component grout, the batch plant uses a simple colloidal mixer and there is no
need for moisture adjustments as there are no aggregates. As with the one-​component plant,
the scales and meters should be tested during commissioning and once every six months
thereafter. Tests that should be done at the batch plant include the following:
 Annular backfill grouting of precast segmental tunnel linings 695

• Density
• Viscosity
• Bleed
• Temperature

A quick test with the mud balance will show if the water/​cementitious ratio is cor-
rect. Air content need be checked only if the water and cement have been weighed but
the density is too low. No test samples for compressive strength are taken at the batch
plant because testing without the accelerator added would not be representative of the
injected grout.

11.13.2 Quality control at the TBM

11.13.2.1 Obtaining fresh grout samples of one-​component grout
Making test cylinders or cubes of one-​component grout at the TBM should be avoided as
the handling and transport of grout test specimens from the TBM to the surface will result
in disturbed specimens that will yield unreliable and inconsistent results. If test specimens
must be taken at the TBM, a safe place with no vibration on the TBM must be designated
for initial curing and storage. The test specimen should be made from grout being injected
halfway through a TBM stroke. The test specimens must be brought to the surface by
someone who will take special care of the specimens during transport. Most TBMs do not
have room for safe storage of test cylinders on the trailing gear. Alternatively, a sample pail
of grout can be brought back to the surface for preparation of grout test cylinders or cubes.
This pail of grout must be promptly transported to the surface, reagitated, and the samples
made within the time limits established by laboratory testing. The grout cubes or molds are
then be stored, undisturbed, in an area protected from freezing.

11.13.2.2 Obtaining fresh grout samples of two-​component grout

Two-​component grouts are easier to batch but difficult to sample as the accelerator is only
added at the point of injection. The grout gels too quickly to collect the sample in a bucket
and for transfer to cylinder molds. When the grout is injected through the segments, the
sample collection should start by dumping accelerated grout from the hose nozzle into a 20
liter (5 gallon) bucket to establish a good flow. Once the flow is established, the nozzle can
be directed into a 75 × 150 mm (3 × 6 inch) test cylinder mold for filling and rodding. Rod
the grout quickly to eliminate the voids before the grout gels. Be careful not to create voids
with the rod. Air voids inside the sample that will cause low test results.
Much more difficult is sample collection from a TBM where the accelerated grout is
injected through the tail shield. The sample is obtained by drilling a 25 mm (1 inch) hole
through a segment near one of the injection ports. The grout extrudes through the hole
and is collected directly into a test cylinder where it is immediately rodded. Because the tail
shield grout line is typically flushed at the end of a TBM stroke, watery grout is possible
if the sample is taken at the beginning of a stroke. A more representative sample can be
obtained from the middle of stroke. The filled test cylinder must be handled like a new-​born
baby and allowed to cure undisturbed for a few hours before being moved out of the tunnel.
This method works well for tunnels where face pressures are low (3 bar or less) or where the
native ground is impermeable clay.
696 Handbook of Precast Segmental Tunnel Lining Systems

11.13.3 Filling the annulus

For soft ground excavation where the grout volume is well defined by the annulus gap,
monitoring the grout volume and pressure is the best guide as to whether the annulus is
filled. With a programmable logic controller (PLC) controlled grouting system, the param-
eters for pressure and volume are preset in the system with alarm levels predefined. The
PLC controlled grouting system tracks and records the injected quantities and checks them
as the TBM advances. Trends provide the best guide as to whether a problem is developing
and provides a means to establish what is a normal grout take for the ground conditions.
One-​component grouts use piston-​type positive displacement pumps. The amount of
grout placed is measured by counting the piston strokes. During commissioning the pumps
should be tested to determine the amount of grout moved by each piston stroke and verify
that the PLC has the correct pump parameters. As the pumps wear, they should be checked
to ensure they are still injecting the programmed amount per stroke.
Two-​component grouts and accelerators are pumped using peristaltic or progressive
cavity pumps controlled by flow meters. These flow meters must be checked on a weekly
basis or any time there is a suspicion that the grout is not working as intended. A mis-​
calibrated accelerator or grout flow meter will not inject accelerator in the correct propor-
tions. Mis-​calibrated flow meters can cause false alarms with regards to the amount of grout
placed. Check the weekly test results against test results obtained during commissioning.
A precast segmental ring that maintains circularity is a good sign that grout is filling the
annular gap and stiffening fast enough to provide good support. The precast segments are
designed with raised faces and a gap at the edges to prevent spalling in the event the segment
pieces rotate slightly. These gaps should be measured and documented during ring erection
and observed as the ring comes out of the tail shield as an indicator of circularity. The trail-
ing gear pulled behind the trailing gear blocks the direct measurement of tunnel diameter
to check for roundness. There are systems that use remote reading inclinometers to read the
deformation of a segmental liner ring but these installations take money and time to install
and are not practical to install on every ring.
Another sign of a quality grouting program is the minimizing of subsidence. Extensometers
and recording systems now have enough accuracy to detect in soft ground when a TBM
approaches, when the cutterhead passes, and when the tail shield and grout operation
passes by. Once that point has passed, minimal subsidence should occur. Once the TBM
and ground behavior has been established in the first part of the tunnel drive, ground sub-
sidence become a measure of the grouting programs effectiveness.

11.13.4 Coring programs

Project specifications often require cores drilling through the segments and grout behind the
TBM to check to grout filling of the annulus and the quality of the grout. There are several
practical limitations to this approach.
Coring is a direct method to check for grout filling at the crown but not a good method
to check grout compressive strength. Most grouts are not designed for high enough com-
pressive strengths, roughly 6.9 MPa (1,000 psi) that a competent core can be extracted for
compressive testing. Because the grout is injected behind the precast liner, and may displace
groundwater, and bentonite from shield injections, the grout may have layers that break off
in the core barrel. If the cores are taken directly behind the TBM at the front of the trailing
gear, the grout will be too weak to extract as single core.
 Annular backfill grouting of precast segmental tunnel linings 697

On smaller TBMs without an upper trailing gear deck there is no room for drilling cores
at the crown position. For larger TBMs, the upper deck is filled with equipment although
there may be small areas that can accommodate core drilling operations. At the tail end of
the trailing gear begins the ventilation line which occupies the crown and prevents any fur-
ther grout investigation until the ventilation line is taken down at the end of the TBM drive.
If the designer wants to verify grout fill by coring, he should specify that a section of TBM
trailing gear be dedicated to that task so that equipment and labor can be included in the
contractor’s bid.
A typical coring program will include more coring and testing at the beginning of the
tunnel; on the order of two proof holes every 5th ring for the first 50 to 100 rings. If the
grouting program is shown to be effective, the proof holes can be eliminated. If the proof
holes show a problem, the proof hole testing can be extended.
Coring is required at the crown because that is the most likely spot for a void but some-
times core holes are required at 10 or 2 o’clock. Observations are made during coring
for penetration rate, water flow and depth to the native ground. Coring is done past the
grout into the ground to see if the grout completely filled the annulus. The core hole is
examined for voids. Core holes are usually 50 mm (2 inches) in diameter. Since the hole
is cored through the segment and is typically 330 to 430 mm (13 to 19 inches) deep, one
needs a fiber optic inspection tool or a hooked probe to determine if there are any voids.
If voids are discovered, proof grout can be injected to fill the void. Use a metered grout
pump to determine the size of the void. Injection pressure is usually groundwater pressure
plus one bar.
If high pressure groundwater is present, and if the ground can be mobilized by water
movement, drilling through a segment to check grout fill is not a good idea. If the grout is
not placed promptly behind the tail shield, the ground will have moved in to fill the void
before one can check for grout and this will be obvious from ground settlement measure-
ments. When a TBM is tunneling through sandy, gravely, or silty ground beneath an inex-
haustible source of water such as a river or estuary, coring through the lining has more risks
than benefits and other methods should be used such as impact-​echo testing.
Placing annular grout in a rock tunnel can be quite different than in soft ground if over-
break is expected and the excavated face is at atmospheric pressure. A good example would
be a large diameter TBM excavating at atmospheric pressure through horizontally stressed
shale. The ground can be expected to break out at the shoulders and rock may be resting on
top of the TBM and the segmental liner. The annular grout is not going to fill large crown
voids. Instead placing additional grout will only push the grout towards the TBM. In this
situation, it makes sense to drill holes on a systematic basis, say one hole every ten rings to
check for voids. When a void is discovered, a grout line tied to the annular grout system
can be used to fill the void with grout. If using a two-​component grout, accelerator may
be reduced to allow the grout to travel farther. In this manner, the contact grout can be
addressed as the TBM advances.

11.14 SAFETY TRAINING AND PRECAUTIONS

Preparing, transporting and injecting annular grout has many hazards. Some of them are
common to all phases of underground construction, including tripping hazards, falls, crush-
ing injuries, and being struck by equipment. Hygienic issues such as excessive noise, dust
exposure and grout burns also exist. A safe construction site has a multipronged program
698 Handbook of Precast Segmental Tunnel Lining Systems

to educate the workers and reduce hazards. This section will look at these safety aspects of
the grouting operation:

• Personnel training
• Grout mix design
• Site layout
• Equipment selection
• Equipment installation and commissioning
• Start-​up
• Operations and maintenance

A constant theme that runs through the above list is that pumping and injecting grout
involves high pressures with the possibility for sudden uncontrolled releases of energy. The
stored energy exists over long distances where people working on the same grout line are
not in sight or in verbal communication with each other. Stored energy in a pipe or hose will
not be evident but is present. Even when the grout is behind the segmental liner, there is a
chance of injury while the pressurized grout is still fluid.

11.14.1 Personnel training

Personnel training starts with an analysis of the work and a written plan by a qualified,
experienced person for each step of the grouting process from receiving the materials, to
batching the grout, to transporting it to the heading, to injecting it through the segments.
Having the right person write the plan ensures that each step has been considered and the
hazards are recognized and eliminated or mitigated. The workplan should be in a standard
format with discrete steps listing the hazards and mitigations for each step. Training the
workers will be primarily on a verbal basis with the written plan as reference material.
Every worker involved should have a copy of the work plan signed and acknowledged by
them. Training must be in a language understood by the worker.
For example, the grout plant operator will be trained on what sequence of buttons should
be pushed to mix the grout and what grout mix to use. They must also know communi-
cation protocols for sending grout down to the tunnel, checking that the silos have oper-
able pressure relief valves, that the baghouses are in good condition, how to keep track of
stockpiles, how to test the grout, and what is the lockout-​tagout procedure for doing main-
tenance work on the plant. There is a lot to remember, and it can seem overwhelming to a
grout plant operator just starting out. The plant operator cannot be expected to perform
well without the training and the written procedures as reference. Operations and safety are
combined. Following the correct procedures leads to a safe operation.
The same can be said for the laborers down in the tunnel extending or cleaning out the
grout line. Everyone needs to be on the same page and that means following set procedures.
When the tunnel grout line is extended, the line needs to be shut-​off, disconnected, drained
and another pipe added before the line is reconnected and pressured-​up. If the laborer for-
gets to shut the line and release pressure before unbolting the coupling, an accident is sure
to happen. The severity of the accident is just a matter of luck. Cleaning the line with a pig
presents special hazards because the grout plant is pushing the pig towards a crew waiting
at the open end of the line. There can be a lot of pressure behind the pig and it can exit with
great force. The laborers must be trained in the safe method to catch the pig at the end of
the line.
 Annular backfill grouting of precast segmental tunnel linings 699

New workers should never be put in the tunnel untrained. They should be trained by a
supervisor in the procedures and hazards to the same standard as the original crew members.

11.14.2 Grout mix design

A grout mix that is hard to pump and prone to plugging the grout line is a safety hazard.
This is more of a problem with one component grouts where the ingredients are chosen
to stiffen rapidly once the grout is in place, but two-​component mixes develop plugs too.
A plugged grout line means the pump will deadhead and produce pressure spikes in the
line as the plug develops, stressing the equipment. Once the line is plugged, the plug must
be found, the pipeline broken apart and the plug cleared. All these procedures carry risks
that are largely eliminated by having an easy to pump grout. Grout plugs are never 100%
eliminated but a good grout mix reduces the risk.

11.14.3 Site layout

The grout plant must be integrated into the overall site layout. Generally, a tunnel main
portal or shaft site has restrictions because of the shape of the site, limited area, access
limitations and proximity to existing businesses, residences and environmentally sensitive
areas. All parts of the tunneling operation compete for space and access and the layout must
consider all tunnel support operations including grout. Most tunnels projects operate on a
multi-​shift basis, so external lighting should illuminate all work areas and walkways.
A clear pedestrian pathway access to the grout plant must be established and all hoses and
pipes should be installed in underground or in trenches (Figure 11.46) to eliminate tripping
hazards. Hose bibs must be provided to help keep the plant clean.
The site layout must look at the location of the grout plant and the flow of traffic to feed
the grout plant. Can the site be laid out to separate vehicle traffic from pedestrians? Can

Figure 11.46 Utility trenches eliminate trip hazards.

Note: Advance site layout planning required.
700 Handbook of Precast Segmental Tunnel Lining Systems

the location of the cement delivery truck (which has to park and unload for 1.5 hours)
be such that the delivery truck driver can make his connections to the silos without being
struck by passing equipment? Are the deliveries of chemicals by tanker or totes compatible
with the parade of trucks hauling muck away from the site? Are the turning radii of large
trucks accounted for? Is the batch plant clear of overhead swinging loads to the shaft? Are
all existing and temporary utilities considered? There are many different solutions to these
problems.
As we saw in the Riachuelo Project (Section 11.9), the interaction of the grout plant
operations with the rail crane servicing the main shaft was largely eliminated by batching
the grout away from the shaft and bringing the grout to shaft with a concrete mixer truck.
This was a good solution for the Riachuelo jobsite but a different jobsite, say in New York
City, would require a different solution.

11.14.4 Equipment selection

All grouting equipment specified and purchased for the job must have the necessary safety
features. That includes guarding moving parts, lock-​out/​tag-​out stations for maintenance
and safe access to maintenance areas. Equipment pressurizing grout lines must be designed
to handle the highest anticipated pressure spikes and must have a means to check for and
bleed off pressure. Used equipment should be examined to see if it will meet current safety
standards or if the item can be economically upgraded. Used equipment should be refur-
bished as necessary before being put to work.
For example, the cement silos, which are about 15 m (50 feet) tall should be purchased
with an access ladder and a guardrail around the top around that meets regulatory require-
ments. The pressure release valve should meet regulatory standards. The baghouse should
have the capacity to accept airflow blown in by delivery truck.
The grout plant must have electrical equipment which meets the regulatory standards of
the jobsite. Access to safely reach and clean the mixer should be included. Electrical con-
trols must be designed so that they can be locked out using a padlock or key switch system.
Interlock switches on access doors or panels on man accessible equipment may be required.

11.14.5 Equipment installation and commissioning

All equipment arriving at the site should be inspected. Complicated pieces of new equip-
ment should be commissioned by a factory representative. For used equipment, a factory
representative or qualified person who is intimately familiar with the equipment can do the
commissioning. Company personnel from other jobsites who have experience with the same
equipment and grout processes may assist.
The grout plant should be assembled under the supervision of an experienced person and
wired up by a qualified electrician. All safety items should be checked for operation before
first use including lockout-​tagout switches, safety interlocks, automatic shutdown for exces-
sive high pressures, and equipment guards. The grout plant should be cycled through all
the normal functions to ensure the plant operates correctly. Emergency stops must be tested
and operational.
The grouting equipment on the TBM trailing gear is usually included a part of the TBM
and consists of tanks and pumps to move the grout and admixtures. This equipment will
be tested as part of the TBM commissioning process. All safety items should be checked or
tested.
 Annular backfill grouting of precast segmental tunnel linings 701

The commissioning process is well defined when a multi-​million dollar piece of equip-
ment like the TBM is assembled on site. Manufacturers have check lists which should be
followed for safety and to ensure the warranty is enforceable. Often, a gap exists between
various equipment systems which the site management must fill with experienced personnel.
For example, the surface grout plant and the TBM probably come from separate manufac-
turers. The site needs to identify the interface, which in this case would be the grout line
between the batch plant and the TBM, and make sure that the two work together smoothly.

11.14.6 Start-​up
Start-​up can be a dangerous process. While the individual pieces of equipment have been
tested in the commissioning process, some parts of the process cannot be tested until grout
is run through the lines and injected behind the segments. The equipment and personnel are
new to the site, the process of batching and delivering the grout is untested, and the pressure
exists for the jobsite to produce. There is an entire complex chain of processes that must
take place for the grout to be safely delivered and injected and the grouting process is only
one of many starting up. There most certainly will be pieces of equipment that do not func-
tion as intended and grout lines that get plugged. As the entire tunneling enterprise cannot
advance until the grout system is operating, the pressure to take shortcuts can be intense.
The risks of start-​up are reduced when the previous steps of personnel training, devel-
oping a good grout mix, selecting the right equipment, and testing equipment are taken
seriously. Daily briefings with the workers to explain what is about to happen and what
hazards exist must be done. Communication from every critical point along the grout pro-
cess must exist whether that is by radio, mine telephone or at pre-​shift meetings. Every
worker must be empowered to stop the process if something amiss is observed.
Personnel responsible for commissioning equipment should also be on hand for startup.
Site leadership needs to exhibit patience and set the example for the crews by following the
written procedures. The first time a grout line is clogged, site supervision needs to lead the
way in finding the plug, releasing the pressure and cleaning out the plug. A determination
must be made on why the plug happened in the first place. Start-​up will include establishing
grout batching and pumping capacities. Fine tuning will be necessary. Procedures changed
in the field need to be immediately followed up with revised work plans that include any
new hazards created by the changes.

11.14.7 Operations and maintenance

Once the TBM is underway, a well-​planned grouting operation will fall into a routine.
Procedures are well established, and the hazards are known, but the potential for injury still
exists. Grout injection pressures are commonly in the 3 to 5 bar (43.5 to 72.5 psi) range
but can go as high as 22 bar (320 psi). That pressure exists as an energy that is stored in
the line between the injection pump and the discharge point. If the tunnel has a long grout
supply line, or comes down a deep shaft, pressures will also be high. A failure in equipment
or procedure resulting in a sudden release of energy can be fatal.
Any grout operation requires cleaning of equipment. Grout builds up inside the storage
tanks, hoses, fittings and pumps. Many of these pumps and mixers are remotely started.
This equipment requires lockout tagout procedures to ensure that equipment is not started
while someone is working on or in it. Grout tank cleaning may require manned entry in
which case confined space procedures shall be followed.
702 Handbook of Precast Segmental Tunnel Lining Systems

Hazards often associated with the grouting operation are as follows:

• Grout burns
• Eye injuries
• Foot injuries

The simplest measures of wearing standard personal protective equipment (hardhat, gloves,
safety glasses, and steel tied boots) will go a long way towards abating some of the hazards.
Other hazards described in more detail below are as follows

• High pressures
• Slips, trips and falls
• Cleaning out the equipment
• Changes in routine and procedure

11.14.7.1 High pressures

Hose bursts
Flexible hoses in the 50 to 75 mm (2 to 3 inches) size are common on TBMs where the grout
line needs to make a difficult transition, or the flexibility is required to allow for movement.
These hoses must be inspected periodically for wear and damage to prevent a burst hose.
One-​component mortar grouts are more likely to cause internal wear, but the hoses can also
be externally damaged by repeated flexing, kinking, dropped objects or abuse. Damaged
hoses must always be replaced with hoses and end fittings having the same pressure rating.
Keep spare hoses and fittings on hand to resist the temptation to continue using a worn hose
and/​or unapproved fittings.

Pipe couplings
Rigid pipe and flexible hoses (Figures 11.47 and 11.48) are assembled with quick connec­­
tion clamps to facilitate cleaning. These clamps have safety pins and clips which should
always be reinserted after taking a coupling apart. Keep plenty of spare pins and clips on
hand. Some couplings use two bolts to hold the coupling together. Bolted couplings must
use the correct bolts. The coupling and hose end must be cleaned off before making up the
connection to ensure that no foreign objects are trapped in the coupling. The couplings
must use the correct gasket. Never use damaged couplings. A good connection has a certain
look to it. Make sure that all connections are good.

Shifting key segment

The grout injected behind the assembled precast liner ring has fluid pressure until the grout
sets. The grout applies a uniform pressure around the liner ring. The resultant forces can
cause a ring to spit out the key, break the gasket seal and cause a sudden expulsion of grout.
Fatalities have occurred when grouting injection pressures were high enough, that the key
shifted and the grout was ejected. Ring designers should consider this possibility when
designing the key piece of the ring and the bolting arrangement. TBM thrust rams should
always press against the assembled liner ring. The ring erector operator and their assistant
should always observe the key piece for unexpected movements when assembling the next
ring and releasing the thrust rams in the key area. In the event there is key movement, a
restraining device may have to be installed.
 Annular backfill grouting of precast segmental tunnel linings 703

Figure 11.47 Tunnel utility lines.

Note: Grout pipeline on top connects to TBM grout hose.

Figure 11.48 Hose reels on TBM trailing gear.

Note: Hoses extend as TBM advances and retract when a new section of pipe is installed.
704 Handbook of Precast Segmental Tunnel Lining Systems

Tunnel grout line plugs and clean-​out

TBM may be supplied by grout in rail tank cars or by a supply line mounted on the tunnel
wall. The tunnel supply line will be as long as the tunnel. Although tunnel supply lines con-
duct very fluid grout, they still can develop plugs and need to be cleaned out on a regular
basis. Finding a plug along a grout line many thousands of meters long is a combination of
art and science. Before breaking the line, make sure that the pressure is relieved at the sup-
ply pump. If the line drops down a shaft, there will still be a head pressure on the grout line
that needs to be drained off at the shaft bottom. Tunnels with long downhill grades will also
develop a head pressure in the pipe that needs to be bled off. Once a plug develops, pushing
on the plug with more pressure usually make the plug worse.
A cleaning pig will be run through the supply line at regular intervals to clean off the pipe
walls. Use water to push the pig and NOT compressed air. Water is an incompressible sub-
stance and once the pig comes out the other end of the line, the pressure will drop rapidly.
Compressed air stores a lot of energy behind the pig. As the pig gets closer and closer to the
end of the line and the resistance drops, the pig will move faster and faster and will shoot
out like a bullet at the pipe end followed by a blast of air. Always put a pig catcher on the
end of the open line to prevent the pig from becoming a projectile.

Cleaning-​out lines on the TBM

The lines on the TBM and trailing gear between the grout pump and the TBM tail shield
will occasionally need to be taken apart and flushed clean. Make sure the grout pumps are
properly locked out and that the pressure from the injection point (tail shield or segment)
is cut-​off. Cleaning pigs (also called “rabbits”) may be run in the line between the pump
on the trailing gear and the tail shield. The same safety rules apply as for cleaning-​out the
tunnel supply line.

11.14.7.2 Slips, trips and falls

Hoses
Hoses are a tripping hazard. Water hoses used for washdown should have a designated stor-
age rack so they can be taken off the walkway when not in use. Have multiple locations with
hose bibs and short hoses to eliminate long hoses over walkways.

Grout
Grout spills, especially of unaccelerated two-​component grout, can be slippery, in the
curved invert of the smoothly finished precast tunneling. Even clean wet inverts are slip-
pery. Steel deck plates can also become slippery. When hoses and pipe are disassembled
for cleaning, the resultant wash water and grout ends up in the curved invert. Any grout
spills should be cleaned. Best practice is to clean up the grout spill immediately into a skip
box and not through the discharge pipe because the grout will eventually plug the dis-
charge line.

11.14.7.3 Cleaning equipment

Maintenance is part of the grouting operation and involves cleaning out equipment, pipes
and hoses. The lowest level laborer will often be assigned to this task without adequate
training.
 Annular backfill grouting of precast segmental tunnel linings 705

Figure 11.49 Lockout/​t agout station on TBM.

Note: Eyewash bottle station at right.

Lock-​out/​tag-​out
Never allow workers to clean powered equipment without making sure the equipment
cannot be accidentally turned on. This applies to electrical equipment but can also apply
to pressurized water and hydraulic lines. Have a lock-​out/​tag-​out station on the TBM
(Figure 11.49).

Confined space
A worker may occasionally be required to enter a space not set up for human occupation
defined as a confined space. Working in these areas requires special training of the entire
work crew. Make sure that proper ventilation is provided, a watch is kept on the worker,
method of extracting on unconscious worker exists are in place, and proper confined space
paperwork is filled out and kept for record purposes.

Pressure washing grout lines in the tail shield

A particular danger exists when cleaning out grout lines embedded in the tail shield. These
grout lines are not readily accessible and pressurized groundwater may exist just outside the
shield. Common methods to clean-​out the lines are to use a flexible steel cable or a high-​
pressure hose with a jetting head at the end.
Careless use of high-​pressure hose cleaning jets has caused the injection of water through
the skin. This is a serious injury that requires immediate medical care. The injuries usually
occur when there is an unexpected exit of the water jet from the grout pipe and nozzle end
whips out. The hose should be marked with tape 600 and 300 mm (2 and 1 feet) from the
end so that the operator knows when the hose is about to come out.
Injuries have also been caused by not having a shut-​off valve within reach of the person
operating the cleaning hose. A trigger style hand operated valve from a pressure washer
wand provides a good way for the clean-​out person to control the water. Protective gloves
706 Handbook of Precast Segmental Tunnel Lining Systems

rated against pressure washer penetration add another layer of protection. Have eyewash
stations on the TBM where grout is handled, or equipment may be disassembled.

11.14.7.4 Changes in routine or procedure

A crew learns to work together and will develop a set routine with assigned tasks. When
a new crew member is added that person will not know exactly how the rest of the crew
accomplishes the task. The person may not be new to the jobsite but just new to the crew.
That new person could be someone from another shift on a crew that does the same task.
Unfortunately, different crews will often perform the same task slightly differently. The shift
supervisor needs to make sure that the new crew member understands the task at hand and
what his role is in the task. Along with making sure the crew member understands the task,
the shift supervisor shall be at the location where the task is taking place and not sitting
in his office. Accidents can also be traced to a change in procedure. Anytime something is
different, the opportunity for an accident increases. All crew members should be trained
to be aware of such changes and empowered to speak up and stop the work if something
is wrong.
Changes in routine compounded by changes in procedure dramatically raise the risk of
an accident. Consider the following scenario: a grout crew member does not show up for
the night shift and the crew was already shorthanded. A laborer stays over from swing
shift to help. Towards the end of the shift, the tail shield injection line is plugged and the
foreman decides to clean out the line with the high-​pressure nozzle. The foreman cannot
find the trigger valve assembly and decides to plumb the pressure washer hose direct to the
pressure washer pump. Now, the only way to turn the pressure washer off is by using the
switch at the pump. The foreman puts the swing shift laborer on the pump as the cut-​off
man. The pump is not in clear view of the tail shield. The pressure washer is turned-​on and
the laborer on the TBM starts cleaning the plugged line. He is not wearing the protective
gloves because they are wet from the previous day’s use. The laborer loses track of how
far the hose is inside the pipe and the hose end comes out unexpectedly and whips around.
The laborer starts screaming in Spanish to shut off the pump but the laborer on the pump,
who has already been up for 16 hours, does not speak Spanish. The water jet slashes the
hand of the laborer holding the hose. The foreman rushes over and shuts off the pump. The
injured laborer requires reconstructive surgery to his hand. How many violations of good
safety practice discussed in this chapter just occurred? How many violations of safety pro-
cedure were required before the accident could occur? The answer is left as an exercise for
the reader.

BIBLIOGRAPHY
For preparation of this chapter, the following references were used, which are categorized into two
categories of authority published documents and authored documents.

AUTHORITY PUBLISHED DOCUMENTS

AMERICAN SOCIETY FOR TESTING MATERIALS (ASTM)

C39/​C39M, Standard Test Method for Compressive Strength of Cylindrical Concrete Specimens.
C109/​C109M, Standard Test Method for Compressive Strength of Hydraulic Cement Mortars (Using
2-​inch or [50mm] Cube Specimens).
 Annular backfill grouting of precast segmental tunnel linings 707

C143/​C143M, Standard Test Method for Slump of Hydraulic-​Cement Concrete.

C393, Standard Test Method for Core Shear Properties of Sandwich Constructions by Beam Flexure.
C403, Standard Test Method for Time of Setting of Concrete Mixtures by Penetration Resistance.
C494, Standard Specification for Chemical Admixtures for Concrete.
C939, Standard Test Method for Flow of Grout for Preplaced-​Aggregate Concrete (Flow Cone Method).
C940, Standard Test Method for Expansion and Bleeding of Freshly Mixed Grouts for Preplaced-​
Aggregate Concrete in the Laboratory.
C1017, Standard Specification for Chemical Admixtures for Use in Producing Flowing Concrete.
C6910/​6910M, Standard Test Method for Marsh Funnel Viscosity of Construction Slurries

EUROPEAN FEDERATION OF NATIONAL ASSOCIATIONS REPRESENTING FOR CONCRETE

EFNARC:2005, Specification and Guidelines for the Use of Specialist Products for Mechanized
Tunneling (TBM) in Soft Ground and Hard Rock.
European Standards/​European Norms (EN)
EN 480-​4:1997, Determination of Concrete Bleeding.
EN 413: 1995, Masonry Cement–​Part 2: Test Methods.

INTERNATIONAL TUNNELLING AND UNDERGROUND SPACE ASSOCIATION (ITA/​A ITES)

ITA Report No.4, Guidelines on Best Practices for Segment Backfilling, May 2014.
National Ready Mix Concrete Association (NRMCA)
Plant Certification Electronic Checklist Version 03.20.20.
Plant Inspector’s Guide.

AUTHORED DOCUMENTS
Ivantchev, A.; Del Rio, J.; 2015, “Two-​Component Backfill Grouting for Double Shield TBMs”,
ITA-​AITES World Tunnel Congress, Dubrobnik, Croatia, May 22–​28, 2015.
Linger, L.; Cayrol, M.; Boutillon, L.; 2008, “TBM’s Backfill Mortars –​Overview –​Introduction to
Rheological Index”, Tailor Made Concrete Structures 2008, J.C. Walraven and D. Stoelhorst,
editors, Taylor & Francis, London, pp. 271–​276.
Maidl, B.; Herrenknecht, M.; Maidl, U.; Wehrmeyer, G.; 2012, Mechanised Shield Tunneling, 2nd
Edition, February 2012, Ernst & Sohn, Berlin, Germany, 490 p.
Mok, P.; Norbert, M.; 2014, “Tunnel Boring Machine Excavation Stability–​Double Shield Tunnel
Boring Machine Advance with Partially Grouted Annulus”, 15th Australasian Tunneling
Conference 2014, Sydney, NSW, Australia, September 17–​19, 2014.
Novin, A.; Tarighazali, S.; Mohammad, F.; Fasihi, E; Mirmehrabi, S; 2015, “Comparison Between
Simultaneous Backfilling Methods with Two Components and Single Component Grouts
in EPB Shield Tunneling”, ITA-​AITES World Tunnel Congress, Dubrobnik, Croatia, May
22–​28, 2015.
Pellegrini, L.; Perruzza, P.; “Sao Paolo Metro Project–​ Control of Settlements in Variable Soil
Conditions Through EPB Pressure and Bicomponent Backfill Grout”, Proceeding of the Rapid
Excavation Tunnel Conference 2009, G. Almararis and B. Mariucci, editors, Society for Mining,
Metallurgy, and Exploration, Las Vegas, NV, June 14–​17, 2009.
Pelizza, S.; Peila, D.; Borio, L.; Dal Negro, E.; Schulkins, R.; Boscaro, A.; 2010, “Analysis of the
Performance of Two Component Backfilling Grout in Tunnel Boring Machines Operating
Under Face Pressure”, ITA-​AITES World Tunnel Congress 2010, Vancouver, Canada, May
14–​20, 2010.
Salini Impregilo /​Webuild S.p.A. Argentine Office, 2021, Riachuelo 3 the Tunnel and the Underground
Worlds in Culture, Webuild S.p.A., Buenos Aires, Argentina.
708 Handbook of Precast Segmental Tunnel Lining Systems

Shirlaw, J.; Richards, D.; Ramond, P.; Longchamp, P.; 2004. “Recent Experience in Automatic Tail
Void Grouting with Soft Ground Tunnel Boring Machines”, Proceedings of the 30th ITA-​
AITES World Tunnel Congress, Singapore, May 22–​27, 2004.
Thewes, M.; Budach, C.; 2009 “Grouting of the Annular Gap in Shield Tunneling–​an Important
Factor for Minimization of Settlements and Production Performance”, Proceedings of the 35th
ITA World Tunnel Congress, Budapest, Hungary.
Vitale, M.; Carlson, J.; Garrod, B.; Gabriel, D.; 2013, “Innovation in Annular Grouting at the Euclid
Creek Tunnel, Cleveland, Ohio”, Proceedings of the Rapid Excavation Tunnel Conference
2013, M.A. DiPonio and C.J. Dixon, eds., SME Press, Engelwood, CO.
Youn, B.; Schulte-​Schrepping, C.; Breitenbucher, R.; 2016, “Properties and Requirements of Two
Component Grouts in Mechanized Tunneling”, ITA World Tunneling Congress 2016, San
Francisco, April 22–​28, 2016.
Chapter 12

Durability and service life of precast

segmental linings
Mehdi Bakhshi and Verya Nasri

12.1 INTRODUCTION
Tunnels are typically designed for a service life of more than 100 years. In segmentally
lined tunnels, the durability of a tunnel is directly related to the durability of the concrete
segments, acting as both the initial support and the final lining. In this chapter, the most-​
frequent degradation mechanisms of concrete linings are briefly discussed. This includes
reinforcement corrosion by chloride attack and carbonation, and sulfate and acid attacks as
major deterioration processes caused by external agents. This discussion will also include
frost attack, freeze-​and-​thaw, and alkali-​aggregate reactions caused by internal chemical
reactions. Stray current-​induced corrosion is also introduced as another major durability
concern specific to the railway and the subway tunnels. Mitigation methods for stray cur-
rent corrosion, including the use of fiber reinforced concrete (FRC) segments are presented,
and durability of segments under the coupling effects of stray current and other conven-
tional degradation mechanisms are explained.
The conventional approach for durability design, based on European codes EN 1992-​1-​
1:2004 and EN 206-​1:2013, American code ACI 318-​19, and Canadian code CSA A23.1-​19
are summarized. Using these standards, recommendations made on concrete strength class,
maximum water-​to-​cement ratio (w/​c), minimum cementitious materials content, minimum
air content and other requirements to ensure tunnel durability are explained. Although the
durability recommendations of other guidelines such as BS PAS 8810 (2016), DAUB:2013,
AFTES (2005), ÖVBB (2011) and LTA (2010) are not covered in this chapter, their method-
ologies are very similar in nature to EN, ACI and CSA methods. Performance-​based service
life design for concrete structures is explained and an example of using this methodology to
design service life of a segmentally lined tunnel is given.

12.2 CONVENTIONAL DEGRADATION MECHANISMS

The durability of bored tunnels needs to be addressed from the perspective of degradation
mechanisms specific to tunnel segmental linings. Due to different geological environments
surrounding tunnels and the specific use of each tunnel, tunnel linings are exposed to differ-
ent aggressive environments. Possible degradation and damage mechanisms in bored tun-
nels include the corrosion of reinforcement by chloride attack and carbonation, sulfate and
acid attacks, alkali-​aggregate reactions and freeze-​and-​thaw damages.

DOI: 10.1201/9781032675541-12
710 Handbook of Precast Segmental Tunnel Lining Systems

12.2.1 Reinforcement corrosion

Unprotected steel materials tend to corrode quickly. However, steel bars embedded in con-
crete as reinforcement are protected by a passive layer formed on the surface of the steel
due to the high alkalinity of hydrated cement in concrete (pH of 12 to 13). Without a pas-
sive layer, steel corrodes at rates at least 1,000 times higher (ACI 222R). The destruction of
the passive layer and subsequent rapid reinforcement corrosion occurs when the alkalinity
of the concrete is reduced by carbonation or when the chloride concentration in concrete
around the reinforcement is increased to corrosion threshold levels. In addition, oxygen
and moisture should also be available to sustain this reaction. As a result, pitting corrosion
reduces the size of the reinforcement and uniform corrosion leads to cracking and spalling
of the concrete.

12.2.2 Corrosion of reinforcement induced by chloride attack

Chloride-​induced reinforcement corrosion is the main cause of degradation in tunnels lined
with reinforced concrete. As shown in Table 12.1, ITA Working Group 6–​Maintenance and
Repair of Underground Structures (1991) presents 13 tunnels that have been significantly
damaged due to reinforcement corrosion by chloride ingress before 1991 (Abbas, 2014).
Chloride-​induced corrosion is a major durability concern in sub-​sea, sea outfall, and
road or rail tunnels. These tunnels are exposed to brackish groundwater and the intrusion
of chloride ions present in saltwater can cause steel reinforcement to corrode. In cold-​
region road or rail tunnels, the major durability concern is the ingress of chloride ions
present in deicing salts used during snow falls. Chloride-​induced corrosion due to saltwater

Table 12.1 Damaged or corroded tunnels due to chloride ingress.

ITA Working Group on maintenance and repair of underground structures 1991

Diameter
Tunnel Completion
Item Tunnel name Location type Metric Imp’l year
1 Basel/​O lten Hauenstein Switzerland Railway 1916
2 Northern Line Old Street to United Metro 3.5m 11.5’ 1924
Moorgate Kingdom
3 Shimonoseki/​M oji Kanmon Japan Railway 1944
4 Mikuni National Route 17 Japan Highway 7.6m 25.0’ 1959
5 Uebonmachi-​N ipponbashi Japan Railway 10.0m 32.8’ 1970
6 Dubai Road Tunnel UAE Road 3.6m 11.8’ 1975
7 Tokyo Underground Japan Road 1976
8 Berlin Tunnel Airport Germany Road 1978
9 Second Dartford Tunnel United Road 9.6m 31.5’ 1980
Kingdom
10 Mass Transit Railway Hong Kong Metro 5.6m 18.4’ 1980
11 Ahmed Hamdi Egypt Road 10.4m 34.1’ 1980
12 Stockholm Underground Sweden Metro 1988
Source: Abbas, 2014.
 Durability and service life of precast segmental linings 711

infiltration initiates from the lining extrados, whereas corrosion due to deicing salts sprayed
from vehicle tires starts from lining intrados.
When the chloride concentration in concrete exceeds the threshold level between 0.15
and 0.6% by mass of cement (ACI 318-​19), iron atoms lose electrons in anode and electrons
flow in the rebar to cathode and react with water and oxygen in the concrete to
form hydroxide (OH-​). The ferrous ions (Fe+​2) then combine with hydroxide to form iron
hydroxides, Fe(OH)2, also known as rust. For more information about electrochemical cor-
rosion reactions in concrete reinforcement, refer to ACI 533.5R-​20 and Wang et al., 2018.
Rust, as the reaction product, has a greater volume than steel, which then exerts pressure on
concrete solid skeleton. When concrete tensile stress exceeds its tensile strength, this results
in cracking, delamination, and spalling in the concrete (Figures 12.1 and 12.2).

Figure 12.1 Loss of reinforcement—​c ross-​s ection view.

Note: Loss of reinforcement section and cracks caused by chloride-​induced steel corrosion.
Source: Adopted from PCA IS536. 

Figure 12.2 Loss of reinforcement—​p an view.

Note: Loss of reinforcement section and cracks caused by chloride-​induced steel corrosion.
712 Handbook of Precast Segmental Tunnel Lining Systems

12.2.3 Corrosion of reinforcement induced by carbonation

Carbonation-​induced corrosion is considered a minor durability issue in reinforced concrete
structures compared to chloride-​induced corrosion. This is mainly due to the limited impact
area of carbonation and reduced strength, which is limited to the extreme outer layer. In
bored tunnels, carbonation is less likely to occur because the tunnel lining extrados is often
permanently wet, while the intrados is constantly dry. It is well-​known that high rates of
carbonation occur when the relative humidity is between 50 and 75% (PCA IS536). In low
relative humidity, the degree of carbonation is insignificant. When the humidity is above this
range, the moisture in concrete pores restricts penetration of CO2 (ACI 201.2R). In tunnels,
only portal areas and entrance zones can maintain a relative humidity in the aforementioned
range because concrete lining in such areas is exposed to cyclic wet and dry conditions.
Developing a high rate of carbonation requires elevated atmospheric carbon dioxide
(CO2) levels, which is only the case in heavily trafficked road tunnels because of CO2 emis-
sions from cars. Therefore, carbonation is a major durability factor in portal areas and
entrance zones of heavily trafficked road runnels. Carbonation can also occur in concrete
linings exposed to bicarbonate (HCO3-​) groundwater, which is often formed by the reac-
tion of carbon dioxide with water and carbonate bedrocks (e.g. limestone and dolomite).
When carbon dioxide or bicarbonate penetrates the concrete, it reacts with hydroxides
from the cement paste such as calcium hydroxide to form carbonates; e.g. calcium car-
bonate (PCA IS536). Calcium carbonate is deposited in concrete pores in the presence of
free water and as shown in Figure 12.3, it reduces the pH of the pore solution from 12 to
13 to as low as 8 to 9 in carbonated region (BTS ICE 2004). When the concrete alkalinity
close to reinforcement reduces to this level, the passive layer on the steel surface becomes
unstable and destroyed. With depassivation and in the presence of water and oxygen, the
rate of steel corrosion increases by three orders of magnitude or higher (ACI 222R-​19).

12.2.4 Reinforcement corrosion mitigation

Both chloride-​induced and carbonation-​induced corrosion can be mitigated by using concrete
with low water-​to-​cement ratio, high compressive strength and high cement content. This, in

Figure 12.3 Carbon dioxide from air or bicarbonate.

Note: CO 2 from the air or bicarbonate (HCO 3-​) ground water reacts with cement hydration
products such as calcium hydroxide, Ca(OH) 2, and reduces the pH of concrete from around 13
to less than 9.
Source: Adopted from ACI 533.5R-​2 0. 
 Durability and service life of precast segmental linings 713

conjunction with sufficient concrete cover over reinforcement, provide high-​quality and dense
concrete that can delay the initiation time of corrosion, also known as propagation time,
beyond the service life of the structure. Further details regarding code recommendations to
reduce chloride attack are provided in Section 12.3. Two other effective mitigation methods,
which are not outlined in the codes are using cements with high amounts of tricalcium alumi-
nate (C3A or 3CaO Al2O3) and the addition of corrosion inhibitors to concrete mixture.

12.3 SULFATE ATTACK

Sulfate attack is a major durability concern for concrete structures in contact with soil or
water that contains deleterious amounts of water-​soluble sulfate ions. Tunnels as under-
ground structures, regardless of their specific use, can be exposed to external sulfate attack
from common sources such as sulfates of sodium, potassium, calcium or magnesium found
in the surrounding ground or dissolved in natural ground water. Ancient sedimentary clays
and the weathered zone of other geological strata (top 30 feet (10 m), and contaminated
ground and groundwater generally contain significant sulfate concentrations (BTS ICE
2004). When tunnel linings are exposed to such conditions, sulfate attack can be a major
concrete deterioration mechanism.
As sulfate ions penetrate in the cement paste, they react with unconstrained alumi-
nate phases, mainly aluminate mono-​sulfate (AFm or Al2O3-​Fe2O3-​mono-​phase), calcium
hydroxide, or both. Main reaction products are ettringite and gypsum, which have higher
volume than the reactants. For example, ettringite volume is approximately 2.2 times higher
than the reactants. As a result, as shown schematically in Figures 12.4 and 12.5, concrete
cracking and spalling are likely to occur.
In the case of magnesium sulfate attacks, along with ettringite and gypsum, brucite or
magnesium hydroxide may also form on the concrete surface. Brucite drops the pH of a
pore solution due to the consumption of calcium hydroxide. This results in decomposition
of the calcium silicate hydrates or C-​S-​H, which is the main product resulting from cement
hydration and is responsible for the concrete strength.

Figure 12.4 External sulfate attack damage process—​e ttringtite.

Note: Formation of ettringite by reaction with sulfate and moisture.
Source: Adopted from Helsing and Mueller, 2013. 
714 Handbook of Precast Segmental Tunnel Lining Systems

Figure 12.5 External sulfate attack damage process.

Note: Expansion and recrystallization causing cracks and voids which are filled with ettringite.
Source: Adopted from Helsing and Mueller, 2013. 

There is also another type of sulfate attack, known as internal sulfate attack, which is
caused by the sulfate present in cement, which is commonly referred to as delayed ettringite
formation (DEF). DEF, however, is observed in concrete only when the initial concrete curing
temperature is high [greater than 160°F (70°C)], or in the case of mass concrete due to
excessive heat of hydration. Project specifications usually prohibit such high curing tempera-
tures for precast concrete tunnel segments. Therefore, it is expected that damages in tunnel
linings due to sulfate attack start on segment extrados and at the interface between the lining
and the ground where sulfate from ground or groundwater can penetrate the concrete.
Sulfate attack can be mitigated in a number of ways:

• By using cements with a low amount of C3A (less than 8%),

• Use of high content of active mineral components, low water/​cement ratio, and
• The use of blended cements with pozzolans.

Code and standard recommendations (ACI 318-​19; EN 206-​1:2013; EN 1992-​1-​1:2004;

CSA A23.1-​19) for mitigation of sulfate attack are based on using a concrete with low
water/​cement ratio, high compressive strength, and high cement content. In addition, codes
require use of sulfate-​resistant cements such as Type II Portland cement (ASTM C150/​
C150M) or in severe cases type V (ASTM C150/​C150M) and pozzolans or slag cement.

12.4 ACID ATTACK

Acid attack can be a major durability issue when a concrete structure is exposed to high
concentrations of aggressive acids, which have the highest degrees of dissociation. The
deterioration of concrete by acids is primarily the result of decomposition of the hydration
products of the cementitious paste (ACI 201.2R). Sulfuric, hydrochloric and nitric acids
are the main inorganic or mineral acids. And acetic, formic and lactic acid are the main
organic acids with a rapid rate of attack on concrete at ambient temperatures. Acids reduce
the pH or alkalinity of the concrete, and once the pH reduces to less than 5.5 to 4.5, severe
damage is imminent. Cement hydration products such as Portlandite [CH or Ca(OH)2] and
 Durability and service life of precast segmental linings 715

calcium silicate hydrates (C-​S-​H) start to decompose when pH drops to approximately 10

(ACI 201.2R). This is the main reason that concrete materials do not have good resistance
to acids.
In tunnels, the concrete lining can be attacked from external sources in the surrounding
ground and groundwater, as well as from internal sources within the tunnels. Concerning
external sources, acidic materials may be found on polluted sites used for industrial waste,
agricultural applications, animal feed and manure, or from natural sources such as peat
soils, clay soils and alum shales. These geological strata, for example, contain sulfide bear-
ing minerals such as pyrite that produce sulfuric acid on oxidation (ACI 201.2R). The rapid
deterioration of concrete, however, only occurs when concrete is subject to the action of
highly mobile acidic water (BTS ICE, 2004).
Regarding internal sources for acid attack, flow of acid-​containing run-​off from outside
the tunnel is not a major concern. However, sulfuric acid solutions resulting from decay of
organic matter by bacterial action in sewage and wastewater tunnels can be a durability
concern. This is due to the high attack rate of sulfuric acid and continuous movement of the
acidic materials inside the tunnel as gravitational flow of sewage in these tunnels is always
guaranteed. Note that sewage is not aggressive to concrete by itself, but hydrogen sulfide
produced by anaerobic bacteria’s reaction within the sludge is subsequently oxidized by
aerobic bacteria to form sulfuric acid. In addition to the decomposition of cement hydra-
tion products, sulfuric acid is particularly aggressive to concrete because the calcium sul-
fate formed from the acid reaction may drive sulfate attack of adjacent concrete that was
unaffected by the initial acid attack (ACI 201.2R; PCA IS536).
Acid attacks can be mitigated by providing a dense and high-​quality concrete, lowering
the water/​cement ratio and increasing compressive strength and cement content. Codes and
standards (ACI 318-​19; EN 206-​1:2013; EN 1992-​1-​1:2004; CSA A23.1-​19) provide spe-
cific limits to achieve very high density and relatively impermeable concrete to reduce the
damage due to acid attack. The type of cement used has an insignificant role in the mitigation
of acid attack. When concrete is exposed to acid attack, surface protection methods such
as coatings, waterproofing membranes, or use of a sacrificial layer should be considered.

12.5 ALKALI-​AGGREGATE REACTION

Alkali-​aggregate reaction (AAR) is a chemical reaction between reactive aggregates and
cement. In most concrete, aggregates are chemically inert or nonreactive. However, AAR as
a chemical attack is a major durability concern when aggregates contain materials that can
be reactive with alkali hydroxides in cement phase. The AAR generates expansive products
and may result in damaging deformation and cracking of concrete over a period of years.
AAR has two main forms: alkali-​silica reaction (ASR) and alkali-​carbonate reaction (ACR).
ASR is often more of a concern compared to ACR as aggregates containing reactive silica
are more common (PCA IS536), whereas aggregates susceptible to ACR are less common
and usually unsuitable for use in concrete. Reactive forms of silica can be found in aggre-
gates such as chert, volcanic glass, quartzite, opal, chalcedony and strained quartz crystals.
Damage to concrete typically occurs when cement alkali content is high, aggregate contains
an alkali-​reactive constituent, and concrete is under wet conditions (BTS ICE 2004). ASR
reactions can be summarized as

Alkalis +​Reactive Silica → Gel Reaction Product

Gel Reaction Product +​Moisture → Expansion
716 Handbook of Precast Segmental Tunnel Lining Systems

Because sufficient moisture is needed to promote the destructive expansion, PCA IS536
reports the internal relative humidity of 80% as a threshold, below which the alkali-​silica
reactivity can be virtually stopped.
Concrete tunnel linings are not different from general types of concrete elements as far as
AAR sources, which are internal reactive aggregates. Therefore, the degradation mechanism
of AAR does not depend on the specific use of each tunnel. Sub-​sea tunnels may be more
susceptible due to exposure to warm seawater-​containing dissolved alkalis that may aggra-
vate ASR. AAR can be mitigated using inert aggregate, controlling the amount of soluble
alkalis in concrete, and using blended cements with pozzolans.

12.6 FROST ATTACK AND FREEZE-​A ND-​T HAW DAMAGE

Frost attack and freeze-​and-​thaw damage are durability concerns in concrete structures
built in cold regions. Water expands by approximately 9% when it freezes and, as a result,
the moisture in concrete capillary pores exerts pressure on the concrete solid skeleton. This
leads to development of excessive tensile stresses in the concrete and rupture of cavities.
Successive cycles of freezing and thawing can disrupt the paste and aggregates and eventu-
ally cause significant expansion and cracking, scaling, and crumbling of the concrete (PCA
IS536). Frost damage is considerably accelerated by deicing salts (ACI 201.2R).
Frost damage at early ages is not noted in precast concrete tunnel segments produced
under high-​quality control conditions of precast plants. In well-​cured concrete segments
with durable aggregate, however, surface scaling may occur in the tunnel which is the
loss of paste and mortar from the surface of the concrete. When water turns to ice, its
volume increases by approximately 9%. Consequently, more than 90% of the capillary
pore volume needs to be filled with water for the ice formation to induce internal stresses
(BTS ICE, 2004). Moisture content near saturation level is usually the case for tunnel linings
as tunnels are often built under the water table and concrete lining can be near saturation
level. However, along most of the tunnel alignment, the temperature rarely falls under the
freezing point because the tunnel is embedded in the ground. However, tunnel entrances,
portals and shafts are parts of the tunnel and underground system that should be considered
for exposure to cycles of freezing and thawing because of the saturation level and potential
exposure to freezing temperatures.
Freezing and thawing attacks are mitigated by controlling water/​cement ratio, com-
pressive strength, and cement content. However, the most effective method to mitigate the
freeze–​thaw attack is controlling the air content in the mixture to a minimum value of 4 to
5% using air-​entraining admixtures. Codes and standards (ACI 318-​19; EN 206-​1:2013;
EN 1992-​1-​1:2004; CSA A23.1-​19) often provide limits for the maximum water/​cement
ratio and the minimum compressive strength. Certain codes, in addition, require use of
frost-​resistant aggregate (EN 206-​1:2013).

12.7 STRAY CURRENT CORROSION

Stray current corrosion is a type of corrosion that is specific to rail tunnels. It is caused by
traction current and results in the rapid migration of the chloride ions and an accelerated
oxidation of metals around the tracks. See “Report on the Damaging Effects of Water on
Tunnels During Their Working Life” (1991) by ITA Working Group No.6.
Electric trains consume at least 20% less power than the diesel-​powered trains (Kemp,
2007). They also have a lower carbon footprint during operation and provide sustainable
 Durability and service life of precast segmental linings 717

solutions to public transportation systems. Therefore, government agencies around the

world are promoting electric trains, and modern railway systems are taking advantage of
railway electrfication. In electric train systems, power transmission is provided either by an
overhead catenary wire or via a conductor rail, which is also known as a “third rail.” Due
to construction limitations and maintenance costs, the running rail connected to nearby sub-
stations is often used as a traction loop through which the return circuit is made. Therefore,
running rails in modern railway systems are used not only for the purpose of mechanical
support and guideways, but also to conduct the electricity in the traction and signaling cir-
cuits (Brenna et al., 2010). Running rails have limited conductivity, and insulation between
the rail and the ground is sometimes reduced or constructed poorly from the beginning. This
causes a fraction of the traction current to leave the rail, leak into the ground and flow back
along the running rail on the return path to the traction substation by the earth diversion,
which is referred to as “stray current.”
When trains run in a lined tunnel, stray current leaks to the tunnel lining and through the
concrete reinforcement. This is shown schematically in Figure 12.6 with a cathode formed
at reinforcement where stray current enters the reinforcing bar and an anode is formed
where stray current leaves the reinforcing bar and flows back to the substation. Corrosion
and severe damage to concrete are anticipated due to hydrogenation and the accumulation
of corrosion products. As shown in Figure 12.7, in a cathode, the reinforcing bar is disen­
gaged from the concrete due to trapped hydrogen isostatic pressure, and in the anode, the
reinforcing bar is oxidized in contact with an electrolytic material, that is, the concrete. As
a result, the accumulation of corrosion products (rust), which have greater volume than the
steel, exert excessive pressure on concrete solid skeleton leading to cracking.
It is noteworthy to mention that this type of corrosion is not limited to the reinforcement
in the concrete lining, as severe corrosion in metal utilities and steel pipelines embedded in
the ground have also been observed in the proximity of railroad tracks.

Figure 12.6 Train catenary stray currents.

Note: Schematics of stray current from a train catenary system picked up by steel reinforcement
in concrete.
Source: Adopted from Bertolini et al., 2007. 
718 Handbook of Precast Segmental Tunnel Lining Systems

Figure 12.7 Corrosive effect of stray currents.

Note: Corrosive effect of stray current on reinforced concrete.
Source: Adopted from Wang et al., 2018. 

12.7.1 Mitigation methods for stray current corrosion

Major mitigation methods for stray current corrosion include, for example,

• Decreasing rail resistance;

• Improving rail-​to-​ground insulation using isolated rail fastening systems or pads;
• Keeping the substation as close as possible to the point of maximum current;
• Developing monitoring systems, devices, and measurement apparatus (Brenna et al.,
2010), and
• Use of fiber-​reinforcement in lieu of reinforcing steel bars.

All of these measures, with the exception of the latter, are effective in reducing the amount
of stray current and, therefore, are considered as general measures for stray current corro-
sion mitigation of all metals embedded in the ground or reinforcement in both cast-​in-​place
and precast segmental linings.
Brenna et al. (2010) used finite element method (FEM) simulations to study specific cases
of reinforcement corrosion in precast segments due to stray current. In their simulations,
stray current was due to the traction power of surface tramway lines in close proximity to the
tunnel, and not from the catenary system of the subway train inside the tunnel (Figure 12.8).
While this particular simulation pertains to a subway line in Milan, located under an oper-
ating surface tramway line, results of this study can be extended to general segmentally lined
tunnels. Figure 12.9 shows the conduction field in the ground and equipotential surfaces
around the segmental ring under a tramway track voltage of 8 volts, corresponding to rush
hour conditions. As shown in Figure 12.10, the area near the joints between two adjacent
segments has been studied as a critical aspect of the model. The current leaves the upper
segment (Segment #6 on Figures 12.9 and 12.10), flows into the ground, and then returns
to the adjacent segment (Segment #5 on Figures 12.9 and 12.10). Particular parts of the
reinforcement where current leaves the reinforcing bar, constitute the anode where corro-
sion can initiate, depending on the reinforcement-​to-​ground potential difference.
Simulation results in this research demonstrate that the potential difference between the
reinforcement and ground, for example for Segment #6 on Figures 12.10 and 12.11, is
greater than the maximum value allowed by EN 50122-​2:2010, indicating the possibility
 Durability and service life of precast segmental linings 719

Figure 12.8 Typical tunnel section.

Note: Typical section of the subway tunnel with the above tramway lines.
Source: Adopted from Brenna et al., 2010. 

Figure 12.9 Current field and equipotential surfaces.

Note: Current field and equipotential surfaces around a segmental ring as results of simulation.
Source: Adopted from Brenna et al., 2010. 
720 Handbook of Precast Segmental Tunnel Lining Systems

Figure 12.10 Current field and equipotential surfaces at tunnel segments.

Note: Equipotential surfaces and current field near tunnel segment lining joints.
Source: Adopted from Brenna et al., 2010. 

Figure 12.11 Stray current mitigation—​f ield layout.

Note: Stray current mitigation using equipotential connection solution provided by the copper
plates/​s traps connecting reinforcement cages of adjacent segments.
Source: Adopted from Dolara et al., 2012. 
 Durability and service life of precast segmental linings 721

Figure 12.12 Stray current mitigation—​s chematic section.

Note: Stray current mitigation using equipotential connection solution provided by the copper
plates/​s traps connecting reinforcement cages of adjacent segments.
Source: Adopted from Dolara et al., 2012. 

of corrosion initiation. Therefore, a potential mitigation method for stray current corrosion
in precast reinforced concrete segments is providing an equipotential connection between
reinforcing bars of adjacent segments and of adjacent rings to prevent zones of high output
current density.
As shown in Figures 12.12 and 12.13, Dolara et al. (2012) modeled this solution by using
copper plates connecting reinforcement cages of adjacent segments together, for the same
tramway line studied by Brenna et al. (2010). In this figure, the traces of reinforcing bars
in adjacent segments were presented with red dashed lines connected together with a plate
connection. Using this solution, the equi-​potential connections between reinforcing bars of
segments in a ring constitute a path with extremely low electrical resistance that allows the
current to flow from one segment to another without passing through the ground (Dolara
et al., 2012).
Electrically connected bars behave like a cylindrical metallic shield that allows the stray
current to distribute in a nearly uniform way, both when entering from the top (crown)
and when leaving from the invert of the tunnel. In this case, the anode is the ring outer sur-
face near the invert and the maximum potential difference between the reinforcement and
ground is reduced by more than 15 times. A conclusion can be drawn that an equipotential
connection between the reinforcement of adjacent segments reduces the reinforcement-​to-​
ground potential difference (voltage) to values well below the standard limits for corrosion
initiation (EN 50122-​2:2010) and provides an effective method to prevent stray current
corrosion.
Steel fiber-​reinforced concrete (SFRC) has been introduced as an effective mitigation
method for stray current-​induced corrosion (Tang, 2017; Solgaard et al., 2013). Results of
studies on stray current corrosion indicate that reinforcing bars are more likely to pick up
the current than short steel fibers under the same conditions (Edvardsen et al., 2017). This
can be attributed to the chloride threshold for the corrosion of reinforcing bars in concrete,
which is between 0.15 and 0.6% by mass of cement (ACI 318-​19). On the other hand,
steel fiber-​reinforced concrete demonstrates a much higher corrosion resistance compared
to rebar with a chloride threshold level of 4% by mass of cement, which is one order of
magnitude higher than the reinforcing bars (Tang, 2017).
722 Handbook of Precast Segmental Tunnel Lining Systems

Figure 12.13 Connection plate located between reinforcement cages.

Note: Effect of using a connection plate between reinforcement cages of adjacent segments on
equipotential surfaces and current field near segment joints.
Source: Adopted from Dolara et al., 2012. 

From a theoretical point of view, corrosion current can be picked up by the reinforcement
if the potential difference between the anodic and cathodic areas at least equals the sum
of the anodic and cathodic polarizations. This polarization of the reinforcement is con-
trolled by the level of the stray currentfield and the distance between the anode and cathode.
Stray current may flow through the rebar or steel fibers embedded in concrete only when a
threshold value of driving voltage in the concrete is reached between the opposite ends of
the reinforcement (Bertolini et al., 1993). Solgaard et al. (2013) performed a comparative
experimental study between a commercial steel fiber [hooked-​end, 1.4 inches (35 mm) long
and 0.022 inch (0.55 mm) in diameter] and a short carbon steel bar 14 inches (350mm) long
and 0.4 inches (10mm) diameter. They evaluated the stray current flow versus the potential
gradient needed for the current pickup.
Views of specimens and schematics of experimental test set-​up for stray current measure-
ments are shown in Figures 12.14 to 12.16. Results of their studies, as shown in Figures 12.17
and 12.18, indicate that the minimum potential gradient required for current circulation in
the steel fibers is about 3.05 to 3.96 V/​ ft (10 to 13 V/​ m) compared to
rebar specimens, with an observed potential gradient of 0.21 to 0.03 V/​ft (0.7 to 0.1 V/​m).
Confirming the theory, this indicates a linear relationship between the required potential
gradient and length of the reinforcement solution for stray current circulation. The rebar
used in these experiments is only 10 times longer than the fibers, whereas rebar used in
practice can be several orders of magnitude longer. Thus, in practice steel fibers are much
less susceptible to transferring current than rebar for concrete structures subjected to the
same level of stray current. Additionally, results of the study conducted by Solgaard et al.
 Durability and service life of precast segmental linings 723

Figure 12.14 Experimental test set-​u p for stray current measurements—​s teel fibers.
Notes: (a) Top view of specimens with fibers; (b) side view of specimens with fibers.
Source: Adopted from Solgaard et al., 2013. 

Figure 12.15 Experimental test set-​u p for stray current measurements—​s teel bars.
Notes: (a) Top view of specimens with short rebar; (b) side view of specimens with short rebar
(Solgaard et al., 2013).
Source: Adopted from Solgaard et al., 2013. 

(2013) indicate that after stray current was picked up by the embedded steel, the stray cur-
rent potential gradients in steel fiber specimens (Figure 12.17) are much lower than those
for rebar specimens (Figure 12.18). This further highlights the significance of steel “length
effect” on the stray current corrosion.
For example, in order for stray current flow of 0.5mA to be circulated through the
embedded steel, the potential gradient required for steel fiber is more than 6.1 V/​ft (20 V/​
m) compared to 0.37 V/​ft (1.2 V/​m) for steel bar. Therefore, the potential gradient required
for the circulation of the current through a single fiber was of the order of several volts per
minute, which is not realistic in actual structures. This is because, in the case of rebar, the
724 Handbook of Precast Segmental Tunnel Lining Systems

Figure 12.16 Experimental test set-​u p for stray current measurements—​g eneral arrangement.
Note: Experimental set-​u p for stray current measurements.
Source: Adopted from Solgaard et al., 2013. 

Figure 12.17 Stray current potential gradients in tunnel segments—​s teel fiber.

Note: Current transferred by embedded steel as a function of the potential gradient between the
left and the right side of (after long-​t erm exposure): (a) steel fiber.
Source: Adopted from Solgaard et al., 2013. 
 Durability and service life of precast segmental linings 725

Figure 12.18 Stray current potential gradients in tunnel segments—​s hort steel bars.
Note: Current transferred by embedded steel as a function of the potential gradient between the
left and the right side of (after long-​t erm exposure): short steel bar.
Source: Adopted from Solgaard et al., 2013. 

Figure 12.19 Schematics of stray current paths.

Note: Schematics of stray current paths through SFRC with a “red path” depicting a very-​h igh
potential gradient path because current should enter and leave each single fiber, and a “green
path” depicting the path of least resistance through concrete.

polarization contributions of the anodic and cathodic sites are less intensified and diluted
along the length of the bar.
A conclusion can be drawn that in real structures where the rebar may be tens of meters
long compared to 14 inches (0.35 m) in this study, the potential gradient required for the
circulation of stray current may be negligible. In contrast, in the case of steel fibers, which
are not electrically interconnected, as shown in Figure 12.19, the stray current should con­
tinuously enter and exit every single fiber, each time dissipating the driving voltage for
anodic and cathodic polarizations. This is presented schematically by a “red path” in
Figure 12.19 compared to the least resistance path presented by a “green path.” Therefore,
“length effect” makes a clear distinction between discrete steel fibers and continuous steel
bars, making the risk of stray current circulation extremely low in the case of concrete rein-
forced only by steel fibers.
726 Handbook of Precast Segmental Tunnel Lining Systems

The higher chloride threshold found for steel fibers was also explained by an effect of cold
drawing, which may increase the polarization resistance and decrease the surface roughness
and micro-​structural defects (Solgaard et al., 2013). Therefore, steel fibers are less suscep-
tible to stray current-​induced corrosion than ordinary rebar; not only because of their short
length but also because of higher intrinsic resistance to chloride-​induced corrosion.
As far as the risk of circulation of stray current through steel fiber is concerned, the posi-
tive effect of short fiber length may be reduced when fibers come into direct electrical con-
tact. The conventional amount of steel fiber added to the concrete is usually around 0.5%
volumetric or 67 lbs/​yd3 (40 kg/​m3) by mass. Although some fibers are likely to touch each
other and thereby form a metallic path through the concrete, it is very unlikely that the steel
fibers generate a continuous metallic path, similar to reinforcing bars. Also, the accidental
presence of a few fibers electrically interconnected by surface contact, would not reduce the
order of magnitude of the threshold values of the potential gradients shown in Figure 12.17;
consequently, extremely low stray current would be expected even in this case.
In summary, the short, discontinuous and discrete nature of steel fibers or the length-​
effect is the main reason for the high corrosion resistance of SFRC, as fibers rarely touch
each other and there is no continuous conductive path for stray currents through the con-
crete (ACI 544.1R). This in addition to higher intrinsic resistance to chloride-​induced cor-
rosion because of the cold-​drawing fabrication process suggests that there would be no risk
in relation to stray current corrosion of steel fibers in concrete. Even if steel fibers corrode,
it will not lead to structural damage in the form of cracking and/​or spalling [Graeff et al.,
2009)].

12.8 DURABILITY UNDER COUPLING MULTI-​D EGRADATION FACTORS

Precast concrete tunnel segments may be subjected to the coupling effects of multiple degrad-
ation factors such as carbonation and sulfate and chloride-​induced corrosion of steel bars
by groundwater and surrounding ground (Bakhshi and Nasri, 2018a). For electrical subway
tunnels, stray current is another major factor that accelerates steel corrosion. A summary of
these major degradation mechanisms and their mitigation methods are shown in Table 12.2.
Li et al. (2014) studied the durability of concrete tunnel segments on a large scale, sub-
jected to various degradation factors of carbonation, sulfate and chloride penetration, cou-
pled with stray-​current corrosion. As shown in Figures 12.20 to 12.25, precast segments
were subjected to a current flow of 1A to simulate the stray current, and the extrados of
segments was immersed in a solution of 3.5% NaCl and 5% Na2SO4 to simulate chloride
and sulfate penetration from the surrounding ground and groundwater. The intrados of
tested segments was exposed to a relative humidity of 70 ±5%, a temperature of 68 ±9oF
(20 ±5oC), and a carbonation set-​up with CO2 concentration of 20% representing the inside
environment of the tunnel. Further details of the testing by Li et al. (2014) are presented in
ACI 533.5R-​20. Results of this study reveal that the stray current accelerates the migration
of chloride ions and also changes its penetration distribution in the section. In the absence
of stray current, the largest concentration of chloride ions is near the surface exposed to
the corrosion solution while in the presence of stray current, the largest concentration of
chloride ions is at the reinforcement level.
Results of the study by Li et al. (2014) also indicate that the chloride ion concentration is
higher for segments immersed only in chloride solutions than those immersed in solutions
with both chloride and sulfate. This can be due to filling of the concrete pores with ettrin-
gite produced by reaction of sulfate ions and cement hydration products, resulting in the
 newgenrtpdf
Table 12.2 Degradation mechanisms for tunnel linings.
Sources, and mitigation methods

Degradation Type of tunnels susceptible to Location of tunnel

Item mechanism factor Main sources of degradation prone to factor Mitigation method
1 Chloride-​ Subsea tunnels Sea/​s altwater Lining extrados Delay corrosion initiation by:
induced • Sufficient cover over rebar
corrosion Sea outfall tunnels Sea/​s altwater • Dense/​h igh quality concrete:
Transportation tunnels in cold Chloride ions present in • Low w/​c ratio
regions deicing salts used during • High compressive strength
snow falls • High cement content
• Cement with high C3A content
2 Carbonate-​ Heavily travelled roadway CO 2 emission from car Lining intrados • Use of corrosion inhibitors
induced tunnels exhaust near portals,
corrosion entrance zones,

Durability and service life of precast segmental linings

shafts
All types of tunnels embedded Bicarbonate (HCO 3) Lining extrados
in carbonate bedrock such as groundwater formed by
limestone or dolomite the reaction of water and
carbonate bedrocks
3 External All types of tunnels embedded Formation of ettringite Lining extrados • Dense/​h igh quality concrete:
sulfate in ancient sedimentary clays due to sulfate reacting • Low w/​c ratio
attack with calcium aluminates or • High compressive strength
All types of shallow tunnels Ca(OH) 2 • High cement content
exposed to weathered zone • Cement with low C3A content
(less than 32 feet [10m]) of (less than 8%)
other geological strata • Pozzolans/​b lended cement
All types of tunnels exposed to
sulfate contamination
4 Internal Sewage and wastewater Formation of H 2S and Lining intrados • Dense/​h igh quality concrete
acid attack oxidation to sulfuric acid • Coatings
• Sacrificial layers
• Use calcareous aggregates

(Continued)

727
 newgenrtpdf
Table 12.2 (Continued)

728
Degradation Type of tunnels susceptible to Location of tunnel

Handbook of Precast Segmental Tunnel Lining Systems

Item mechanism factor Main sources of degradation prone to factor Mitigation method
5 Alkali All types of tunnels built with Volcanic glass No specific • Use inert aggregate
aggregate reactive silica aggregate Opal/​c halcedony location • Control amount of soluble alkalis
reaction Deformed quartz in concrete
(AAR) • Pozzolans/​b lended cement
Subsea tunnels Warm seawater containing Lining extrados
dissolved alkalis
6 Frost All types of tunnels in cold Surface scaling due to Lining intrados • Dense/​h igh quality concrete:
attack and regions increase in volume when near portals, • Low w/​c ratio
freeze–​ water turn to ice near entrance zones, • High compressive strength
thawing saturation shafts • High cement content
• Air-​e ntraining admixture
7 Stray Subway tunnels Stray current leaking into Near rebar • Reduce amount of current:
current lining when returning from • Decrease rail resistance
corrosion running rail • Improve rail/​g round insulation
• Substation close to maximum
electrical current
• Use of straps connecting bars
• Use fiber reinforcement
Source: Bakhshi and Nasri, 2018a.
 Durability and service life of precast segmental linings 729

Figure 12.20 Study on stray current with multiple concrete degradation mechanisms—​g eneral.
Note: Size and layout of experimental segment samples and reinforcement.
Source: Li et al., 2014. 

Figure 12.21 Study on stray current with multiple concrete degradation mechanisms—​s ection.
Note: Schematics of test set-​u p for combined effect of chloride and sulfate penetration and
stray-​c urrent.
Source: Li et al., 2014. 

Figure 12.22 Study on stray current with multiple concrete degradation mechanisms—​t ests.
Note: Views of carbonation test set-​u p.
Source: Li et al., 2014. 
730 Handbook of Precast Segmental Tunnel Lining Systems

Figure 12.23 Study on stray current with multiple concrete degradation mechanisms—​d evices.
Note: Views of carbonation test set-​u p.
Source: Li et al., 2014. 

Figure 12.24 Study on stray current with multiple concrete degradation mechanisms—​rebar cage.
Note: Corrosion in main reinforcement and stirrups.
Source: Li et al., 2014. 

Figure 12.25 Study on stray current with multiple concrete degradation mechanisms—​s tirrup.
Note: Corrosion in main reinforcement and stirrups.
Source: Li et al., 2014. 
 Durability and service life of precast segmental linings 731

obstruction of the concrete pores through which chloride ions migrate. In addition, results
demonstrate that the carbonation depth in segment intrados is only 0.04 to 016 inches (1
to 4 mm), concluding that carbonation is not a controlling durability factor for concrete
segments when compared to the chloride and sulfate ion penetration and stray current
corrosion. Considering the results of this study, a conclusion may be drawn that coupling
factors of chloride ion penetration and stray current has the most detrimental effect on the
durability of precast concrete tunnel segments.

12.9 DESIGN FOR SERVICE LIFE

Durability design based on prescriptive approaches is often performed in accordance with
major national and international structural codes (ACI 318-​19; EN 206-​1:2013; EN 1992-​
1-​1:2004; CSA A23.1-​19). Such codes specify characteristics of concrete, such as concrete
strength or maximum water/​cement ratio, based on the exposure condition or environ-
mental classes that concrete element is exposed to.

12.10 CODE APPROACHES

Durability design according to major code approaches of European Norm (EN 1992-​1-​
1:2004 and EN 206-​1:2013), American Concrete Institute (ACI 318-​19), and Canadian
Standards Association (CSA A23.1-​19) are presented in the following.

12.10.1 European norm (EN) method

EN 1992-​1-​1:2004, used in conjunction with EN 206-​1:2013, is one of the world’s most
well-​known design codes, providing comprehensive, prescriptive-​based specifications for
concrete exposed to environmental actions. In EN 1992-​ 1-​
1:2004, the main exposure
classes are the following:

• XC for carbonation-​induced corrosion

• XD for chloride-​induced corrosion from other sources than sea water
• XS for chloride-​induced corrosion from sea water
• XF for freezing and thawing
• XA for chemical attacks

Depending on the severity of exposure class, ranges of XC1 to XC4, XD1 to XD3, XS1
to XS3, XF1 to XF4, and XA1 to XA3 are provided for different exposure conditions. For
clarity, as shown in Table 12.3, EN 1992-​1-​1:2004 describes the environmental condition
for each exposure class and provides informative examples where each exposure classmay
occur. Nonetheless, EN 1992-​1-​1:2004 does not directly provide informative examples for
underground construction. EN 206-​1:2013 provides specific limiting values for exposure
classes in the case of chemical attack from natural soil and groundwater, which is presented
here on Table 12.4.
For the specific case of tunnel lining, suggested exposure classes for carbon dioxide-​
induced carbonation are XC3 to XC4, for seawater chloride-​induced corrosion are XS2
to XS3, for deicing salt chloride-​induced corrosion are XD2 to XD3, for freezing and
thawing are XF3 to XF4, and for harmful ions other than chloride (such as Mg+​2, SO42-​
) are XA1 to XA3 (Helsing and Mueller, 2013). The concrete requirements specified by
732 Handbook of Precast Segmental Tunnel Lining Systems

Table 12.3 Concrete exposure classes.

Concrete exposure classes per EN 1992-​1 -​1 :2004 and EN 206-​1 :2013

Exposure Description of the Informative examples where

Item class environment exposure classes may occur
1 No risk of corrosion or attack
For concrete without reinforcement or embedded metal
1.1 X0 All exposures except where • Concrete inside buildings with very
there is freeze/​t haw, abrasion or low air humidity.
chemical attack.
For concrete with reinforcement
or embedded metal: very dry
2 Corrosion induced by carbonation
Where concrete containing reinforcement or other embedded metal is exposed to air
and moisture, the exposure shall be classified as follows.
2.1 XC1 Dry or permanently wet • Concrete inside buildings with low
air humidity
• Concrete permanently submerged in
water
2.2 XC2 Wet, rarely dry • Concrete surfaces subject to long-​
term water contact
• Many foundations
2.3 XC3 Moderate humidity • Concrete inside buildings with
moderate or high air humidity.
• External concrete sheltered from
rain
2.4 XC4 Cyclic wet and dry • Concrete surfaces subject to water
contact, not within exposure
class XC2
3 Corrosion induced by chlorides other than from seawater
Where concrete containing reinforcement or other embedded metal is subject to
contact with water containing chlorides, including deicing salts, from sources other
than from seawater, the exposure shall be classified as follows.
3.1 XD1 Moderate humidity • Concrete surfaces exposed to
airborne chlorides
3.2 XD2 Wet, rarely dry • Swimming pools and concrete
exposed to industrial waters
containing chlorides
3.3 XD3 Cyclic wet and dry • Parts of bridges exposed to spray
containing chlorides.
• Pavements, car park slabs
4 Corrosion induced by chlorides from seawater
Where concrete containing reinforcement or other embedded metal is subject to
contact with chlorides from seawater or air carrying salt originating from sea water,
the exposure shall be classified as follows.
4.1 XS1 Exposed to airborne salt but not in • Structures near to or on the coast
direct contact with sea water
4.2 XS2 Permanently submerged • Parts of marine structures
4.3 XS3 Tidal, splash, and spray zones • Parts of marine structures
 Durability and service life of precast segmental linings 733

Table 12.3 (Continued)

Exposure Description of the Informative examples where

Item class environment exposure classes may occur
5 Freeze /​ thaw attack with or without de-​icing agents
Where concrete is exposed to significant attack by freeze/​t haw cycles whilst wet, the
exposure shall be classified as follows.
5.1 XF1 Moderate water saturation, • Vertical concrete surfaces exposed
without de-​i cing to rain and freezing
agent
5.2 XF2 Moderate water saturation, with • Vertical concrete surfaces of road
de-​i cing structures exposed to freezing
agent • Airborne deicing agents
5.3 XF3 High water saturation, without • Horizontal concrete surfaces
de-​i cing exposed to rain and freezing
agent
5.4 XF4 High water saturation, with de-​ • Road and bridge decks exposed to
icing agent or sea water deicing agents
• Concrete surfaces exposed to direct
spray containing deicing agents and
freezing
• Splash zones of marine structures
exposed to freezing
6 Chemical attack
Where concrete is exposed to chemical attack from natural soils and groundwater, the
exposure shall be classified as follows.
6.1 XA1 Slightly aggressive chemical • Concrete exposed to natural soil and
environment according to EN 206-​ ground water
1, Table 2
6.2 XA2 Moderately aggressive chemical • Concrete exposed to natural soil and
environment according to EN 206-​ ground water
1, Table 2
6.3 XA3 Highly aggressive chemical • Concrete exposed to natural soil and
environment according to EN 206-​ ground water
1, Table 2

EN 1992-​1-​1:2004 include the assumption of an intended design service life of 50 years.

However, almost all tunnels are designed for a service life of over 100 years. Table 12.5,
which presents Table 4.3N of EN 1992-​1-​1:2004, must be taken into account for longer
service life of tunnels (≥100 years), which provides requirements of increases in exposure
class for structures with intended service life of more than 50 years. In the case of tunnels,
exposure classes must be increase by “2” for most exposure conditions during the service
life of the structure.
The main specifications set by EN 206-​1:2013 (Table F.1) and EN 1992-​1-​1:2004 for
each exposure class are shown in Table 12.6, which includes maximum water/​cement ratio,
minimum strength class, and minimum cement content. This is because the quality of the
concrete outer layer (cover) as the main protection layer is achieved by limiting the max-
imum water/​cement ratio, minimum cement content, and minimum concrete strength (EN
1992-​1-​1:2004).
734 Handbook of Precast Segmental Tunnel Lining Systems

Table 12.4 Limiting values for exposure classes.

Limiting values for exposure classes for chemical attack from natural soil and
groundwater

Exposure class
Chemical Reference
Item characteristic test method XA1 XA2 XA3
A Groundwater
1 SO 42-​ mg/​l EN 196-​2 ≥200 and ≤600 >600 and >3,000 and
≤3,000 ≤6,000
2 pH ISO 4316 ≥5.5 and ≤6.5 <5.5 and ≥4.5 <4.5 and ≥40
3 CO 2 mg/​l; aggressive EN 13577 ≥15 and ≤40 >40 and ≤100 > 100 up to
saturation
4 NH 4+​ mg/​l ISO 7150-​1 ≥15 and ≤30 >30 and ≤60 > 60 and ≤100
5 Mg 2+​ mg/​l EN ISO ≥300 and ≤1,000 >1,000 and > 3,000 up to
7980 ≤3,000 saturation
B Soil
1 SO 42-​ mg/​l [1]‌
total EN 196-​2 [2]‌ ≥2,000 and >3,000 [3]‌ and >12,000 and
≤3,000 [3]‌ ≤12,000 ≤24,000
2 Acidity according to EN 16502 >200 Not encountered in practice
Baumann Gully ml/​k g
Source: EN 206-​1 :2013, Table 2.
Notes:
[1]
Clay soils with a permecability below 10 -​5 m/​s may be moved into a lower class.
[2]
The test method prescribes the extraction of SO 42-​ by hydrochloric acid; alternatively, water extraction
may he used, if experience is available in the place of use of the concrete.
[3]
The 3,000 mg/​k g limit shall be reduced to 2,000 mg/​k g where there is a risk of accumulation of sulfate
ions in the concrete due to drying and wetting cycles or capillary suction.

Additional specifications such as minimum air content, aggregate characteristics for

the freeze–​thaw attack, and sulfate-​resisting cement for chemical attack can be found in
Table 12.6. For corrosion protection of steel rebar for concrete exposure to carbonation
(XC classes) or chloride ion penetration (XD/​XS classes), values of minimum concrete cover
for structural class S4 in accordance with EN 1992-​1-​1:2004, Table 4.4N should be met
(Table 12.7). As an example, for CX4 and XD3/​XS3 exposure classes, which are usually
used in the design of tunnel linings with an intended service life of 100 years, the minimum
required covers are 1.20 inches (30 mm) and 1.80 inches (45 mm), respectively.

12.10.2 American Concrete Institute (ACI) method

ACI 318-​19 is another well-​known and widely used concrete code to design durability of
tunnel segments. ACI 318-​19 approach, similar to EN 1992-​1-​1:2004 and EN 206-​1:2013,
is based on the exposure categories such as category F, S, W and C defined in ACI 318-​19,
Table 19.3.1.1, and the requirements of ACI 318-​19, Table 19.3.2.1. These two tables are
presented here as Tables 12.8 and 12.9. However, there are a few differences between ACI
and European (EN) codes.
 Durability and service life of precast segmental linings 735

Table 12.5 Recommended structural classification.

Recommended structural classification per EN 1992-​1 -​1 :2004, Table 4.3N

Structural classification
Exposure class according to Table 4.1 on EN 1992-​1 -​1 :2004
(Table 12.3 here)
XD2/​ XD3/​
Item Criterion X0 XC1 XC2/​XC3 XC4 XD1 XS1 XS2/​X S3
1 Design Increase Increase Increase Increase Increase Increase Increase
working life of Class Class Class Class Class Class Class
100 years by 2 by 2 by 2 by 2 by 2 by 2 by 2
2 Strength ≥ C30/​ ≥ C30/​ ≥ C35/​ ≥ C40/​ ≥ C40/​ ≥ C40/​ ≥ C45/​
class according 37 [1]‌ 37 [1]‌ 45 [2]‌ 50 [3]‌ 50 [3]‌ 50 [3]‌ 55 [4]‌
to EN 1992-​1 -​1 : reduce reduce reduce reduce reduce reduce reduce
2004 Class Class Class Class Class Class Class
by 1 by 1 by 1 by 1 by 1 by 1 by 1
3 Member with reduce reduce reduce reduce reduce reduce reduce
slab geometry Class Class Class Class Class Class class
(position of by 1 by 1 by 1 by 1 by 1 by 1 by 1
reinforcement
not affected by
construction
process)
4 Special quality reduce reduce reduce reduce reduce reduce reduce
control of Class Class Class Class Class Class Class
the concrete by 1 by 1 by 1 by 1 by 1 by 1 by 1
production
ensured
Special Notes:
[1]
The strength class C30/​3 7 based on Eurocode is equivalent to concrete with f′ c =​ 4,350 psi (30 MPa)
‌ The strength
[2]
class C35/​4 5 based on Eurocode is equivalent to concrete with f′ c =​ 5,100 psi (35 MPa)
‌ The strength
[3]
class C40/​5 0 based on Eurocode is equivalent to concrete with f′ c =​ 5,800 psi (40 MPa)
‌ The strength
[4]
class C45/​5 5 based on Eurocode is equivalent to concrete with f′ c =​ 6,550 psi (45 MPa)
Table Notes:
1
The strength class and water/​c ement (w/​c) ratio are considered to be related values. A special composition
(type of cement, w/​c value, fine fillers) with the intent to produce low permeability may be considered.
2
The limit may be reduced by one strength class if air entrainment of more than 4% is applied.

• First, ACI 318-​19 has an additional exposure category W, for the durability of con-
crete in contact with water but not exposed to freezing and thawing, chlorides, or
sulfates.
• Second, exposure category C applies to concrete exposed to all conditions that require
protection against corrosion of reinforcement regardless of the corrosion source. In
general, exposure category C in ACI 318-​19 can be compared to XC and XD/​XS
exposure classes in EN 1992-​1-​1:2004, which provides more insight into how speci-
fications are set differently with respect to corrosion sources.
• Third, ACI 318-​19 considers only sulfate attack requirements category S exposure,
whereas EN 206-​1:2013 and EN 1992-​1-​1:2004 cover a wide range of chemical
attacks induced by different types of ions and acids, not limited to sulfate.
 newgenrtpdf
736
Table 12.6 Recommended characteristics of concrete.
Recommended characteristics of concrete by EN 206-​1 :2013

Handbook of Precast Segmental Tunnel Lining Systems

Exposure classes
Chloride-​i nduced corrosion
No risk of
corrosion or Carbonation-​i nduced Chloride other than Freeze and Aggressive chemical
attack corrosion Sea water water thaw attack environments
Item X0 XC1 XC2 XC3 XC4 XS1 XS2 XS3 XD1 XD2 XD3 XF1 XF2 XF3 XF4 XA1 XA2 XA3
1 Maximum water/​c ement ratio
-​ 0.65 0.60 0.55 0.50 0.50 0.45 0.45 0.55 0.55 0.45 0.55 0.55 0.5 0.45 0.55 0.50 0.45
2 Minimum strength class according to EN 1992-​1 -​1 :2004
C12/​1 5 C20/​ C25/​ C30/​ C30/​ C30/​ C35/​ C35/​ C30/​ C30/​ C35/​ C30/​ C25/​ C30/​ C30/​ C30/​ C30/​ C35/​
[1]‌
25 30 37 37 37 45 45 37 37 45 37 30 37 37 37 37 45
[2]‌ [3]‌ [4]‌ [4]‌ [4]‌ [5]‌ [5]‌ [4]‌ [4]‌ [5]‌ [4]‌ [3]‌ [4]‌ [4]‌ [4]‌ [4]‌ [5]‌

3 Minimum cement content, lb/​yd 3 (kg/​m 3)

-​ 440 470 470 505 505 540 575 505 505 540 505 505 540 575 505 540 575
(260) (280) (280) (300) (300) (320) (340) (300) (300) (320) (300) (300) (320) (340) (300) (320) (340)
4 Minimum air content (%)
-​ -​ -​ -​ -​ -​ -​ -​ -​ -​ -​ -​ 4.0 4.0 4.0 -​ -​ -​
5 Other requirements
-​ -​ -​ -​ -​ -​ -​ -​ -​ -​ -​ Aggregate in accordance Sulfate
with EN 12620 with resisting
sufficient freeze and thaw cement
resistance
Special Table Notes:
‌ The
[1]
strength class C12/​1 5 based on Eurocode is equivalent to concrete with f′ c =​ 1,750 psi (12 MPa)
‌ The
[2]
strength class C20/​2 5 based on Eurocode is equivalent to concrete with f′ c =​ 2,900 psi (20 MPa)
[3]‌
The strength class C25/​3 0 based on Eurocode is equivalent to concrete with f′ c =​ 3,600 psi (25 MPa)
[4]
‌ The strength class C30/​3 7 based on Eurocode is equivalent to concrete with f′ c =​ 4,350 psi (30 MPa)
‌ The
[5]
strength class C35/​4 5 based on Eurocode is equivalent to concrete with f′ c =​ 5,100 psi (35 MPa)
 Durability and service life of precast segmental linings 737

Table 12.7 Values of minimum reinforcement cover.

Values of minimum cover over steel reinforcing bars per EN 1992-​1 -​1 :2004

Environmental requirement for c min,dur, in inches (mm)

Exposure class according to Table 4.1 on EN 1992-​1 -​1 :2004

Structural
Item class X0 XC1 XC2/​XC3 XC4 XD1/​X S1 XD2/​X S2 XD3/​X S3
1 S1 0.4 (10) 0.4 (10) 0.4 (10) 0.6 (15) 0.8 (20) 1.0 (25) 1.2 (30)
2 S2 0.4 (10) 0.4 (10) 0.6 (15) 0.8 (20) 1.0 (25) 1.2 (30) 1.4 (35)
3 S3 0.4 (10) 0.4 (10) 0.8 (20) 1.0 (25) 1.2 (30) 1.4 (35) 1.6 (40)
4 S4 0.4 (10) 0.6 (15) 1.0 (25) 1.2 (30) 1.4 (35) 1.6 (40) 1.8 (45)
5 S5 0.6 (15) 0.8 (20) 1.2 (30) 1.4 (35) 1.6 (40) 1.8 (45) 2.0 (50)
6 S6 0.8 (20) 1.0 (25) 1.4 (35) 1.6 (40) 1.8 (45) 2.0 (50) 2.2 (55)

Table 12.8 Concrete exposure categories.

Summary of exposure categories for concrete by exposure class by per ACI 318-​1 9,
Table 19.3.1.1

Exposure
Item Category class Condition
1 Freezing and F0 Concrete not exposed to freezing-​a nd-​t hawing cycles
thawing (F)
F1 Concrete exposed to freezing-​a nd-​t hawing cycles with
limited exposure to water
F2 Concrete exposed to freezing-​a nd-​t hawing cycles with
frequent exposure to water
F3 Concrete exposed to freezing-​and-​thawing cycles with frequent
exposure to water and exposure to deicing chemicals
2 Sulfate (S) Water-​s oluble sulfate (SO 42–​) Dissolved sulfate (SO 42–​)
in soil, percent (%) by mass [1]‌ in water, ppm [2]‌
S0 SO 42–​ < 0.10 SO 42–​ < 150
S1 0.10 ≤ SO 42–​ < 0.20 150 ≤ SO 42–​ < 1,500 or
seawater
S2 0.20 ≤ SO 42–​ ≤ 2.00 1,500 ≤ SO 42–​ ≤ 10,000
S3 SO 42–​ > 2.00 SO 42–​ >10,000
3 In contact W0 Concrete dry in service
with water
(W) W1 Concrete in contact with water and low permeability is required
W2 Concrete in contact with water where low permeability is required
4 Corrosion C0 Concrete dry or protected from moisture
protection of
reinforcement C1 Concrete exposed to moisture but not to an external source of
(C) chlorides
C2 Concrete exposed to moisture and an external source
of chlorides from deicing chemicals, salt, brackish water,
seawater, or spray from these sources
Special Table Notes:
‌ Percent sulfate by mass in soil shall be determined by ASTM C1580.
[1]

‌ Concentration of dissolved sulfates in water, in ppm, shall be determined by ASTM D516 or ASTM D4130.
[2]
 newgenrtpdf
Table 12.9 Requirements for concrete by exposure class.

738
Summary of requirements for concrete by exposure class by per ACI 318-​1 9, Table 19.3.2.1

Handbook of Precast Segmental Tunnel Lining Systems

Additional requirements
Exposure Maximum Minimum f c′, Limits on cementitious
Item class w/​c ratio [1,2] psi (MPa) Air content materials
1 F0 N/​A 2,500 (17.2) N/​A N/​A
2 F1 0.55 3,500 (24.1) ACI 318-​1 9, Table 19.3.3.1 N/​A
3 F2 0.45 4,500 (31.0) ACI 318-​1 9, Table 19.3.3.1 N/​A
4 F3 0.40 [3]‌
5,000 [3]‌
(34.5) ACI 318-​1 9, Table 19.3.3.1 ACI 318-​1 9, Art.
26.4.2.2(b)
Cementitious materials [4]‌—​t ypes
Calcium Chloride
ASTM C150 ASTM C595 ASTM C1157 Admixturess

5 S0 N/​A 2,500 (17.2) No type restriction No type restriction No type No restriction

restriction
6 S1 0.50 4,000 (27.6) II [5]‌[6] Types IP, IS, or IT with MS No restriction
(MS) designation
7 S2 0.45 4,500 (31.0) V [6]‌ Types IP, IS, or IT with HS Not permitted
(HS) designation
8 S3 0.45 4,500 (31.0) V plus pozzolan or Types IP, IS, or IT with HS plus Not permitted
slag cement [7]‌ (HS) designation plus pozzolan or
pozzolan or slag cement [7]‌ slag cement [7]‌
0.40 5,000 (34.5) V [8]‌ Types with (HS) HS Not permitted
designation
9 W0 N/​A 2,500 (17.2) None
10 W1 N/​A 2,500 (17.2) ACI 318-​1 9, Article 26.4.2.2(d)
11 W2 0.50 4,000 (27.6) ACI 318-​1 9, Article 26.4.2.2(d)
 Maximum water-​s oluble chloride ion (Cl –​)
content in concrete, percent by weight of
cement [9,10] Additional
provisions
Non-​p restressed Prestressed concrete
concrete
12 C0 N/​A 2,500 (17.2) 1.00 0.06 None
13 C1 N/​A 2,500 (17.2) 0.30 0.06
14 C2 0.40 5,000 (34.5) 0.15 0.06 Concrete cover [11]
Special Table Notes:
‌ The w/​c is based on all cementitious and supplementary cementitious materials in the concrete mixture.
[1]

‌ The maximum w/​c limits do not apply to lightweight concrete.

[2]
[3]
For plain concrete, the maximum w/​c shall be 0.45 and the minimum f c′ shall be 4500 psi.

Durability and service life of precast segmental linings

[4]‌
Alternative combinations of cementitious materials to those listed are permitted for all sulfate exposure classes when tested for sulfate resistance and
meeting the criteria in 26.4.2.2(c).
[5]‌
For seawater exposure, other types of Portland cements with tricalcium aluminate (C3A) contents up to 10 percent are permitted if the w/​c m does not
exceed 0.40.
[6]‌
Other available types of cement such as type I or type III are permitted in exposure classes S1 or S2 if the C A contents are less than 8 percent for exposure
class S1 or less than 5 percent for exposure class S2.
‌
[7]
The amount of the specific source of the pozzolan or slag cement to be used shall be at least the amount that has been determined by service record to
improve sulfate resistance when used in concrete containing type V cement. Alternatively, the amount of the specific source of the pozzolan or slag cement
to be used shall be at least the amount tested in accordance with ASTM C1012 and meeting the criteria in 26.4.2.2(c).
[8]‌
If type V cement is used as the sole cementitious material, the optional sulfate resistance requirement of 0.040 percent maximum expansion in ASTM C150
shall be specified.
[9]‌
The mass of supplementary cementitious materials used in determining the chloride content shall not exceed the mass of the Portland cement.
[10]
Criteria for determination of chloride content are in 26.4.2.2.
[11]
Concrete cover shall be in accordance with 20.5.

739
740 Handbook of Precast Segmental Tunnel Lining Systems

Another difference between the two codes is that ACI 318-​19 does not set requirements for
minimum cement content. Also, in contrast to EN 1992-​1-​1:2004, the minimum amount
of concrete cover required by ACI 318-​19 for corrosion protection of reinforcement is a
function of casting type. The two concrete casting types considered in ACI 318-​19 for this
matter are cast-​in-​place and precast concrete manufactured under plant conditions. For pre-
cast elements such as tunnel segments, ACI 318-​19 categorizes the minimum cover based
on exposure conditions into two categories of exposed or not exposed to weather or in con-
tact with the ground. Additional considerations for concrete cover over reinforcement in

Table 12.10 Specified concrete cover.

Specified concrete cover for precast non-​p restressed or prestressed concrete
members manufactured under plant conditions per ACI 318-​1 9, Table 20.5.1.2.3

Specified cover
Concrete Structural
Item exposure member Reinforcement details Metric Imperial
1 Exposed to Walls No.14 and No.18 bars; 38 mm 1-​1 /​2 ”
weather or tendons larger than 1-​1 /​2 ”
in contact (38 mm) diameter
with ground
No.11 bars and smaller, W31 19 mm 3/​4 ”
or D31 wire and smaller;
tendons and strands 1-​1 /​2 ”
(38 mm) diameter and smaller
All other No.14 and No.18 bars; 51 mm 2”
tendons larger than 1-​1 /​2 ”
(38 mm) diameter
No.6 through No.11 bars, 38 mm 1-​1 /​2 ”
tendons and strands larger
than 5/​8 ” (16 mm) diameter
through 1-​1 /​2 ” (38 mm)
diameter
No.5 bar, W31 or D31 32 mm 1-​1 /​4 ”
wire and smaller; tendons
and strands 5/​8 ” (16 mm)
diameter or smaller
2 Not exposed Slab, joists No.14 and No.18 bars; 32 mm 1-​1 /​4 ”
to weather and walls tendons larger than 1-​1 /​2 ”
or in contact (38 mm) diameter
with ground
Tendons and strands 1-​1 /​2 ” 19 mm 3/​4 ”
(38mm) diameter and smaller
No.11 bar, W31 and D31 wire 16 mm 5/​8 ”
and smaller sizes
Beams, Primary reinforcement Greater of
columns, d b and 5/​
pedestals 8” (16 mm)
and tension and need not
ties exceed 1-​1 /​2 ”
(38 mm)
Stirrups, ties, spirals and 9.5 mm 3/​8 ”
hoops
 Durability and service life of precast segmental linings 741

Table 12.11 Total air content for concrete.

Total air content for concrete exposed to cycles of freezing and thawing per ACI
318-​1 9, Table 19.3.3.1

Nominal maximum aggregate size Target air content (%)

Item Metric Imperial F1 F2 F3
1 9.5 mm 3/​8 ” 6.0 7.5 7.5
2 12.7 mm 1/​2 ” 5.5 7.0 7.0
3 19.1 mm 3/​4 ” 5.0 6.0 6.0
4 25.4 mm 1” 4.5 6.0 6.0
5 38.1 mm 1-​1 /​2 ” 4.5 5.5 5.5
6 50.8 mm 2” 4.0 5.0 5.0
7 76.2 mm 3” 3.5 4.5 4.5

ACI 318-​19, are the types of members categorized into two categories of walls or all other
(Table 12.10).
The last parameter affecting the concrete cover selection considered in ACI 318-​19, is the
size of reinforcing bars. For the tunnel lining, the concrete and member categories should
be selected as “exposed to weather or in contact with ground” and “all other,” respectively.
Because reinforcing bars smaller than or equal to No.6 (metric bar size No.19) are commonly
used in precast tunnel segments, for corrosion exposure class C2 typically encountered in
tunnels, ACI 318-​19 specifies a minimum cover of 1.25 to 1.50 inches (32 to 38 mm). This
is in contrast with EN 1992-​1-​1:2004, which requires different minimum concrete covers,
ranging from 0.40 to 1.8 inches (10 to 45 mm), based on the source of corrosion that can
be either carbonation, chloride ions other than seawater, or seawater exposure.
Following recommendations of ACI 318-​19, when a concrete structure is subject to
freezing-​and-​thawing (exposure classes F1, F2 or F3), the concrete should be air entrained.
The minimum air content should be selected based on the nominal maximum aggregate
size (MAS) of concrete as presented in Table 12.11 (ACI 318-​19, Table 19.3.3.1). However,
this code allows for a reduction of air content by 1% from the values in Table 12.11, when
specified compressive strength is higher than 5,000 psi (35 MPa). Precast concrete seg-
ments always have a higher strength than this value, and the nominal MAS is often between
0.50 inches (12.5 mm) and 0.75 inches (19 mm).
For precast concrete tunnel segments subject to freezing-​and-​thawing, an exposure class
F2 or F3 is commonly selected. Therefore, referring to Table 12.11, and considering the
MAS and 1% reduction allowance for concrete strength of 5,000 psi (35 MPa) or higher,
air content of 5 and 6% are recommended for concrete with MAS of 0.75 inches (19 mm)
and 0.50 inches (12.5 mm), respectively.
Consideration of a common exposure class example, such as when the concrete lining
is exposed to chloride-​induced corrosion, can help provide insight into how disparate the
ACI and EN concrete codes are. For this purpose, requirements of exposure category C2 of
ACI 318-​19 are compared with exposure class XS3/​XD3 of EN 1992-​1-​1:2004. ACI code
requires a maximum water/​cement ratio of 0.40 and a minimum compressive strength of
5,000 psi (35 MPa) whereas EN code requires a maximum water/​cement ratio of 0.45, a
minimum compressive strength of 5,000 psi (35 MPa), and a minimum cement content of
742 Handbook of Precast Segmental Tunnel Lining Systems

575 lbs/​yd3 (340 kg/​m3). Minimum concrete cover of 1.5 inches (38 mm) specified by ACI
318-​19 for reinforcing bar No.6 (metric bar size No.19) can be compared to the 1.8 inches
(45 mm) EN 1992-​1-​1:2004 requirement. This example shows that, despite the differences
between the two methods, the concrete requirements set forth by both codes for the same
exposure condition are similar and will most likely result in similar, if not identical, concrete
specifications (Bakhshi and Nasri, 2018b).

12.10.3 Canadian Standards Association (CSA) method

The Canadian Standards Association code CSA A23.1-​19 is commonly used in Canada for
the design and construction of concrete structures. As far as durability is concerned, CSA
A23.1-​19 defines the following exposure classes: C, F, N, A, S and R. The definition of these
exposure classes is given in Table 12.12 (CSA A23.1-​19, Table 1). Table 12.13 (CSA A23.1-​
19, Table 2) presents the mix design requirements for these exposure classes.

Table 12.12 Exposure classes by Canadian Standards Association.

Definitions of C, F, N, A, S and R exposure classes per Canadian Standards
Association

Exposure
Item class Definition and condition
1 C-​X L Structurally reinforced concrete exposed to chlorides or other severe
environments with or without freezing and thawing conditions, with
higher durability performance expectations than the C-​1 classes.
2 C-​1 Structurally reinforced concrete exposed to chlorides with or without
freezing and thawing conditions. Examples:
• Bridge decks, parking decks and ramps
• Portions of structures exposed to seawater located within the tidal
and splash zones
• Concrete exposed to seawater spray, and saltwater pools
For seawater or seawater-​s pray exposures the requirements for S-​3
exposure also have to be met.
3 C-​2 Non-​s tructurally reinforced (i.e. plain) concrete exposed to chlorides
and freezing and thawing. Examples:
• Garage floors, porches and step,
• Pavements, sidewalks, curbs, and gutters
4 C-​3 Continuously submerged concrete exposed to chlorides, but not to
freezing and thawing. Examples:
• Underwater portions of structures exposed to seawater.
For seawater or seawater-​s pray exposures the requirements for S-​3
exposure also have to be met
5 C-​4 Non-​s tructurally reinforced concrete exposed to chlorides, but not to
freezing and thawing. Examples:
• Underground parking slabs on-​g rade
6 F-​1 Concrete exposed to freezing and thawing in a saturated condition, but
not to chlorides. Examples:
• Pool decks, patios and tennis courts
• Freshwater pools and freshwater control structures
 Durability and service life of precast segmental linings 743

Table 12.12 (Continued)

Exposure
Item class Definition and condition
7 F-​2 Concrete in an unsaturated condition exposed to freezing and thawing,
but not to chlorides. Examples:
• Exterior walls and columns
8 N Concrete that when in service is neither exposed to chlorides nor
to freezing and thawing nor to sulfates, either in a wet or dry
environment. Examples:
• Footings and interior slabs
• Walls and columns
9 N-​C F Interior concrete floors with a steel-​t rowel finish that are not exposed
to chlorides, nor to sulfates either in a wet or dry environment.
Examples:
• Interior floors, surface covered applications (carpet, vinyl tile)
• Surface exposed applications (with or without floor hardener)
• Ice-​h ockey rinks and freezer warehouse floors
10 A-​X L Structurally reinforced concrete exposed to:
• Severe manure and/​o r silage gases, with or without freeze-​– ​t haw
exposure
• Concrete exposed to the vapor above municipal sewage or industrial
effluent, where hydrogen sulfide gas might be generated, with higher
durability performance expectations than A-​1 class
11 A-​1 Structurally reinforced concrete exposed to severe manure and/​
or silage gases, with or without freeze-​t haw exposure. Concrete
exposed to the vapor above municipal sewage or industrial effluent,
where hydrogen sulfide gas might be generated. Examples:
• Reinforced beams, slabs, and columns over manure pits
• Silos, canals, and pig slats; and access holes, enclosed chambers
• Pipes that are partially filled with effluents
12 A-​2 Structurally reinforced concrete exposed to moderate to severe
manure and/​o r silage gases and liquids, with or without freeze-​t haw
exposure. Examples:
• Reinforced walls in exterior manure tanks
• Silos and feed bunkers, and exterior slabs
13 A-​3 Structurally reinforced concrete exposed to moderate to severe
manure and/​o r silage gases and liquids, with or without freeze-​
thaw exposure in a continuously submerged condition. Concrete
continuously submerged in municipal or industrial effluents.
Examples:
• Interior gutter walls, beams, slabs, and columns
• Sewage pipes that are continuously full (e.g., force mains)
• Submerged portions of sewage treatment structures
14 A-​4 Non-​s tructurally reinforced concrete exposed to moderate manure
and/​o r silage gases and liquids, without freeze-​t haw exposure.
Examples:
• Interior slabs on grade
15 S-​1 Concrete subjected to very severe sulfate exposures (Tables 2 and 3).
(Continued)
744 Handbook of Precast Segmental Tunnel Lining Systems

Table 12.12 (Continued)

Exposure
Item class Definition and condition
16 S-​2 Concrete subjected to severe sulfate exposure (Tables 2 and 3).
17 S-​3 Concrete subjected to moderate sulfate exposure and to seawater or
seawater spray
(Tables 2 and 3).
18 R-​1 Residential concrete for footings for:
• Walls and columns
• Fireplaces and chimneys
19 R-​2 Residential concrete for:
• Foundation walls
• Grade beams and piers, etc.
20 R-​3 Residential concrete for interior slabs on ground not exposed to
freezing and thawing or de-​icing salts.
Source: CSA A23.1:19, Table 1.
Table Notes:
1
“C” Classes pertain to chloride exposure
2
“F” Classes pertain to freezing and thawing exposure without chlorides
3
“N” Class is exposed to neither chlorides not freezing and thawing
4
All classes of concrete exposed to sulphates shall comply with the minimum requirements of Class S, as
noted in Tables 2 and 3 (of CSA A23.1:19). In particular, exposure classes A1 to A-​4 and A-​X L in municipal
sewage elements could be subjected to sulphate exposure.
5
No hydraulic cement concrete will be entirely resistant in severe acid exposures. The resistance of
hydraulic cement concrete in such exposures is largely dependent on its resistance to penetration of fluids.
6
Decision of exposure class should be based upon the service conditions of the structure or structural
element, and not upon the condition during construction.

The CSA code takes an approach towards concrete exposure classes that is relatively
similar to that of ACI 318-​19. However, one of the differences between the two codes is
the exposure class W, which is defined in ACI 318-​19 for concrete elements in contact with
water but does not exist in the Canadian code. Also, exposure class N in CSA A23.1-​19,
which does not exist in ACI 318-​19, is basically equivalent to exposure classes C0, F0 and S0
combined. Another difference between the two codes is the two additional exposure classes
in CSA code: exposure class A and exposure class R. exposure class A in CSA A23.1-​19 is
related to concrete exposure to the manure and silage gases or liquids, and the vapor above
the municipal sewage or the industrial effluent. Examples of such structures are tanks, silos,
canals and pipes, which are considered environmental structures.
ACI set durability requirements for such structures not in ACI 318-​19, but in another
code specific to environmental structures: ACI 350-​06. Exposure class A in CSA can be
applied to sewage and waste-​water tunnels. One last major difference between ACI and
CSA codes regarding exposure classification, is the exposure class R in CSA code which
does not exist in ACI 318-​19 and is specifically defined for residential concrete. This specific
exposure class is not relevant to tunnels, and therefore, will not be discussed in this chapter.
The exposure category C in CSA A23.1-​19 corresponds to all reinforced concrete exposed
to chloride. Therefore, this CSA category is equivalent to category C in ACI 318-​19 and can
be compared to XC, XD/​XS exposure classes in EN 1992-​1-​1 (2004) and EN 206-​1 (2013).
Because tunnel linings are usually designed for a service life of more than 100 years, the rec-
ommended exposure class for chloride-​induced corrosion based on CSA code is C-​XL. This
 newgenrtpdf
Table 12.13 Requirements for exposure classes by the Canadian Standards Association.
Definitions of C, F, N, A and S exposure classes per the Canadian Standards Association

Air content category Curing type

per CSA Table 4 d (see Table 19 of CSA 23.1-​1 9)
Minimum specified Chloride ion
Maximum water-​ compressive Not Exposed penetrability
Exposure to-​c ementing strength (MPa) Exposed to Cycles to Cycles of Normal requirements
Item class a materials ratio b and age at test b,i of Freeze/​T haw Freeze/Thaw Concrete HVSCM-​1 HVSCM-​2 and age at test c
1 C-​X L 0.40 50 MPa 1 e
3 3 3 < 1,000
or within 56 days coulombs
A-​X L within 91 days
2 C-​1 or 0.40 35 MPA 1 e
2 3 2 < 1,500
A-​1 within 56 days coulombs

Durability and service life of precast segmental linings

within 91 days
3 C-​2 or 0.45 h 32 MPa 1 N/​A 2 2 2 —​
A-​2 at 28 days
4 C-​3 or 0.50 30 MPa N/​A e
1 2 2 —​
A-​3 at 28 days
5 C-​4 e or 0.55 25 MPa N/​A e
1 2 2 —​
A-​4 at 28 days
6 F-​1 0.50 j 30 MPa 1 N/​A 2 3 2 —​
at 28 days
7 F-​2 or 0.55 j 5 MPa 2f N/​A 1 2 2 —​
R-​1 or at 28 days
R-​2
8 N As per the mix For structural None e
1 2 2 —​
design for the design
strength required
9 N-​C F g 0.55 25 MPa N/​A e
1 2 2 —​
or R-​3 at 28 days

(Continued)

745
 newgenrtpdf
Table 12.13 (Continued)

746
Air content category Curing type

Handbook of Precast Segmental Tunnel Lining Systems

per CSA Table 4 d (see Table 19 of CSA 23.1-​1 9)
Minimum specified Chloride ion
Maximum water-​ compressive Not Exposed penetrability
Exposure to-​c ementing strength (MPa) Exposed to Cycles to Cycles of Normal requirements
Item class a materials ratio b and age at test b,i of Freeze/​T haw Freeze/Thaw Concrete HVSCM-​1 HVSCM-​2 and age at test c
10 S-​1 0.40 35 MPa 1 e
2 3 2 —​
within 56 days
11 S-​2 0.45 j 32 MPa 1 e
2 3 2 —​
within 56 days
12 S-​3 0.50 j 30 MPa 1 e
1 2 2 —​
within 56 days
Source: CSA A23.1:19, Table 2.
Special Table Symbols:
a
See Table 1 for a description of concrete exposure classes.
b
The minimum specified compressive strength may be adjusted to reflect proven relationships between strength and the water to cementitious materials ratio
provided that freezing and thawing and de-​icer scaling resistance have been demonstrated to be satisfactory. The water-​t o-​c ementitious materials ratio shall
not be exceeded for a given exposure class.
c
In accordance with CSA A23.2-​2 3C, an age different from that indicated may be specified by the owner. Accelerated moist curing in accordance with CSA
A23.2-​2 3C may be specified by the owner; in such cases, the age at test shall be 28 days. Where calcium nitrite corrosion inhibitor is to be used, the same
concrete mixture, without calcium nitrite, shall be qualified to meet the requirements for the permeability index in this Table. For field testing, the owner shall
specify the type of specimen and location from which it is taken. If cores are required, the concrete cores shall be taken in accordance with CSA A23.2-​2 3C.
d
Air entrained concrete shall not receive a steel troweled finish. See Note 4 to clauses 7.7.4.3.1 and 7.7.4.3.2.
e
Class N-​C F concrete shall not contain an air entraining admixture. Other classes of concrete falling into this air content category have no requirement to
provide entrained air, however, the producer may choose to add entrained air in order to modify plastic concrete properties such as bleeding, workability,
cohesiveness, etc. No air entrainment shall be added to concrete which is to receive a steel trowel finish.
f Air entrainment shall be waived for F-​2 exposure class frozen in an air dry condition and receiving very limited cycles of freeze/​t haw. Interior ice rink slabs
brought up to sub-​z ero levels before the introduction of water and dry freezer slabs have been found to perform satisfactorily without entrained air when
steel troweled.
g
See Clause 7.1.2 for concrete mixes for concrete floors.
h
The minimum water-​t o-​c ementitious material ratio for HVSCM-​1 concrete in an exposure class C-​2 condition shall not exceed 0.40.
i
A different age at test may be specified by the owner to meet structural or other requirements.
j
For reinforced concrete surfaces exposed to air and not directly exposed to precipitation, with depths of cover less than 50 mm, the water-​t o-​c ementitious
materials ratio shall not be greater than 0.40 for HVSCM-​1 concrete and not greater than 0.45 for HVSCM-​2 concrete. This requirement is intended to min-
imize the risk of corrosion of embedded steel due to carbonation of the concrete cover. The exposure conditions that present the greatest risk are the soffits
of suspended slabs and balconies and exposed vertical surfaces that receive little direct precipitation. For concrete that is continuously moist, the process of
carbonation will be very slow.
 Durability and service life of precast segmental linings 747

exposure class is defined for structures with higher durability performance expectations
than the C-​1 class, which is often defined for a service life of less than 100 years.
The recommended exposure class for sewage and waste-​water tunnels is A-​XL, which
is applied to structures with higher durability expectations than exposure class A-​1, with
usual service life expectancy of less than 100 years. Exposure category F deals with con-
crete elements exposed to freezing and thawing but not subject to chloride penetration. This
category corresponds to ACI 318-​19 exposure class F but is less specific. Once tunnels are
subject to freezing-​and-​thawing, near portal areas for example, exposure class F-​1 should
be selected for specifying concrete requirements. Similar to ACI 318-​19, CSA A23.1-​19 only
considers sulfate attack category S, while EN 1992-​1-​1 (2004) and EN 206-​1 (2013) cover a
wide range of chemical attacks to different types of ions and acids other than sulfate. When
tunnels are subject to sulfate attack, an exposure class between S1 and S3 is selected based
on the amount of water-​soluble sulfate found in soil and groundwater (Table 12.14).
Much like ACI 318-​19 and in contrast with Eurocode, CSA code does not have any
requirements for the minimum cement content for different exposure classes. As shown in
Table 12.13, the main concrete requirements of CSA are the maximum water-​to-​cementing
materials ratio (water/​cement ratio), the minimum compressive strength, the minimum air
content (Table 12.15), the curing type, and the maximum chloride ion penetrability. Note
that curing type requirements of CSA A23.1-​19 do not apply to precast elements and are
only applicable to cast-​in-​place concrete.
When the tunnel lining is exposed to chloride penetration or deterioration may occur due
to sewage exposure, concrete requirements of Table 12.13 for exposure class C-​XL or A-​XL
should be met. These include a maximum water/​cement ratio of 0.40, a minimum compres-
sive strength of 7,250 psi (50 MPa) within 56 days, a minimum air content according to
Table 12.15 (CSA A23.1-​19, Table 4), and a maximum chloride ion penetration of 1,000
coulombs after 91 days of curing. Because maximum aggregate size (MAS) for precast seg-
ments is usually between 0.55 to 0.78 inches (14 to 20 mm), the required air content for
exposure class C-​XL or A-​XL based on CSA code is 5 to 8%.
For the tunnel lining subject to freezing and thawing, concrete requirements for exposure
class F-​1 is water/​cement ratio of less than 0.50, minimum compressive strength of 4,350
psi (30 MPa) at 28 days, and minimum air content of 5 to 8% (see Table 12.13 and 12.15).
Concrete requirements of CSA A23.1-​19 for structures exposed to sulfate attack (exposure
class S) are presented in Tables 12.13 and 12.14, which may vary according to the amount
of the soluble sulfate that the tunnel lining is exposed to.
For durability against chloride penetration, similar to ACI 318-​19 and EN 1992-​1-​
1:2004, the minimum concrete cover required by CSA is a function of exposure conditions.
As shown in Table 12.16 (CSA A23.1-​19, Tables 17), CSA A23.1-​19 categorizes the min­­
imum cover primarily based on exposure conditions and element types. Additionally, CSA
code limits the ratios between the concrete cover and the nominal bar diameter, as well as
the concrete cover and nominal MAS. Nonetheless, in precast tunnel segments with MAS
and rebar diameter of less than 0.78 inches (20 mm), CSA limitations of the minimum cover
related to these parameters do not govern.
Because tunnel linings are always cast against and permanently exposed to earth, CSA
requires a minimum cover of 2.95 inches (75 mm) for all different exposure conditions.
However, CSA A23.4-​16 (2021), which provides specific standards for precast concrete in
Canada, allows a reduction of concrete cover from the values set forth by CSA A23.1-​19 for
cast-​in-​place concrete. Greater dimensional control of formed concrete, tighter tolerances
on placement of reinforcing, and better quality of concrete in plant-​controlled conditions
 newgenrtpdf
748
Table 12.14 Additional requirements for concrete subject to sulphate attack.
Definition of additional requirements for concrete subject to sulphate attack per Canadian Standards Association (CSA

Handbook of Precast Segmental Tunnel Lining Systems

A23.1:19, Table 3). (SI unit mg/​l is equivalent to 0.0000083 lb/​g al)

Performance
requirements 6
Water soluble
Water-​s oluble sulfate (SO 4) Sulfate (SO 4) Maximum expansion
sulfate (SO 4) in in recycled in groundwater Cementing when tested using CSA
Exposure Degree of soil sample 2 aggregate sample samples 3 materials A3004- ​C 8 (%)
Item class exposure (%) (%) (mg/​l ) to be used 4 6 months 12 months 7
1 S-​1 Very severe > 2.0 > 2.0 > 10,000 HS 5, HSb, HSLb or 0.05 0.1
Hse
2 S-​2 Severe 0.20 to 2.0 0.60 to 2.0 1,500 to 10,000 HS 5, HSb, HSLb or 0.05 0.1
Hse
3 S-​3 Moderate 0.10 to 0.20 0.20 to 0.60 150 to 1,500 MS, MSb, MSe, MSLb, 0.1
(including seawater LH, LHb, HS 5, HSb,
exposure) 1 HSLb or Hse
Table Notes:
1
For seawater exposure, also see CSA A23.1:19, clause 4.1.1.5.
2
In accordance with CSA A23.2-​3 B.
3
In accordance with CSA A23.2-​2 B.
4
Where combination of supplementary cementitious materials and Portland, Portland-​limestone, or blended hydraulic cements are to be used in the concrete
mix design instead of the cementitious materials listed, and provided they meet the performance requirements demonstrating equivalent performance against
sulphate exposure, they shall be designated as MS equivalent (MSe) of HS equivalent (HSe) in the relevant sulphate exposures (see CSA 23.1 clause 4.1.1.6.2,
4.2.1.1, 4.2.1.3 and 4.2.1.4.
5
Type HS cement shall not be used in reinforced concrete exposed to bot chlorides and sulphates, including seawater. See CSA A23.1, clause 4.1.1.6.3.
6
For demonstrating equivalent performance, use the testing frequency in Table 1 of CSA A3004-​A 1 and see the applicable notes to Table A3 in CSA A3001 with
regard to re-​e stablishing compliance of the composition of the cementitious materials used to establish compliance charges.
7
If the expansion is greater than 0.05% at 6 months but less than 0.10% at one year, the cementitious materials combination under test shall be considered to
have passed.
Special Notes:
1
Limestone fillers shall not be used in concrete for any exposure class S listed in CSA A3001, Tables 1 to 3. Portland-​limestone cement shall not be used as
the sole cementitious materials in concrete for any exposure class S in CSA A3001, Table 1 to 3. However, blended hydraulic cements, or combinations of
Portland-​limestone cement and the minimum levels of supplementary cementitious materials listed in CSA A3001, Table 9, and also meeting the test require-
ments in CSA A3001, Table 5 may be used in any exposure class S listed in listed in Tables 1 to 3.
 Durability and service life of precast segmental linings 749

Table 12.15 Requirements for air content categories.

Requirements for air content categories per the Canadian Standards Association

Range in air content *for concretes with indicated nominal

maximum sizes of coarse aggregate
Air content 10 mm 14 to 20 mm 28 to 40 mm
Item category (0.40 inches) (0.55 to 0.78 inches) (1.10 to 1.57 inches)
1ᵻ 1 6 to 9 5 to 8 4 to 7
2 2 5 to 8 4 to 7 3 to 6
Source: CSA A23.1:19, Table 4.
Table Notes:
ᵻ
At the point of discharge from the delivery equipment, unless otherwise specified.
*
For hardened concrete, see CSA A23.1, clause 4.3.3.2.
Special Notes:
1
The above difference in air contents has been established based upon the difference in mortar fraction
volumes required for specific coarse aggregate sizes.
2
Air contents measured after pumping or slip forming can be significantly lower than those measured at the
end of the chute

Table 12.16 Concrete cover according to exposure condition.

Recommendations for concrete cover according to exposure condition per the
Canadian Standards Association (CSA A23.1-​1 9, Table 17) (unit is in mm; multiply
by 0.039 to convert to inches)

Exposure class (see Tables 1 and 2)

F-​1 , F-​2 , S-​1 , S-​2 , C-​X L, A-​X L, C-​1 ,
N, N- ​C F, R-​3 S-​3 , R-​1 , R-​2 C-​3 , A-​1 , A-​2 , A-​3
Item Exposure condition Metric Imp. Metric Imp. Metric Imp.
1 Cast against and 75 mm 3.0” 75 mm 3.0” 75 mm 3.0”
permanently exposed to
earth, including footings and
piles
2 Beams, girders, and columns 30 mm* 1.20” 40 mm 1.60” 60 mm 2.40”
3 Slabs, walls, joists, shells, 20 mm* 0.80” 40 mm 1.60” 60 mm 2.40”
and folded plates
4 Ratio of cover to nominal 1.0 mm* 0.04” 1.5 mm 0.06” 2.0 mm 0.08”
bar diameter†
5 Ratio of cover to nominal 1.0 mm*‡ 0.04” 1.5 mm 0.06” 2.0 mm 0.08”
maximum aggregate size
Special Notes:
*
This refers only to concrete that will be continually dry within the conditioned space (i.e., members
entirely within the vapor barrier of the building envelope).
†
The cover for a bundle of bars shall be the same as that for a single bar with an equivalent area.
‡
The specified cover from screeded surfaces shall be at least 1.5 times the nominal Maximum Aggregate Size
to reduce interference between aggregate and reinforcement where variations in bar placement result in
a cover smaller than specified.
 n

750 Handbook of Precast Segmental Tunnel Lining Systems

are among the reasons provided by CSA A23.4-​16 for justifying this reduction. Concrete
cover limitations by CSA A23.4-​16 for precast elements as a function of exposure condition
is shown in Table 12.17 (CSA A23.4-​16, Table 1). Piles are the closest element type in this
table to the tunnel lining and, therefore, for exposure classes C-​XL or F-​1/​F-​2, designing a
minimum cover of 1.57 inches (40 mm) is adequate in this example.

12.11 PERFORMANCE-​B ASED APPROACHES

Prescriptive design methodologies require concrete specifications to achieve a very dense
and high-​quality concrete. However, the major flaw of prescriptive approaches is the lack
of direct connection between the limiting requirements and the main source of degrad-
ation mechanisms for each specific type of concrete damage. In addition, these methods
do not explicitly provide quantitative service life predictions for structures. In contrast,
performance-​based design approaches, despite all challenges related to these methods, pro-
vide significant benefits to owners by focusing on the specific sources of concrete damages
in a project-​specific fashion (Swiss Standard SIA 262 2003).
In order to achieve a performance design, rapid, easy and reliable test methods are needed
to assess properties of the structural concrete. Studies such as Rashidi and Nasri (2012), Sigl
et al. (2000) and Li et al. (2015) provide insights into such design methods for combined
sewer overflow tunnels, subway and subsea road tunnels, respectively. Performance-​based
service life design for carbonation and chloride-​induced steel corrosion is briefly explained
in the following section.

12.11.1 Service life design for reinforcement corrosion

Service life prediction models for steel corrosion due to chlorides or carbonation, divide the
concrete deterioration time into two time periods (Figure 12.26). The initial time period
without considerable damage is called incubation. This period ends at the time of corrosion
initiation time, Ti. Within this time period, chloride ions or carbonation penetrate through
the concrete surface until critical chloride concentration of concrete at the level of reinforce-
ment is reached. At this time, steel reinforcement becomes depassivated and is likely to
initiate the corrosion process. Chloride-​induced corrosion, Ti is often associated with the
service life of the element. The following time period is called corrosion propagation. The
end of this period, Tp, is a function of limit state conditions. For example, as Figure 12.26
shows, Tp1 is the corrosion propagation time for localized cracking and spalling considered
as the serviceability limit state (SLS). Tp2 is the ultimate limit state (ULS) corrosion propa-
gation time, which corresponds to the bearing capacity failure due to loss of bond, and
reduced steel and concrete sections.
Most well-​known service life prediction models such as DuraCrete (Engelund, et al,
2000), fib Bulletin No. 34 (2006), and Life-​365 (2012) use the simplified Fick's second law
for non-​steady-​state diffusion for modeling chloride penetration in concrete. Assuming the
coefficient of diffusion and the chloride concentration at the concrete surface are constant,
Fick’s second law can be presented as

  x 
C ( x, t ) = C0 + (Cs − C0 ) 1 −   (12.1)
  2 Dt  
 newgenrtpdf
Table 12.17 Cover to reinforcement, tendon sheaths and ducts.
Recommendations for minimum cover to reinforcement, tendon sheaths and ducts per the Canadian Standards Association (CSA
A23.4-​1 6, Table 1) (unit is in mm; multiply by 0.039 to convert to inches)

Exposure
Chlorides, sulphates, manure,
sewage and industrial effluents
Protection systemsⴕ
Not Earth or Unprotected
Item Description exposedⱡ weather concrete A B, C, D, E F, G, H, J, K
1 Exposure condition (see N F-​1 , F-​2 C-​1 , C-​3 , A-​1 , C-​X L C-​1 , C-​3 , A-​ C-​1 , C-​3 , A-​1 , C-​1 , C-​3 ,
Table 2 of CSA A23.1) A-​2 , A-​3 , S-​1 , 1, A-​2 , A-​3 , A-​2 , A-​3 , S-​1 , A-​1 , A-​2 ,
S-​2 , S-​3 S-​1 , S-​2 , S-​3 S-​2 , S-​3 A-​3 , S-​1 ,
S-​2 , S-​3

Durability and service life of precast segmental linings

2 Piles –​ 40 mm 50 mm 40 mm –​ 40 mm 40 mm
3 Beams, girders, and columns 20 mm 30 mm 50 mm 40 mm –​ 40 mm 30 mm
4 Slabs, walls, joists, shells, and 20 mm 25 mm 50 mm 40 mm 40 mm 40 mm 35 mm
folded plates
5 Ratio of cover to nominal bar 1 1.5 1.5 1.5 1.5 1.5 1.5
diameter§ (cover ≤ (cover ≤ (cover ≤ (cover ≤ 60 mm) (cover ≤ 60 mm)
60 mm) 60 mm) 60 mm)
6 Ratio of cover to nominal 1.0** 1.5 1.5 1.5 1.5 1.5 1.5
Maximum Aggregate Size (cover ≤ (cover ≤ (cover ≤ (cover ≤ 60 mm) (cover ≤ 60 mm)
60 mm) 60 mm) 60 mm)
7 Cover to mesh reinforcing in 20 mm 25 mm 40 mm 35 mm 30 mm 35 mm 30 mm
all precast products

(Continued)

751
 newgenrtpdf
Table 12.17 (Continued)

752
Table Notes:
*
The cover for precast concrete is reduced from the cover used in cast-​in-​p lace concrete because of greater dimensional control of formed concrete, tighter

Handbook of Precast Segmental Tunnel Lining Systems

tolerance on placement of reinforcing, and better quality of concrete in plant-​c ontrolled conditions.
ᵻ
A
Membrane
B
Concrete with a rapid chloride permeability test result (ASTM C1202) of less than 1,000 coulombs within 91 days and black reinforcing within 100 mm of the
exposed surface.
C
Corrosion inhibitor and black reinforcing within 100 mm of the exposed surface.
D
Sealer and black reinforcing within 100 mm of the exposed surface.
E
Galvanized reinforcing within 100 mm of the exposed surface
F
Sealer and galvanized reinforcing within 100 mm of the exposed surface
G
Concrete with a rapid chloride permeability test result (ASTM C1202) of less than 1,000 coulombs within 91 days and galvanized reinforcing within 100 mm
of the exposed surface.
H
Corrosion inhibitor and galvanized reinforcing with 100 mm of the exposed surface.
J
Sealer and corrosion inhibitor.
K
Stainless steel reinforcing within 100 mm of the exposed surface.
ⱡ
Refers only to concrete that will be continuously dry with a conditioned space (i.e. member entirely within the vapor barrier of a building envelope).
§
The cover for a bundle of bars shall be determined from a circle circumscribing the perimeter of the bundle.
**
The specified cover from screeded surfaces shall be at least 1.5 times the nominal maximum aggregate size to reduce interference between aggregate and
reinforcement where variations in bar placement can result in a cover smaller than specified.
Special Notes:
1
Greater cover or protective coatings can be necessary where there is exposure to industrial chemical, food processing chemicals, and other corrosive materi-
als. See CSA A23.1 and PCA IS001.
 Durability and service life of precast segmental linings 753

Figure 12.26 Reinforced concrete deterioration model.

Note: Reinforced concrete deterioration model by Tuutti (1982) expanded by Nilsson (2012) for
reinforcement corrosion.

where
C ( x, t ) is the chloride concentration in concrete at depth x and time t ;
C0 is initial chloride concentration in concrete;
Cs is the chloride concentration at the surface of concrete;
erf is the error function; and
D is the chloride diffusion coefficient.

The corrosion initiation time, Ti, happens when the chloride concentration at the reinforce-
ment cover depth of d reaches a critical value Ccr . The initial chloride concentration in
concrete, C0 , can be assumed zero ( C0 = 0 ). When solving the above equation for the cor-
rosion initiation time, Ti, C ( x, t ) is replaced by Ccr ( C ( x, t ) = Ccr ) while t is replaced by
Ti ( t = Ti ), and x is replaced by d or x = d . The solution to Equation (12.1) is, therefore,

d2 2
Ti = A (12.2)
4D

where

1
A= (12.3)
 Ccr 
 1 − C 
s

Note that is the inverse of the error function. Following experimental observations, most
service life prediction models predict the chloride diffusion coefficient in concrete will
decrease with time. A common equation in use is presented in Equation (12.4).
754 Handbook of Precast Segmental Tunnel Lining Systems

a
t 
D = D0  0  for t ≤ t a (12.4)
t

where
D is the chloride diffusion coefficient at age t ;
D0 is the chloride diffusion coefficient measured at reference age t0 ; and
a is an exponent that indicates the effect of aging.

t0 is typically taken as 28 days and, therefore, D0 is the chloride diffusion coefficient

measured at 28 days. Well-​known service life prediction models such as DuraCrete
(Engelund et al., 2000) and fib Bulletin No.34 (2006) are based on Equations (12.2–​4).
During the design phase, two major parameters of d (reinforcement cover depth) and
D0 (chloride diffusion coefficient measured at 28 days) are specified in order to achieve
a service life, Ti , of 100 years or more. As such, performance-​based models, as opposed
to prescriptive-​based models, can provide an explicit prediction of the service life of rein-
forced concrete structures. Nevertheless, caution must be taken when using such models
because the computed service life parameter ( Ti ) is extremely sensitive to parameters
Cs , Ccr , a and t a .

• First, it is very difficult to determine parameters Cs and Ccr , and their ratio may affect
the computation of Ti by a factor of up to 4 (Torrent, 2017).
• Second, there are uncertainties with regard to validity of the power decay function
[Equation (12.4)] and the proposed values for age exponent a (Gulikers, 2006, 2011).
• Third, the variation of values considered for exponent a can lead to variation of Ti by
several orders of magnitude (Torrent, 2013, 2017).

While Life-​365 (2012) limits t a to 25 years, DuraCrete (Engelund et al., 2000) and fib
Bulletin No.34 (2006) models do not set limitations for this parameter, which seems
unreasonable. Limiting the t a parameter to a year or less can be justified based on obser-
vations ensuring that after an appropriate curing period, there will be no substantial
decrease in chloride permeability under jobsite conditions (Di Pace et al., 2019). At the
time of service life design, determining parameter a is practically impossible for design-
ers due to the long-​term nature of this exponent. Designer’s reliance on the service life
models, especially on proposed values for parameters a and t a , which may not match
with the jobsite conditions can result in incorrect computation of service life parameter
( Ti ) and is problematic.
To alleviate this problem, a simple yet robust method has been proposed by Torrent
(2015). Although this method is based on Equations (12.2) to (12.4), unclear parameters
are taken-​out to remove uncertainties. This simple method can be used for preliminary
service-​life design, and designers would be able to check whether a sophisticated calculation
can provide a reasonable prediction. Solutions of this simplified model are presented as a
chart in Figure 12.27 for the exposure class XD3/​XS3 per EN 1992-​1-​1:2004. As discussed
in section 12.10.1, this exposure class (XD3/​XS3) is applicable to the case of tunnel linings
subject to chloride penetration and can be directly used for preliminary service-​life design
of precast tunnel segments.
While using these models for predicting service life is reasonable for structures subject
to severe environments, for less severe environments, these models may result in too conser-
vative predictions (Di Pace et al., 2019). Particularly in the case of carbonation-​induced
 Durability and service life of precast segmental linings 755

Figure 12.27 Service life parameters.

Note: Service life parameter (Ti) as a function of d and D 0 for exposure class XS3/​ XD3 per EN
1992-​1 -​1 :2004.
Source: Torrent, 2017. 

corrosion, it is necessary to take propagation time into account. For example, in high-​
performance concrete with concrete covers of 3.0 to 3.5 inches (75 to 90 mm), predicted
propagation time could be more than 20 years before any localized cracking or spalling, as
the serviceability limit state (SLS) appears (Di Pace et al., 2019).
During the design phase, in addition to guidelines on mixture characteristics, a com-
prehensive experimental program should be carried-​out on concrete mixtures to include
chloride permeability tests. One of the most common tests is the rapid chloride perme-
ability test (RCPT) per ASTM C1202 (2019), which measures the electrical charge, Q
(in coulombs), passed through the concrete in 6 hours. Equation (12.5) proposed by Olek
et al. (2002) can be used to estimate the chloride diffusivity from the electrical charge, Q,
determined by RCPT.

D0 (10−12 m2 /s ) = 0.4 + 0.002Q (12.5)

Experimental results for precast concrete tunnel segments, when tested according to ASTM
C1202 (2019), usually range between 300 to 700 coulombs. Using Equation (12.5), this
translates into a diffusion coefficient of 1 to 1.8 ×10−12 m2 /s . An example following the
simplified service life prediction model of Figure 12.27 considers the design service life of
a tunnel exposed to chloride-​induced corrosion at 100 years; the designed concrete cover
over reinforcement is in the range of 1.25 to 1.8 inches (32 to 45 mm). This result is in
agreement with prescriptive-​based code recommendations of ACI 318-​19, CSA A23.1-​19
and EN 1992-​1-​1:2004, which require a minimum cover of 1.5 inches (38 mm), 1.6 inches
(40 mm) and 1.8 inches (45 mm) for a similar exposure class, respectively.
756 Handbook of Precast Segmental Tunnel Lining Systems

12.11.2 Example of using performance-​based service life design for

Green Heart Tunnel
The Green Heart Tunnel is a 4.5 mile (7.2 km) long high-​speed railway bored tunnel between
Amsterdam and Brussels. The tunnel is lined by precast reinforced concrete segments and
each ring consists of nine ordinary segments and one key segment (9 +​1 segmental ring
configuration). Internal diameter of the tunnel is 43.65 feet (13.3 m), the thickness of seg-
ments is 2.0 feet (0.6 m) and the width of segments (ring length) is 79 inches (2.0 m). The
design service life of the tunnel is 100 years. As mentioned in their study, frost attack and
deicing agents, and biological and chemical (e.g. sulfate) attacks were not considered as
relevant degradation mechanisms to this tunnel. Durability against alkali-​silica reaction
(ASR) was mitigated by the selection of proper cement and by preventing use of potential
reactive aggregates. The major durability concern in this project was reinforcement cor-
rosion due to carbonation and chloride ion penetration. Han et al. (2003) identified four
critical locations, which are subject to these two degradation mechanisms.

• First, segment extrados which is constantly exposed to the chloride in groundwater

similar to submerged conditions in seawater.
• Second, location of the segment intrados, which is constantly exposed to the air and
carbonation. Due to possible leakage into segment intrados at the invert, chloride
penetration can also be a concern. This can be treated as the splash zone as far as the
chloride exposure condition is concerned.
• Third, two critical locations are segments circumferential and longitudinal joints
where concrete located outside of gaskets are constantly in contact with the saltwater.
• Fourth, the small area of concrete located behind the gasket toward the inside of tun-
nel could possibly be in contact with the leaking saltwater from the gasket. This can
be considered as a tidal zone as far as exposure condition is concerned.

DuraCrete Service life model (Engelund et al., 2000) was used in Han et al. (2003) study
to evaluate the required service life of 100 years. For service life design when concrete is
exposed to carbonation, a deterioration model based on Fick's first law of diffusion was
adopted, which is presented in Equation (12.6).

a
2.ke .kc .kt .Cs t 
x (t ) = . t. 0  (12.6)
RCarb t

where
x (t ) is the carbonation depth at the exposure time t ;
RCarb is the effective carbonation resistance of concrete;
Cs is concentration of the passivator at the concrete surface;
ke is a constant parameter, which considers the effect of environment on RCarb (e.g. real-
istic moisture history at the concrete surface during use);
kc is a constant parameter, which considers the effect of execution on RCarb (e.g. influ-
ence of curing);
kt is a constant parameter which considers the effect of test method on RCarb ;
t0 is the reference period (e.g. 1 year); and
a is a constant parameter which considers the effect of meso climatic conditions (e.g.
orientation and placing of structure).
 Durability and service life of precast segmental linings 757

For chloride induced corrosion, Equations (12.7) and (12.8) based on Fick’s second law
of diffusion were adopted.

a
t 
x (t ) = 2.k. DRCM ,0 .ke .kc .kt .t.  0  (12.7)
t

 C 
k =  1 − cr  (12.8)
 Cs 

where
x (t ) is the depth of critical chloride concentration at the exposure time of t ;
DRCM,0 is the chloride diffusion coefficient at defined compaction, curing and environ-
mental conditions, measured at time t0 ;
Ccr is the critical chloride concentration;
Cs is the surface chloride concentration;
ke is a constant parameter which considers the effect of environment on DRCM ,0 ;
kc is a constant parameter which considers the effect of execution on DRCM,0 (e.g. influ-
ence of curing);
kt is a constant parameter which considers the effect of test method on DRCM ,0 ;
t0 is the reference period (e.g. 28 days); and
a is the age exponent, which takes into account the effect of age and the environmental
condition on the measured material property.

Equations (12.7) to (12.8) are basically the same as Equations (12.2) to (12.4), only
with a different organization when D0 in Equations (12.2) and (12.4) are replaced by
DRCM ,0 .ke .kc .kt . This is because the DuraCrete Service life model (Engelund et al., 2000) takes
into account the effect of the differing initial chloride diffusion coefficient between jobsite
and the pre-​construction laboratory testing conditions by proposing ke , kc , and kt as mul-
tipliers to D0 .
In this project, the corrosion initiation time was conservatively considered as a service-
ability limit state (SLS) because of the difficulty and the cost impact of repair and mainten-
ance during the service life of the tunnel. Project technical requirements defined a reliability
index β of 1.8 for the probabilistic calculations for tunnel structures with a design service
life of 100 years. As presented in Equation (12.9), reliability-​based calculations were carried
out according to the DuraCrete method (Engelund et al., 2000).

g  x (t ) = d − x (t ) = 0 (12.9)

where
g  x (t ) is the limit state function for carbonation or chloride diffusion;
d is the designed concrete cover; and
x (t ) is the carbonation depth or chloride penetration depth calculated by Equation
(12.6) or (12.7), whichever applies.
758 Handbook of Precast Segmental Tunnel Lining Systems

Considering Ti as the intended design service life, the reliability calculation can be
expressed by Equation (12.10).

Pf (Ti ) = P { g ( x, t )} < Ptargert = Φ ( −β ) (12.10)

where
Pf (Ti ) is the probability of the critical carbonation/​chloride penetration depth exceed-
ing the designed concrete cover over reinforcement within the intended design service
life Ti ;
Ptargert is the accepted maximum value of the failure probability;
Φ is the standard normal distribution function (with mean value of 0 and standard devi-
ation of 1); and
β is the reliability index (e.g. 1.80).

During the design phase, after performing a comprehensive experimental program, an

optimized concrete mix proportion was selected, which is shown in Table 12.18. As a
result, the average chloride diffusion coefficient in the laboratory at 28 days, DRCM ,0 , was
determined as 3.4 ×10−12 m2 /s with a standard deviation of 1.1 ×10−12 m2 /s . Han et al.
(2003) presented assumptions to be considered for the calculation of Equations (12.7) to
(12.9) for the location of segment joints, when the tunnel is exposed to chloride penetra-
tion. As shown in Table 12.19, all variables are stochastic with characteristics such as dis­
tribution type, mean and standard deviations determined either based on the experimental
data or using the DuraCrete database (Engelund et al., 2000). Considering the design
concrete cover of 1.38 inches (35 mm), the probability of a critical chloride penetration
depth exceeding this number, calculated using STRUREL software (RCP 1997), is as low
as 0.03 or 3.3%.
Han et al. (2003), additionally presented variations of the reliability index ( β ) as a func-
tion of exposure time ( t ) when the designed concrete cover is 1.38 inches (35 mm), which
is shown in Figure 12.28. This result indicated that the reinforced concrete mixture designed
for precast segments in this project is durable for the worst exposure condition of chloride
penetration during the expected service life of the tunnel (i.e. 100 years) when the designed
concrete cover is 1.38 inches (35 mm).

Table 12.18 Concrete mix proportions for the Green Heart Tunnel.

Precast concrete segmental tunnel lining mix proportions used in
Green Heart Tunnel

Concrete mix proportions

Item B62.5 precast concrete Metric Imperial
1 CEM III/​A 52.5 LA Obourg 450 kg/​m ³ 760 lbs/​yd 3
2 Water/​c ement ratio 0.39 0.39
3 Aggregate 1,785 kg/​m ³ 3,010 lbs/​y d 3
Source: Han et al., 2003.
 Durability and service life of precast segmental linings 759

Table 12.19 Stochastic variables.

Stochastic variables and their characteristics used for calculation of depth of
critical chloride and reliability function at the location of segment joints

Values
Item Parameters Unit of measure Mean Standard deviation Distribution
1 xc mm 35 5 Lognormal
2 D RCM,0 10 –​12 m²/​s 3.4 1.1 Normal
3 C cr wt.–​% /​ binder 1.0 0.2 Normal
4 N –​ 0.8 0.07 Beta a =​ 0; b =​ 1
5 Kt –​ 1.0 –​ Deterministic
6 Ke –​ 2.704 1.292 Gamma
7 kc –​ 1.8 0.4 Beta a =​ 1; b =​ 4
8 Cs wt.–​% /​ binder 2.7 0.11 Normal
9 t0 year 0.0767 Deterministic
Source: Han et al., 2003.

Figure 12.28 Reliability index for durability of Green Heart Tunnel segment joints.
Note: Reliability index ( ² ) for durability of Green Heart Tunnel segment joints calculated by
Han et al. (2003) as a function of chloride exposure time ( t ) when designed concrete cover is
1.38 inches (35 mm).

12.12 REPAIR OF DEFECTS

The repair or replacement of damaged and misaligned segments is essential to maintain the
structural integrity, durability, and watertightness of segmental lining systems.
Pre-​erection repairs for concrete segments involve addressing minor damage by removing
defective or damaged areas until reaching sound concrete. These areas are then repaired
using appropriate materials and procedures. Segments with excessive cracks, voids, or
defects should be examined to determine the cause. Accepted repairable segments must
 n

760 Handbook of Precast Segmental Tunnel Lining Systems

be classified according to the defect class outlined in Table 12.20, which provides proce­
dures for repairing segment defects, including descriptions of damage types and extents, in
accordance with ACI 224R-​01 and ACI 503.4-​92(03). Note that for certain defect types,
the repair procedure must be provided by the segment manufacturer, contractor, or both,
and reviewed by the designer before approval and implementation. Segments identified for
repair but listed as “segment rejection” in Table 12.20 should be immediately discarded and
not used for installing segmental rings.
For post-​erection repairs, it is important to address all oil stains and unsightly segments
by cleaning and repairing them before erection. Any stained segments should be cleaned
either during or after erection. Cracks in the segments and water infiltration through gaskets
after erection must be repaired using materials and procedures that have been examined
and approved by the designer. Table 12.21 presents procedures for repairing segment and
joint defects, providing descriptions of the type and extent of damage, as well as repair
procedures for segment defects that occur during and after installation by the tunnel boring
machine.
 newgenrtpdf
Table 12.20 Materials and procedures for repairing segment defects

Class of damage/​ defect Description Location Extent Remedy

Class A1 Blow holes and air All locations except Diameter > 0.8 inches (20 mm) or Repair procedure 2
non-​s tructural voids gasket groove, intrados depth > 0.4 inches (10 mm)
patching & caulking groove (if
used)
Class A2 Intrados, caulking Diameter > 0.8 inches (20 mm) or Repair procedure 2
non-​s tructural patching groove & gasket groove depth > 0.2 inches (5 mm)
Class B Gasket groove Diameter > 0.08 inches (2mm) Repair procedure 1
non-​s tructural patching
Class C Spalling Gasket groove edges Area: length > 0.8 inches (20 mm) × Repair procedure 2

Durability and service life of precast segmental linings

non-​s tructural cosmetic depth > 0.2 inches (5 mm)
Class D All locations except as Area > 1.6 inches × 1.6 inches (40 mm Repair procedures 2
non-​s tructural cosmetic noted in C × 40 mm) or depth > 0.6 inches or 3
(15 mm)
Class E1 Local protrusions Non-​f ormed surfaces Height or width > 0.2 inches (5 mm) Stone rub or grind
surface irregularities
Class E2 Joint faces Height or width > 0.2 inches (5 mm) Stone rub or grind
surface irregularities Check mold
Class F localized surface Minor non-​ Gasket groove & Cracks > 0.012 inches (0.3 mm) width Designer review for
cracking & crazing structural local extrados edge on joint to be assessed by examination approval of repair
defects faces procedure
Class G Through segment All locations Reject segment
structural cracks
Class G1 Damage exposing To be assessed by examination Designer review for
structural damage reinforcing approval of repair
procedures 2 or 3
Class I Exposed aggregate/​ Maximum depth 2 inches (50 mm) Repair procedure 3
honeycombing poor consolidation Depth > 2 inches (50 mm) to be
assessed by examination

761
762 Handbook of Precast Segmental Tunnel Lining Systems

Notes for Table 12.20

1 Repair procedure 1
a. Materials
i. Cement type MS
ii. Silica sand
iii. Mix one part cement with 2.5 parts sand, using a water/​cement ratio of 0.4.
b. Procedure
i. Clean and wire brush off all dirt and dust from areas to be filled. Dampen repair
area with water.
ii. Measure and mix cement and sand with water in accordance with the instruc-
tions. Do not add additional water to the mixture after it has already been mixed
to adjust its consistency, which can negatively affect the strength and quality of
the repair.
iii. Fill the repair area, and sack rub the finished surface.
2 Repair procedure 2
a. To be used in areas less than 1.6 inches (40 mm) long and 0.6 inches (15 mm) deep.
b. Material
i. SIKADUR 31 epoxy mortar or equivalent mortar is acceptable.
ii. Mix in a one-​to-​one ratio by volume by adding oven-​dried silica sand until a uni-
form and consistent mix is achieved. Do not mix a quantity larger than can be
used within 30 minutes.
c. Procedure
i. Repair area must be dry. Remove any dust, laitance, grease, oils, or loose materi-
als from the area to be repaired and wire brushed.
ii. Place mixed materials into the void, working the material by trowel or spatula to
ensure bond. Strike off level to existing concrete.
iii. Cure the epoxy mortar at a minimum temperature of 5 degrees Celsius.
iv. Ensure accurate profile by removing any excess mortar by grinding.
3 Repair procedure 3
a. To be used in areas more than 1.6 inches (40 mm) long and 0.6 inches (15 mm) deep.
b. Material
i. Master Builders EMACO S88-​CA or equivalent.
ii. Mix at the rate of 55 lb (25 kg) Bag EMACO with 0.7 gallons (2.7 liters) to 1
gallons (3.8 liters) of water (10.5–​15% by weight).
c. Procedure
i. Saw-​cut all edges of repair to a depth of 0.6 inches (15 mm) with a circular saw.
ii. Break back to sound concrete and remove surplus material by low impact
method. Clean and wire brush off loose particles, dirt and dust from areas to
be filled. Soak burlap or sponge over the repair area to dampen for a period of
2 hours.
iii. Measure and mix the patching compound with water in accordance with the
manufacturer’s instructions. Do not retemper mixture with water.
iv. Fill the repair area, finish open edges with a steel trowel and use temporary form-
work when necessary, ensuring that repair material is thoroughly compacted.
Strike off level to existing concrete.
v. Place a damp cloth over the repair area and leave it damp for a period of 7 days
until the repair material has sufficiently hardened.
 Durability and service life of precast segmental linings 763

Table 12.21 Materials and procedures for repairing segment defects after installation by the
tunnel boring machine

Class of
damage/​ defect Description Location Extent Remedy
Class G2 All types of All locations Crack opening Use Sikadur 55 SLV or
structural cracks in between equivalent. Repair procedure
damage dry or damp 0.012 inches to be submitted for designer’s
conditions (0.30 mm) and approval
0.12 inches
Class G3 All types of (3 mm) Use Sika Injection-​2 16 or
structural cracks in wet equivalent. Repair procedure
damage conditions to be submitted for designer’s
approval
Class G4 All types of Opening > Repair procedure to be
structural cracks 0.12 inches submitted for designer’s
damage (3 mm) approval
Class J Gaskets Water Use Sika Inject-​2 15 or
water infiltration equivalent. Repair procedure
infiltration through according to section 10.7 of
gaskets ACI 533.5R-​2 0

BIBLIOGRAPHY
For preparation of this chapter, the following references were used, which are categorized into two
categories of authority published documents and authoreddocuments.

AUTHORITY PUBLISHED DOCUMENTS

AMERICAN CONCRETE INSTITUTE (ACI)

ACI 201.2R-​16, Guide to Durable Concrete.

ACI 222R-​19, Guide to Protection of Reinforcing Steel in Concrete against Corrosion.
ACI 224R-​01 –​Control of Cracking in Concrete Structures
ACI 318-​19, Building Code Requirements for Structural Concrete and Commentary.
ACI 350-​06, Code Requirements for Environmental Engineering Concrete Structures.
ACI 503.4-​92(03) –​Standard Specifications for Repairing Concrete with Epoxy Mortars
ACI 533.5R-​20, Guide for Precast Concrete Tunnel Segments.
ACI 544.1R-​96(09), Report on Fiber Reinforced Concrete.

AMERICAN SOCIETY FOR TESTING MATERIALS (ASTM) INTERNATIONAL

ASTM C150/​C150M-​18, Standard Specification for Portland Cement.

ASTM C1202-​19, Standard Test Method for Electrical Indication of Concrete’s Ability to Resist
Chloride Ion Penetration.
ASTM C1580, Standard Test Method for Water-​Soluble Sulfate in Soil.
ASTM D516, Standard Test Method for Sulfate Ion in Water.
ASTM D4130, Standard Test Method for Sulfate Ion in Brackish Water, Seawater, and Brines.

AUSTRIAN SOCIETY FOR CONCRETE AND CONSTRUCTION TECHNOLOGY (ÖVBB)

ÖVBB 2011, Guideline for Concrete Segmental Lining Systems.

764 Handbook of Precast Segmental Tunnel Lining Systems

BRITISH STANDARDS INSTITUTION (BSI)

BS PAS 8810:2016, Tunnel Design—​Design of Concrete Segmental Tunnel Linings—​Code of Practice.

BRITISH TUNNELING SOCIETY (BTS) AND THE INSTITUTION OF CIVIL ENGINEERS (ICE)

BTS ICE 2004, Tunnel Lining Design Guide.

CANADIAN STANDARDS ASSOCIATION (CSA)

CSA A23.1-​19, Concrete Materials and Methods of Concrete Construction.

CSA A23.4-​16(21), Precast Concrete—​Materials and Construction.

EUROPEAN STANDARDS/​E UROPEAN NORMS (EN)

EN 206-​1:2013, Concrete—​Specification, Performance, Production and

Conformity.
BS EN 1992-​1-​1:2004, Eurocode 2: Design of Concrete Structures—​Part 1-​1: General Rules and
Rules for Buildings.
EN 50122-​2:2010, Railway Applications—​Fixed Installations—​Electrical Safety, Earthing and the
Return Circuit Part 2: Provisions Against the Effects of Stray Currents Caused by D.C. Traction
Systems.
EN 196-​2, Method of testing cement—​Part 2: Chemical analysis of cement.
EN 13577, Chemical Attack on Concrete—​Determination of Aggressive Carbon Dioxide Content
in Water.

FRENCH TUNNELING AND UNDERGROUND ENGINEERING ASSOCIATION (AFTES)

AFTES:2005, Recommendations for the Design, Production, and Installation of Segmental Ring
Recommendation for the Design, Sizing and Construction of Precast Concrete Segments
Installed at the Rear of a Tunnel Boring Machine (TBM)s.

GERMAN TUNNELING COMMITTEE (DAUB)

DAUB:2013, Lining Segment Design: Recommendations for the Design, Production, and Installation
of Segmental Rings.

INTERNATIONAL FEDERATION FOR STRUCTURAL CONCRETE (FIB)

fib bulletin 34, Model Code for Service Life Design.

INTERNATIONAL ORGANIZATION FOR STANDARDIZATION (ISO)

ISO 4316, Surface Active Agents—​Determination of pH of Aqueous Solutions—​

Potentiometric
Method.
ISO 7150-​1, Water quality—​Determination of Ammonium—​ Part 1: Manual Spectrometric
Method.
ISO 7980, Water Quality—​ Determination of Calcium and Magnesium—​ Atomic Absorption
Spectrometric Method.

PORTLAND CEMENT ASSOCIATION (PCA)

PCA IS536:2002, Types and Causes of Concrete Deterioration.

 Durability and service life of precast segmental linings 765

SINGAPORE LAND TRANSPORT AUTHORITY (LTA)

LTA 2010, Civil Design Criteria for Road and Rail Transit Systems.

SWISS SOCIETY OF ENGINEERS AND ARCHITECTS (SIA)

SIA 262:2003, Concrete Structures.

AUTHORED DOCUMENTS
Abbas, S., 2014, Structural and Durability Performance of Precast Segmental Tunnel Linings, PhD
dissertation, University of Western Ontario, London, Ontario, Canada.
Bakhshi, M., and Nasri, V., 2018a, “Tunnel Segmental Lining Durability,” Proceedings of TT2018-​
TAC/​NASTT-​NW Tunnelling and Trenchless Conference, Edmonton, Alberta, Canada.
Bakhshi, M., and Nasri, V., 2018b, “Durability Design of Segmental Linings for Intended Service
Life of Tunnels,” Proceedings of Sixth International Conference on Durability of Concrete
Structures, Leeds, United Kingdom.
Bertolini, L., Carsana, M., and Pedeferri, P., 2007, “Corrosion Behaviour of Steel in Concrete in the
Presence of Stray Current,” Corrosion Science, Volume 49, No. 3, pp. 1056–​1068. doi: 10.1016/​
j.corsci.2006.05.048.
Brenna, M., Dolara, A., Leva, S., and Zaninelli, D., 2010, “Effects of the DC Stray Currents on
Subway Tunnel Structures Evaluated by FEM Analysis,” Power and Energy Society General
Meeting, 2010 IEEE, Minneapolis, MN, pp. 1–​7.
Di Pace, G., Morales, L.E., Torrent, R., and Bueno, V., 2019, “From Cradle to Maturity,” Concrete
International, Volume 41, No.4, pp. 47–​54.
Dolara, A., Foiadelli, F., and Leva, S., 2012, “Stray Current Effects Mitigation in Subway Tunnels,”
IEEE Transactions on Power Delivery, Volume 27, No.4, pp. 2304–​ 2311. doi: 10.1109/​
TPWRD.2012.2203829.
Edvardsen, C., Müller, S., Nell, W., and Eberli, M., 2017, “Steel Fibre Reinforced Concrete for Tunnel
Lining Segments—​Design, Durability Aspects and Case Studies on Contemporary Projects,”
Proceedings of STUVA Conference 2017 (49 Forschung +​Praxis: STUVA-​ Tagung 2017),
Stuttgart, Germany, pp. 184–​189.
Engelund, S., Edvardsen, C., and Mohr, L., 2000, General Guidelines for Durability Design and
Redesign: DuraCrete—​ Probabilistic Performance Based Durability Design of Concrete
Structures, Contract BRPR-​ CT95-​ 0132, Project BE95-​ 1347, Volume 15, Lyngby,
Denmark, 109 p.
Graeff, A., Pilakoutas, K., Lynsdale, C., and Neocleous, K., 2009, “Corrosion Durability of Recycled
Steel Fibre Reinforced Concrete,” Article No.7, Intersections/​Intersectii, Volume 6, No.4,
pp. 77–​89.
Gulikers, J., 2006, “Considerations on the Reliability of Service Life Predictions Using a Probabilistic
Approach,” Journal of Physics IV, France, Volume 136, pp. 233–​241.
Gulikers, J., 2011, “Practical Implications of Performance Specifications for Durability Design of
Reinforced Concrete Structures,” Proceedings of the fib Workshop on Performance-​ Based
Specifications for Concrete, F. Dehn and H. Beushausen, editors, June 2011, Leipzig, Germany,
pp. 341–​350, 2011.
Han, N., Taibi, Y., and Kooiman, A., 2003, “Performance and Reliability Based Service Life Design
for the Green Hart Tunnel,” Proceedings of the World Tunnelling Congress 2003, Amesterdam,
Netherlands, pp. 12–​17.
Helsing, E., and Mueller, U., 2013, Beständighet av cement och betong i tunnelmiljö (Resistance
to Cement and Concrete in Tunnel Environment), Seminarium Vatten i anläggningsbyggande
(Seminar on Water in Construction), Göteborg, Sweden.
ITA Working Group on Maintenance and Repair of Underground Structures, 1991, “Report on the
Damaging Effects of Water on Tunnels During their Working Life,” Tunnelling and Underground
Space Technology, Volume 6, No.1, pp. 11–​76. doi: 10.1016/​0886-​7798(91)90005-​O.
766 Handbook of Precast Segmental Tunnel Lining Systems

Kemp, R., 2007, Traction Energy Metrics, Rail Safety and Standards Board, Lancaster University,
United Kingdom.
Li, Q., Yu, H., Ma, H., Chen, S.; and Liu, S., 2014, “Test on Durability of Shield Tunnel Concrete
Segment under Coupling Multi-​Factors,” Open Civil Engineering Journal, Volume 8, No.1,
pp. 451–​457. doi: 10.2174/​1874149501408010451.
Li, Q., Li, K., Zhou, X., Zhang, Q., and Fan, Z., 2015, “Model-​Based Durability Design of Concrete
Structures in Hong Kong-​Zhuhai-​Macau Sea Link Project,” Structural Safety, Volume 53,
March 2015, pp. 1–​12.
Life-​365 Service Life Prediction ModelTM and Computer Program for Predicting the Service Life and
Life-​Cycle Cost of Reinforced Concrete Exposed to Chlorides, 2012, Life-​365 Consortium II,
80 p. Washington DC.
Nilsson, L.O., 2012, “Transport Processes in the Microstructure of Concrete and Their Relevance
for Durability,” Second International Conference on Microstructural-​Related Durability of
Cementitious Composites, Amsterdam, The Netherlands, 8 p.
Olek, J., Lu, A., Feng, X., and Magee, B. (2002). “Performance-​related specifications for concrete
bridge superstructures.” Vol. 2, High-​Performance Concrete, Rep. No. FHWA/​INDOT/​JTRP-​
2001/​08-​II, Joint Transportation Research program, Purdue Univ., West Lafayette, Ind.
Rashidi, S., and Nasri, V., 2012, Mitigation of the Corrosion Risk for Large Concrete Sewer Tunnels,
ITA-​AITES World Tunnel Congress (WTC) 2012, Bangkok, Thailand.
RCP, 1997, “STRUREL—​A Structural Reliability Analysis Program System,” RCP Consulting
GmbH, München.
Sigl, O., Raupach, M., and Rieker, L., 2000, “Durability Design of Concrete Tunnel Lining Segments,”
25th Conference on Our World in Concrete and Structures, 23–​24 August 2000, Singapore.
Solgaard, A.O.S., Carsana, M., Geiker, M. R., Küter, A., and Bertolini, L., 2013, “Experimental
Observations of Stray Current Effects on Steel Fibres Embedded in Mortar,” Corrosion Science,
Volume 74, pp. 1–​12. doi: 10.1016/​j.corsci.2013.03.014.
Tang, K., 2017, “Stray Current Induced Corrosion of Steel Fibre Reinforced Concrete,” Cement and
Concrete Research, Volume 100, pp. 445–​456. doi: 10.1016/​j.cemconres.2017.08.004.
Torrent, R., 2013, “Service Life Prediction: Theorecrete, Labcrete and Realcrete Approaches,” SCTM3
Conference, Kyoto, Japan.
Torrent, R., 2017, “The Reference Approach to Service Life Design,” High Tech Concrete: Where
Technology and Engineering Meet, D.A. Hordijk and M. Luković, editors, Springer Cham,
Maastricht, The Netherlands, pp. 2224–​2233.
Tuutti, K., 1982, Corrosion of Steel in Concrete, Research Report No.4.82, CBI (Swedish Cement and
Concrete Research Institute), Stockholm, Sweden, 468 p.
Wang, C., Li, W., Wang, Y., Xu, S., and Fan, M., 2018, “Stray Current Distributing Model in the
Subway System: A Review and Outlook,” International Journal of Electrochemical Science,
Volume 13, pp. 1700–​1727. doi: 10.20964/​2018.02.16.
Chapter 13

Innovative products and applications

Lead author: David Klug
Contributing authors: Jon Kaneshiro, Richard Cubeta,
David Green, and Ralf Winterberg

13.1 CURRENT POSITION OF PRECAST SEGMENTAL TUNNEL LININGS

IN TUNNEL INDUSTRY
Since the mid-​1970’s, precast segmental tunnel linings have continued to gain in market
share for tunnels driven with circular mechanical tunnel excavation machines, first with
open faced digger shields and since the late 1990’s they have been used almost exclusively
with shielded tunnel boring machines (TBMs). The advancement in one pass precast seg-
ment technology has been in parallel with the advancement in TBM technology as segments
were first used in soft ground tunnel with a homogeneous geology. With TBMs now being
engineered to deal with multiple geological conditions in a specific drive, one-​pass precast
segmental tunnel linings have become the support option of choice through various industry
innovations.
Owners and tunnel design engineers are now specifying one-​pass precast segmental tun-
nel linings for hard rock tunnels that in the past would be driven with the multiple pass
method of bored excavation, placement of tunnel waterproofing if required and placement
of a cast-​in-​place final concrete lining. Each separate pass or operation adds additional costs
and time to the schedule and owners/​project financiers want the tunnel to begin generating
revenue at the earliest possible date and this is now more critical with PPP funded infra-
structure projects.
The earlier chapters of this book deal with the various detailed aspects of precast seg-
mental design, production and installation. This chapter reviews the various innovations
that have helped to make precast segmental tunnel linings more adaptive to the tunnel
industry marketplace, with easier, safer and more efficient installation procedures that have
improved overall installation quality.
PSTL have evolved from being a material replacement for cast iron linings in soft ground
tunnel applications only to being the prevalent lining system for all types of tunneling condi-
tions ranging from soft ground to very hard rock tunneling conditions. The main reason for
this transformation has been the many major innovations brought to the tunnel industry by
private enterprise.

DOI: 10.1201/9781032675541-13
768 Handbook of Precast Segmental Tunnel Lining Systems

13.2 MAJOR INNOVATIONS THAT IMPACTED PRECAST SEGMENTAL

TUNNEL LINING MARKET WORLDWIDE

13.2.1 Innovations in TBM technology

One of the major industry innovations has been the advancements made in TBM tech-
nology. TBMs can now mine variable ground conditions in the same tunnel reach without
making major TBM modifications. This capability is very accommodating to PSTL as one
lining can be designed and used for the full length of the tunnel. Standardization brings effi-
ciency to tunnel operations.

13.2.2 TBM design improvements

As stated earlier in this book and this chapter and in Chapter 1, the first use of precast seg­
mental tunnel lining systems was with open faced digger shields. With innovative cutter
head design TBMs can now mine through varying geological conditions using one TBM.
Design changes may have been made to the cutting tools and cutting heads to address
changes in geological conditions in the same drive. This machine versatility of design has
not only lowered the cost of many projects but has made many projects feasible to construct
at a desired tunnel elevation even with changing geological conditions.

13.2.3 TBM guidance systems

A major improvement in TBM technology has been the introduction of electronic guid-
ance and control systems installed on TBMs to maintain proper line and grade and curve
execution. The old system of a single laser point required the TBM operator to control the
curve by manually moving the shove jacks controls to control the TBM. Frequently TBMs
became “iron-​bound” meaning the lining pinched against one side of the shield preventing
any adjustments in line and grade. Frequently PSTL were shimmed with wood or plastic
shims to negotiate a curve, adding shims as required.

13.2.4 TBM segment erector technology

Originally, segments were handled in the tunnel by lifting them on a single steel ball screwed
into the center of each segment as shown in Figure 13.1. This frequently caused damage to
the segments during the lifting and placement of the segment as all the lifting stresses are
transferred through a single point in the middle of the segment. Refer to Figure 13.2.

13.2.5 Vacuum lifting for segment handling

A major innovation that has helped to minimize these handling issues has been the introduc-
tion of the vacuum lifting system whereby vacuum pads are used to stabilize the segments
during the lifting process, this stabilizing the lifting load across the greater segment face
area, thus decreasing segment damage during the lifting and placement operations. Along
with the introduction of vacuum lifting system came a dramatic improvement in the design
of the TBM segment lifting arm handling systems. With innovations in hydraulic control
systems using microprocessors to control the lifting arm hydraulic control systems workers
are now able to better control the orientation of each segment during the installation
 Innovative products and applications 769

Figure 13.1 Segment centroid ball lifting figure system in TBM.

Note: Steel lifting ball being used to move the segment within the trailing gear of the TBM so
that it can be in position to be connected to the segment erector.

process thus better enabling the TBM team to build “round rings”. There is an old saying
within the tunnel segment industry “a round ring solves many problems”.

13.3 INNOVATIONS IN SEGMENT CONNECTOR COMPONENTRY

SYSTEMS
In the early development of PSTL steel bolts were used to connect the concrete segments as
steel bolts were used to connect the steel liners. Frequently the ring build was out of toler-
ance and water leakage occurred and segments would crack if eccentric forces were put on
the lining from the shove jacks during the TBM advancement. In the 1980’s new advance-
ments such as dowel connectors for the circumferential face connection that sped up ring
erection, provided better segment-​to-​segment alignment and eliminated packets in the lining
that improved lining hydraulic efficiency. Other innovation occurred such as guide rods
for radial joint alignment and improved grout lifting insert assemblies that assisted in bet-
ter handling and in the proper placement of backfill grout. These and other innovations
reviewed in this section have contributed to more efficient and higher-​quality ring build.

13.3.1 Joint connection innovations

From the inception of PSTL design in the 1960’s (refer to Chapter 1) wood dowels, steel
bolts, straight and curved have been used to connect diameter in the radial joint faces
770 Handbook of Precast Segmental Tunnel Lining Systems

Figure 13.2 Segment vacuum lifting system in TBM.

Note: A typical vacuum lifting device connected to the TBM erector arm that will be used to
place the segment in its proper location in the ring build.

[segment-​to-​segment and circumferential joint faces (ring-​to-​ring)]. Wood pins were the
first dowels, they were introduced in the 1960’s as ring-​to-​ring alignment devices for pri-
mary segments. These were cones tapered on each end that were installed in tapered holes
cast into the face of each segment. When the dowel engaged it helped to align each segment
and when the TBM propel jacks pushed each segment home the friction force between the
dowel and the wall of the cast-​in hole created a resistance restricting segment movement
when the shove jacks propelled the digger shield forward mining for the next ring. Wood
dowels were used on segments without gaskets as the wood dowels could not keep gasket
seals compressed.

13.3.2 Wood dowel connections

A “Conex” wood dowel system was used in the 1990’s on the one-​pass diameter for the
MWRA Effluent Outfall Tunnel in Boston with a six-​piece trapezoidal ring 7.78 m (25.5
foot) outside diameter. 7.70 m (25.25 foot) inside diameter. There were two Conex dowels
per segment. The PSTL did not have a sealing gasket as part of the lining design. This was
the last time this wood dowel system was used in North America (Figure 13.3).
Bolts are a very positive connection once tightened to the proper torque, for the early
“waffle” segments straight bolts worked fine as they were simple to use (see Figure 13.4).
 Innovative products and applications 771

Figure 13.3 C onex wood dowels –​ MWRA Effluent Outfall Tunnel Project –​ Photo Courtesy
Sehulster Tunnels.
Note: Conex wood dowels installed in the circumferential joint of the test ring at the cast-
ing plant.

Figure 13.4 Waffle segment bolts being installed.

Note: Tunnel worker tightening the steel bolts in the circumferential face of a precast segment
in a project in Milwaukee, WI in the 1980’s. Note that the segment joints are not staggered as
is standard industry practice today.
772 Handbook of Precast Segmental Tunnel Lining Systems

13.3.3 Steel bolt connectors

When the industry evolved to the use of solid concrete segments with a smooth intrados face
the industry required a different connection system and went to straight bolts (commonly
called spear bolts [Figure 13.5] or curved bolts [Figure 13.6]) for joint connections.
One problem with steel bolts is that if the segments are not properly aligned during
the ring erection sequence the bolts would not properly engage in the cast-​in anchoring
sockets. Curved bolts have their own installation issues. Frequently workers would “ham-
mer” in the steel bolts in an attempt to get them engaged into the embedded plastic socket.
This frequently damaged the sockets and the concrete around the socket thus allowing for
external water leaks through the bolt sockets area and prevent the bolts from performing
as designed.
There was a confirmed incident on a subway project whereby an inspector standing on a
catwalk in the tunnel during the mining operation was injured when a circumferential joint
connecting bolt approximately 355 × 25 mm (14 × 1.0 inch) diameter came loose from the
connecting joint location due to vibration from the TBM and fell-​out and hit the inspector.
Investigations discovered that workers could not get the bolt installed due to the ring being
out of round thus preventing proper bolt engagement. The workers hammered the bolt into
the socket destroying the socket threads and thus preventing proper bolt engagement then
allowing the bolt to migrate loose and fall injuring the inspector.
Throughout the world it is a common practice to remove the steel bolts after ring erection
and set-​up of the backfill grout as bolts are assumed to be a ring assembly aid. Some

Figure 13.5 Typical spear bolt connection.

Figure 13.6 Typical U-​b olt connection.

 Innovative products and applications 773

high-​speed rail systems and highway tunnels; i.e., Port of Miami Tunnel, Alaskan Way
Tunnel remove the bolts because over time even properly installed bolts can vibrate loose
and then become a potential safety hazard to vehicles traveling through the tunnel. This
practice varies by standard industry practice in a particular country and the position of the
design engineer regarding final lining design requirements.

13.3.4 Fastlock steel dowel system

In the 1980’s Sehulster Tunnels introduced many innovations in the upper Midwest tunnel
market of the United States inclusive of one-​pass precast segmental tunnel linings for sewer
tunnel applications. Refer to Chapter 1, for details. To facilitate the smooth bore one pass
lining Sehulster developed the “fast lock” dowel system for circumferential joint connection
(Figure 13.7). Refer to Chapter 1. This connection innovation changed the industry as it
aligned the segments into a circular ring and locked the segments into its final location with
the TBM propel jacks. The locking characteristics kept the gaskets compressed thus creating
a more watertight tunnel.
The fast lock system was licensed to various international tunnel lining manufacturers
and used on projects throughout the world. From the steel locking dowel system came the
next development of plastic friction lock dowels that anchored into unlined sockets cast
into the segment circumferential joint faces. This development came from manufacturers
in Austria, Italy and France. The plastic dowels had a center core of steel or hard plastic
encased with a softer more flexible outer plastic shell that gripped on the concrete walls of
the unlined socket holes.

13.3.5 Connecting dowels and bicone innovations

When segments progressed to their current concept of mass concrete with a relatively
smooth inside intrados face, this smooth face facilitated better hydraulics for water and

Figure 13.7 Sehulster fast lock dowel system.

Note: This innovation was developed and patented in the United States and was sold throughout
the world as it reduced segment erection time and improved ring build, especially for tunnels
under 4.5 m (15.0 feet) in diameter.
774 Handbook of Precast Segmental Tunnel Lining Systems

sewer tunnel and more efficient air flow for transportation tunnels. The international tunnel
industry wanted a circumferential joint connection system that met the following criteria:

• Make a positive connection ring-​to-​ring

• Kept the gaskets compressed when the TBM shove jacks were retracted
• Provided shear resistance from ring-​to-​ring movement
• Provided corrosion resistance for a life span up to 100 years for transit tunnels
• Provided for proper segment alignment during ring assembly operations
• Was simple to understand and install in the tunnel on a production basis
• Improve ring erection time over a conventional bolted system

With the new concept, designers and innovative precasters began to develop better circum-
ferential joint connection systems that would meet the above market objectives.

13.3.6 Solid plastic friction lock dowel system

The international development of the solid plastic dowel that had fins that were engineered
to grip the sides of a hole cast into the circumferential face of the segment. They were suc-
cessfully marketed by component manufacturers in Austria, France and Italy. The size of
the plastic dowels was typically 70 mm (2.75 inches) diameter with a length of 270 mm
(10.6 inches) with a pull-​out rating of 40 kN (10,000 lbs) as shown in Figure 13.8. This
product met some of the above listed industry requirement, it did prove that the dowel
concept would work but further developments were required to meet international tunnel
industry requirements.

13.3.7 Anchored plastic dowel with steel center axis

From the success of the friction dowel came the next industry innovation as the industry
was needing a dowel system that would provide up to 100 kN (22,500 lbs) of pull-​out
resistance and 100 kN (22,500 lbs) of shear resistance to meet the design needs of upcoming
major transportation tunnels. This objective was met by using a dowel system whereby the
centering dowel was mechanically anchored into plastic female sockets that were cast into
the segments that would engage the fins on the dowels. To provide greater shear resistance

Figure 13.8 Sof-​C lip friction dowel.

Note: This dowel does not have female sockets embedded into the concrete into which to
anchor the dowel. It was a friction connection between the plastic Sof-​C lip dowel and the con-
crete walls of a hole cast into the circumferential face of the segment. This dowel provides excel-
lent ring-​t o-​r ing alignment characteristics.
 Innovative products and applications 775

Figure 13.9 Typical symmetrical dowel with steel center core.

Note: The need developed in the industry for a dowel that would withstand greater shear
forces at the connection thus the above dowel was developed with a steel axis center core. The
embedded female sleeve helped to create greater pull-​o ut strength and thus a greater overall
performance.

Figure 13.10 Sof-​F ix anix dowel system –​ symmetrical ends.

Note: If the segments are required to be rotated for line or grade adjustment a symmetrical
dowel is required, as the dowel are installed in the tunnel on the trailing face of the last installed
segment.

the dowel can be manufactured with a high-​strength steel axis that is completely embedded
in the dowel during the injection molding process, thus providing a high-​strength dowel
that will not corrode as shown in Figure 13.9.
In early version of this concept, the dowel was symmetrical on both ends, as illustrated in
Figures 13.9 and 13.10. Workers had to seat the dowels in the previously installed ring by
“hammering” the dowel to where it would remain straight for engagement to the next ring.
The force of the TBM shove jacks will close the ring and then the dowels would be fully
anchored. The next development in the product development process has been to remove
the “hammering” operation by the tunnel workers to initially seat the dowel thus greatly
simplifying and standardizing ring assembly in the tunnel.
This was achieved by having a threaded end on one end of the dowel as shown in
Figure 13.11. The workers then can screw the dowel’s threaded end into a plastic socket that
is cast into the circumferential face of the previously installed ring. This can be done using an
air wrench and socket thus improving on installation time and ring build quality as the dowel
now cannot be easily damaged. This has proven to be a very effective industry innovation.

13.3.8 Bicones and center cones for circumferential joint applications

The industry developed a need for a device that could be used with steel bolts to help align
the segment for proper bolt engagement by helping to keep the ring round before the com-
pletion of the back filling operation and to provide additional long-​term increased shear
resistance in the ring-​to-​ring joint orientation after bolt removal.
776 Handbook of Precast Segmental Tunnel Lining Systems

Figure 13.11 Sof-​F ast dowel system –​ threaded end.

Note: Installing a dowel can be a time-​c onsuming process in the tunnel. A newer innovation used
with universal tapered segments on both sides for line an grade adjustments is to use a dowel
that is installed in the tunnel with threads into an embedded threaded socket that can be tight-
ened mechanically thus saving time and minimizing installation errors and dowel damage.

Figure 13.12 Bicones and centering cones.

Note: Bicones and centering cones increase shear resistance and assist in ring erection time.

The industry developed bicones that are American football shaped plastic cones that are
available in different diameters, lengths, some have additional center core reinforcement of
a hard plastic or a steel axis to provide increased shear resistance as shown in Figure 13.12.
These devices are placed in block-​outs cast into the circumferential joint faces of the segment
and are installed in the ring at the time of erection in the tunnel. For projects where the
steel bolts are removed after ring assembly, bicones are typically used to provide long-​term
shear resistance for the circumferential joint connection, project example Port of Miami
Tunnel, Miami, Florida and Alaskan Way Tunnel, Seattle, Washington in the United States.

13.3.9 Innovations in radial joint connecting systems

The international tunnel market has been experimenting for many years in the development
of a mechanical connection system for the radial or longitudinal joints for precast segmental
 Innovative products and applications 777

Figure 13.13 Rad-​link radial joint connection system.

Note: Mechanical radial joint connectors are a relatively new innovation as designers and con-
tractors are looking for ways to eliminate the bolting typically associated with radial joint con-
nection, these connectors also eliminate bolt pockets required for bolted joints. To date, the
Rad-​L ink radial joint mechanical connection system have been successfully implemented on tun-
nel projects in Japan and Europe.

tunnel lining system that provides a smooth intrados surface (without pockets). Pockets in
final lining segments create waster turbulence in water and sewer tunnels and wind turbu-
lence in ventilated transportation tunnels. A common practice now is to use guide rods, as
noted in Section 13.2.11, then remove the bolts after the backfill grout has set. Based on
certain geological conditions designers will prefer that the radial joints have a permanent
structural connection.
In the 2000’s, the Japanese developed and experimented with a steel radial joint con-
nection system that was used on some domestic projects but failed to gain any acceptance
in the international tunnel construction industry. Other systems have been attempted that
connected reinforcing cages through small pockets in the lining but tended to be project
specific experiments.
In the 2010’s, the French tunnel contractors gave serious consideration to the need to
develop a mechanical connection system for the radial joints that would be structurally
sound, non-​corrosive and simple to install. Optimas Solutions, a French based segment
componentry supplier developed a system that met the criteria of the tunnel contractors.
The system consists of two main components. A plastic male tapered dowel that during ring
assembly enters a mating plastic cone. When the TBM shove jacks push the segment home
the tapered cone properly aligns the radial joint as it enters the female cone. Once in place
and the circumferential connectors are engaged the radial joint is permanently fixed and the
segmental lining has “pocket free” intrados face. Refer to Figure 13.13 that shows details
of the connection concept.

13.3.10 Bicones with tie-​rods

To keep segment rings compressed during a tunnel start-​up or to strengthen ring-​to-​ring
connections in cross-​passage break-​out areas and in seismic application the industry devel-
oped bicones with tie-​rods as illustrated below in Figure 13.14.
778 Handbook of Precast Segmental Tunnel Lining Systems

Figure 13.14 Bicones with tie-​r ods –​ typical installation arrangement.

Note: The tie-​rods tighten the bicones in a cast-​in pocket thus aligning the segments while
increasing shear resistance and strengthening the lining to forces caused by seismic events.
Additional reinforcement may be required for this system.

These are standard high-​performance bicones whereby a steel sleeve with an internal thread
that replaces the standard solid steel axis. Steel tie-​rods with threaded ends, are installed
through tubes cast in the segment, connect into the bicones. When tightened during segment
installation to a predetermined torque, force pulls the recently assembled ring segments
towards the previously installed ring creating a very high-​ tension force thus keeping
the gasket compressed. In seismic situations the bicone helps to increase the survivability
rate of the segmental tunnel lining. The system has been incorporated into the Los Angeles,
California Metro Purple Line, Sections 2 and 3 in the cross-​passage break-​out areas. The
system eliminates the building of a steel framework commonly called “hamster cages” in
the tunnel.

13.3.11 Guide rods for radial joint alignment

To obtain proper gasket performance both segment faces, the circumferential and radial
(longitudinal) faces, must align properly during the segment assembly operation in the tun-
nel. Getting the radial joint faces to properly align was always an industry problem but was
solved by an innovative solution, guide rods as shown in Figures 13.15 and 13.16.
These are plastic rods ranging in diameter from 30 mm to 80.0 mm (1.20 to 3.15 inch)
diameter. The rods are fixed to a groove cast into the radial joint face of the segment.
Embedding one half of the plastic rod, during segment ring assembly, the adjoining seg-
ment radial joint is abutted to the segment with the guide rod installed. With the two seg-
ment joints sharing a common guide rod the radial joints become aligned facilitating bolt
installation and faster ring build time. Once the radial joint bolts are tightened the shear
resistance provided by the guiding rods improves ring construction tolerance and joint shear
resistance.
First design guiding rods were typically glued in at the segment casting plant, but a new
design innovation allows for the guiding rods to be installed without gluing by using a
 Innovative products and applications 779

Figure 13.15 Guide rod on radial joint.

Note: Typical installation for the guide rods after casting. This system eliminates the need for
gluing using the two holes for connecting the plastic dowel with the two pins as shown in
Figure 13.16.

Figure 13.16 Anchored guide rod system.

Note: The above pins are cast into the segment thus eliminating gluing, a substantial cost savings.

special embedded stud system. The new design eliminates the need to have to glue the
plastic rods in casting plant. Throughout the world gluing operations using a hazardous
material now requires special plant ventilation isolation systems for worker protection. The
embedded guide rod design creates a positive connection to the segment that will withstand
extended periods of outside storage even in cold weather climates.

13.4 INNOVATIONS IN PRECAST SEGMENTAL TUNNEL LINING

SEALING SYSTEMS
In early segment design sealing gaskets were not used as wood dowels were used in the cir-
cumferential joint faces, the dowel had minimal pullout resistance thus they could not keep
a gasket compressed when the shove jacks on the digger shield were retracted. When the
industry evolved into the “waffle” segments in the 1970’s designers started to add sealing
780 Handbook of Precast Segmental Tunnel Lining Systems

systems to segment faces which was an attempt to seal the segmental lining. The primary
purpose of the seals was to seal the lining to prevent backfill grout from entering into the
working area of the tunnel. Permanent waterproofing was to be achieved by a final lining
steel or concrete pipe (water and sewer tunnels) or a cast-​in-​place concrete lining (transpor-
tation tunnels).

13.4.1 Hydrophilic (swelling) seals

Sealing systems for precast segmental tunnel lining systems originated in two different loca-
tions with two different designs. In Japan they developed what is called the hydrophilic
system that was designed to expand when the hydrophilic material came into contact with
groundwater. When the material expansion occurred between to fixed concrete faces a seal
was created that prevented groundwater from entering tunnel. The hydrophilic material
was manufactured and supplied in strips, approximately 25 mm (1.0 inches) in width
and was glued to the precast segment. There gasket grooves may not be required to be cast
into the segment faces with this system based on project design as shown in Figure 13.17.
The system had limitations in that it could not typically restrain high external water pres-
sures, >10 bar (>145 psi) and the hydrophilic performance could be reduced if the ground-
water contained certain contaminants, i.e. hydrocarbons or sea water. The hydrophilic
system has been successfully used to seal many tunnels worldwide and is still used today.
In the European and North American markets, hydrophilic circular strips are frequently
combined with polymer rubber seals to create a composite sealing system.

13.4.2 Polymer rubber seals

The British tunnel industry was the leader in the development of the polymer rubber tunnel
segment sealing system as they were developing new tunnel lining concepts to be used in
conjunction with mechanical tunnel excavation. The tunnel construction companies were
requiring that the precast segment enter the tunnel to be erected with minimum worker
involvement. British designers developed the concept of attaching an extruded rubber pro-
file to the precast element in what is described, even today, as the “door frame system”,
meaning that gasket profiles re applied to all four sides of the precast segment into prede-
termined gasket grooves that are cast into each segment. The early profiles were simple
extrusions designed to seal when adjacent profiles were compressed together during ring
assembly underground. Please refer to Figure 13.18.
The early gasket profiles were typically manufactured from neoprene or other polymer-​
based product that were deemed to be resistant to hydrocarbons, other external criteria

Figure 13.17 Typical hydrophilic seal.

Note: Hydrophilic gaskets can be installed on a flat concrete face or in a gasket groove.
 Innovative products and applications 781

Figure 13.18 Early style polymer rubber sealing gasket profiles.

Note: Early gasket profiles did not have openings in their design to control deformation forces.

impact the selection of the gasket material. Over time it was determined that the best
material to be used for tunnel gaskets was EPDM (ethylene propylene diene monomer rub-
ber) is a type of synthetic rubber that is used in many industrial applications. Tunnel seg-
ments may have to be stored outside for extended periods of time if a project is delayed
plus owners were requiring a design life of 100 years, i.e. Channel Tunnel located between
France and England, it has been determined by the industry that EBDM best meets these
performance requirements and thus EPDM has become the standard polymer rubber com-
pound used in the tunnel industry.
Neoprene or other materials are still used in projects where high concentrations of hydro-
carbons will be encountered but special precautions are needed for the outside storage of
the segments or the gasket installation must occur at the project site just before the segments
are taken into the tunnel.

13.4.3 Innovations in sealing gasket profile design

Early gasket profile design consisted of wide relatively low-​profile design extruded from a
relatively hard rubber compound. The designers preferred a wide gasket profile of up to
50 mm (2.0 inch) as they wanted to insure gasket engagement as installation tolerances
commonly exceeded 25 mm (1.0 inch) in the ring build diameter in standard 6.0 m (20.0
foot) diameter subway tunnels. With the introduction of improved connection systems
and improved TBM erector technology, as described above ring build tolerances began to
improve thus allowing gasket manufacturers to begin offering gasket profile designs that
were more efficient and effective in sealing. The standard gasket width today for a 6.0 m
(20.0 feet) diameter subway tunnel is 26.0 mm (1.02 inch) designed to seal with a 15 mm
(0.60 inch) offset and a 5.0 mm (0.12 inch) gap This is only possible with improved ring
assembly technology as discussed previously in this chapter.
Gasket manufacturers, with the use of computer aided design have developed gasket
profile designs that provide higher sealing capabilities while exerting a lower reaction load
on the concrete groove area thus helping to minimize concrete failure in this most critical
area as shown in Figure 13.19. These same design innovations have also been applied to
improve gasket corner design whereby the corner is softer thus exerting lower forces in
the corner area while maintaining a tight seal. Leaking corners are typically the limiting
performance criteria when gaskets are tested in a steel fixture for their performance rating
certification.

13.4.4 Innovation in sealing gasket attachment procedures –​anchored

gaskets
In the past, the typical method for attaching gaskets to the precast segment has been to
glue the rubber gasket to the concrete segment after the segment had been removed from
the post casting curing area. This procedure required a secondary operation regardless if
782 Handbook of Precast Segmental Tunnel Lining Systems

Figure 13.19 Modern gasket profiles with molded corner.

Note: Modern gasket profiles are designed with engineered cavities and rubber compounds that
control gasket deformation and the reaction forces that seal the tunnel lining more efficiently.
These profiles must be glued into a gasket groove cast into the precast segment.

Figure 13.20 Modern anchored gasket profiles.

Note: Anchored gaskets do not require gluing that has become a casting plant environmental
issue in many locations throughout the world, the feet anchor into the concrete thus the gaskets
are attached to the segment when demolded offering overall cost savings.

the segments were manufactured on a carousel or stationary casting system. A more recent
industry innovation has been the introduction of “anchored gaskets”. This gasket design
has feet at the base of the gasket that are designed to protrude into the concrete matrix at
the time of casting thus when demolded the gasket will be “anchored” to the precast seg-
ment as shown in Figure 13.20. The gasket is placed in the casting form prior to casting, it is
anchored to the mold with fins on the gasket profile that are anchored into grooves
machined into the steel mold side flanges. The gaskets will release from the side molds when
the segment is demolded. The gasket is now permanently anchored to the segment and can
now be sent to the storge yard.

13.5 INNOVATIONS IN PRECAST SEGMENTAL TUNNEL LINING

REINFORCEMENT SYSTEMS
Various international companies have recognized the need for precast segmental tunnel lin-
ing reinforcement products that improve the long-​term performance of segmental tunnel
linings, which in many cases require a design life of over 100 years and also the quality of
the final products.

13.5.1 High strength welded wire fabric (WWF)

Most design codes for transportation tunnels require standard rebar reinforcement. The use
of high strength welded wire fabric products in accordance with ASTM A-​496 Grade 80
(550 MPa) become an industry standard. The material is produced in flat sheets and bent
 Innovative products and applications 783

Figure 13.21 Fabricated high-​s trength WWF cage.

Note: Ready for placement into casting forms.

Figure 13.22 Typical steel fiber for segments.

Note: These fibers are directly dispensed into the concrete mix.

to the proper intrados and extrados radii before being fabricated into cage assemblies that
are then placed in the steel casing molds at the casting facility. Please refer to Figure 13.21.
The cages are fabricated to very tight dimensional tolerances to meet tight reinforcement
placement tolerances as specified by the lining designer.

13.5.2 Steel fiber reinforcement

Steel fibers were introduced in the 1970’s and have become a standard reinforcement product
for PSTL systems. Please refer to Figure 13.22. Details on steel fiber design and application
are presented in Chapter 5. They have contributed in the following ways:

• Fibers provide reinforcement throughout the concrete matrix.

• gasket grooves now have reinforcement
• improved strength around grout port areas
• improved pullout resistance if lifting ball used
• improved strength around connection area
784 Handbook of Precast Segmental Tunnel Lining Systems

• no special reinforcement coil required

• Fibers eliminate cage manufacturing
• provides overall lower manufacturing cost
• eliminates the need for specialist workers –​welders
• Fibers can be easily dispersed into concrete mix
• with modern dispersing equipment fiber dosing is easily measured and controlled.

13.5.3 Macro synthetic fiber reinforcement –​single component plastic

Macro synthetic fiber (MSF) is an innovative construction material first introduced com-
mercially in the 1990’s and has now been used in hundreds of roads, rail and utility tunnel
linings, including more recently in precast segmental tunnel linings. Please refer to Figures
13.23 ad 13.24.

Figure 13.23 BarChip macro synthetic fiber (MSF).

Note: Before dispensed into the concrete mix.

Figure 13.24 Fiber reinforced segments –​ Blacksnake Creek CSO Tunnel.

Note: Finished product ready to be shipped to the tunnel site.
 Innovative products and applications 785

For durability reasons these polymer fibers are predominantly made of polypropylene,
having a nominal length of typically 50 to 65 mm (2.0 to 2.6 inches) and a tensile strength
ranging from 500 to 700 MPa (72,500 to 101,500 psi). The fibers are manufactured in con-
formity to ASTM C1116 and DIN EN 14889-​2. The key advantages are that they provide a
more uniform reinforcement matrix as compared to traditional steel bars, contributing to a
reduction in cracking and spalling during transient and permanent load stages of the tunnel
segments. Furthermore, MSF is non-​corrosive, a property that has driven its use in many
applications where durability is a concern. The non-​corrosive property may also allow for a
reduction of concrete cover required solely for serviceability limit states, such as corrosion
prevention, which in return reduces material usage. Further environmental benefits include
reduced carbon footprint.
Extensive full-​scale segment testing has been conducted by BarChip at the University of
Brescia, Italy, which demonstrated MSF reinforced concrete can meet the requirements of
MC 2010 for flexural requirements, minimum shear reinforcement, splitting and spalling
requirements. For test details refer to the “Performance of Tunnel Segments Reinforced
with Synthetic Macro Fibers” by Ralf Winterberg.
To date BarChip MSF has been used in multiple segmental tunnel linings, including the
Harefield to Southall Gas Transfer Tunnel, UK (2009), Pista Nueva Malaga Rail Tunnel,
Spain (2009), Santoña-​ Laredo Central Interceptor Collector, Spain (2016), and the
Blacksnake Creek CSO Tunnel, USA (2020). In North America the use of MSF is currently
limited to sewer tunnel applications as some designers are concerned about the long-​term
creep “elongation” if applied to transportation tunnels. The industry is addressing these
issues and for sewer tunnels under compression this is not a major concern.
BarChip MSF fibers were successfully implemented into the precast segmental lining for
the Blacksnake Creek CSO Tunnel in St. Joseph, MO, United States. The sewer tunnel
project consisted of 2.0 km (1.25 miles) of a six-​piece universal ring, 2.74 m (9.0 foot)
internal diameter, 1.2 m (4.0 foot) segment width with a 12.7 mm (0.5 inch) ring taper.
Dowels were used in the circumferential joint connection with steel bolts used in the radial
joint connection. A steel ball lifting device was used for segment erection in the TBM. As
discussed earlier in this chapter this ball lifting method can induce increased stresses in the
segment during handling. At the utilized dosage rate of 7.0 kg/​m3 (12 lbs/​yd3) the MSF
system worked fine.
The tunnel drive using an EPB TBM was initiated in soft ground (23%), then transitioned
into rock (63%), then to a mixed face condition (14%). The segment design using the MSF
system performed well in all three geological conditions. This was the first application of
macro synthetic fibers in a full-​scale tunnel project in North America and the project was
both a technical and commercial success.

13.5.4 Synthetic fiber reinforcement–​bi-​component polymer

The issue of the potential for long-​term creep has been addressed by the Swiss fiber manufac-
turer Contec Fiber, AG who developed a macro synthetic fiber that is a unique bi-​component
polymer. The fiber consists of a high E-​modulus core for high strength with a structured
polymer shell designed for concrete bonding as illustrated below in Figure 13.25. The bi-​
component fiber has a tensile strength of 600 N/​mm2 (87 ksi). The company has confirmed
its resistance to long-​term creep by institutional testing. The manufacturer offers a life-​cycle-​
assessment and is certified based on ISO 14025–​Environmental Labels and Declarations –​
Type III Environmental Declarations –​Principles and Procedures and European standard
786 Handbook of Precast Segmental Tunnel Lining Systems

Figure 13.25 Contec bi-​c omponent polymer fiber.

Note: The two-​c omponent fiber, inner component for tensile strength and outer component for
bonding to the concrete, makes the fiber suitable for final lining applications.

Figure 13.26a Polypropylene fiber –​ typical matrix.

EN 14889-​2–​Fibres for Concrete–​Part 2: Polymer Fibres–​Definitions, Specifications and

Conformity.

13.5.5 Synthetic fiber for precast segmental tunnel lining

fire resistance
A very important tunnel industry innovation has been the introduction of small synthetic
polypropylene fiber to the precast segmental tunnel lining concrete designed to minimize
segment damage in event of a fire inside a transportation tunnel. In the 1990’s it was dis-
covered that high-​intensity fires will cause precast segmental lining to spall concrete as the
heat causes the moisture in the segments to turn to steam and expand thus causing the con-
crete spall.
The introduction of small, average 6 mm (0.23 inch) long, micro synthetic fiber filaments
to the concrete mix created a more durable precast segmental tunnel lining in the event of
a fire in a transportation tunnel. Refer to Figure 13.26. This was confirmed by full scale
testing at various laboratories in Europe thus this has become an international standard for
transportation tunnels. Refer to Figure 13.26b below for test verification of concept.
 Innovative products and applications 787

Figure 13.26b Polypropylene fiber showing fire resistance benefits.

13.5.6 Composite reinforcement design

A common trend in Europe and North America is to design precast segmental tunnel lining
with a composite reinforcement system. This is when steel rebar is used in conjunction with
steel fibers thus creating a composite design. Typically, a lighter weight rebar cage is used as
this provide stability for segment handling both during the demolding and segment hand-
ling and installation in the tunnel. The steel fiber (can be substituted with macro synthetic
fibers) provide the benefits of less cracking and damage due to the concrete having a fully
reinforced matrix. This is discussed in Chapter 3 and Chapter 5. This design was success­­
fully implemented on the recently completed San Francisco Municipal Transit Authority
(SFMTA) Central Subway Tunnel Project in San Francisco, California.

13.5.7 GFRP cage for segment reinforcement

Glass fiber-​reinforced polymer (GFRP) rebars in concrete structure reinforcement is an
innovative solution to specific tunnel industry issues where an alternative to traditional
steel rebar reinforcement is required. Please refer to Figure 13.27. When compared to steel
GFRP rebar offers higher tensile capacity, a lower elastic modulus and lower weight than
steel rebar. GFRP rebar will not corrode thus providing an excellent reinforcement in highly
corrosive PSTL tunnel applications. With its non-​corrosive characteristics, a thinner lining
can be designed as the concrete cover over the bar can be reduced.
On the recent TMCLK Tunnel in Hong Kong that was mined with at a 17.6 m (57.75
foot) diameter under Honk Kong Harbor. Because of the tunnel diameter and geological
conditions, it was determined that the best manner to construct the cross-​passages was to
mine them with a slurry MTBM supported with concrete pipes, the cross-​passage tunnel
was mined to 3.65 m (12 feet) diameter. For the TBM entry portal and exit location in the
parallel tunnel the running tunnel segments were cast with GFRP rebar reinforcement. The
GFRP bars were pre-​bent where required, assembled into cages and replaced the standard
steel cages on the segment production line. The system was very innovative in its overall
design and worked as planned, the cross passage tunnels were driven safely and efficiently.
Another characteristic of GFRP rebar is that it can be cut without damaging TBM
cutterhead cutting tools. GFRP rebar is frequently used to create “soft eyes” in TBM
788 Handbook of Precast Segmental Tunnel Lining Systems

Figure 13.27 GFRP rebar as segment reinforcement.

Note: GFRP reinforcement can be manufactured in different shapes and profiles to create prefab-
ricated cages, as shown above, for placement into steel casting molds (Image courtesy of Firep
International AG).

recovery shafts, the GFRP bars are made part of the slurry reinforcement cage at TBM
break-​out “tunnel eye” area so the TBM head can easily cut through the support-​of-​
excavation wall without damaging the TBM cutterhead, this has become a common
industry practice.

13.6 INNOVATIONS IN PRECAST SEGMENTAL TUNNEL LINING

CORROSION RESISTANCE
While concrete has been used to line tunnels for over 100 years, all concrete sewer linings
cast-​in-​place or precast segmental tunnel linings, are subject to concrete corrosion caused by
microbial growth in the tunnel. Sulfides are generated in sewers as bacteria colonizes above
the water line, consume hydrogen sulfide and oxidize it to sulfuric acid. This process can be
accelerated by warm conditions in the tunnel. Over time the sulfuric acid will deteriorate
the concrete tunnel lining if not properly protected.
Designers of precast segmental tunnel liners are well informed about this industry problem
and are taking steps to address the issue. Some common current practices are as follows:

13.6.1 Adding sacrificial concrete to increase lining thickness

A common current industry practice is to increase the segment concrete thickness allowing
for concrete deterioration while not impacting the long-​term structural design capabilities
of the lining.
 Innovative products and applications 789

Figure 13.28 Intrados HDPE membrane system.

Note: System used in sewer tunnel in Doha, Qatar.

13.6.2 Integrated plastic intrados face membrane with welded joints

Cast a membrane liner on the inside face of the segment during initial casting then weld the
joints of the segments with membrane strips after the lining has been installed in the tun-
nel. This is commonly called the Doha System as a major sewer tunnel project was recently
completed in Doha, Qatar using this system as shown in Figure 13.28. Refer to Chapter 6
for details on this system.

13.6.3 Combisegment system

Cast the segments with a fully integrated HDPE liner/​gasket system located near the
segment intrados that when cast creates a fully sealed lining when installed in the tun-
nel. This system requires no secondary operations to seal the lining after installation.
Herrenknecht, AG has developed the system and it is marketed under the trade name
Combisegment. Refer to Chapter 8. The system has been successfully installed in a pre­
cast segmental tunnel lining sewer tunnel near Toronto, Ontario, Canada as shown
above in Figure 13.29.

13.6.4 Membrane application –​ post-​tunnel construction

Another option used is to install a membrane lining in the tunnel as a stand-​alone operation
post-​tunnel construction whereby the lining has external feet that adhere to bonding grout
that is installed behind the lining bonding the membrane to tunnel substrate. The tunnel
will be protected 270 to 360° with a membrane lining. This system is very common in sewer
rehabilitation projects and can be used on new construction. This system was used on a tun-
nel in Columbus, Ohio.
790 Handbook of Precast Segmental Tunnel Lining Systems

Figure 13.29 Herrenknecht Combisegment System –​ with HDPE membrane system.

Note: System used in the West Trunk Sewer Tunnel No.2 in Toronto, Ontario, Canada.

13.6.5 Post-​construction application of protective coatings

In areas with a high potential for corrosion such as in drop shaft landing areas with high
turbulence. A corrosion-​resistant coating can be spray applied to the newly installed con-
crete segments. This may not be a long-​term solution as reapplication may be required
on a predetermined interval, but the system can be effective if properly maintained.
Sauereisen Inc. supplies such products to the underground construction industry for this
application.

13.6.6 Design note regarding membrane plastic linings

Potential concerns with membrane sealing systems are that they have a limited resistance of
approximately 1 to 2 bar (14.5 to 29 psi) of external pressure that may act upon the lining
if a leak would develop in the precast segmental tunnel lining system. Check closely with the
various manufacturers as they typically are aware of this potential problem and are working
to improve membrane resistance to external groundwater forces.

13.7 INNOVATIONS IN PRECAST SEGMENTAL TUNNEL LINING DESIGN

The following innovations have enabled precast segmental tunnel lining to be more adaptive
to wide range of geological conditions and end use applications.

13.7.1 Composite segmental tunnel lining

For the recent Los Angeles County Metropolitan Transportation Authority (LACMTA)
Westside Purple Line Extension Project Sections 2 and 3, the owner specified at the time of
bid that for the areas where the tunnel crossed various active faults, the contractors provide
a special fabricated steel tunnel lining system as a direct replacement for the precast seg-
mental tunnel lining specified for the running tunnels. The fabricated steel tunnel liner was
 Innovative products and applications 791

Figure 13.30 Composite liner with seismic expansion joint.

Note: This seismic design concept was initially developed for the LACMTA West Side Expansion
Program.
Source: Del Amo, et al, 2021.

very expensive due to the extensive fabrication costs and the special fire-​resistant coating
that was required post-​fabrication.
The tunnel contractor, the same for both contract sections, in conjunction with their
tunnel Design Engineer of Record, developed an alternative to the fabricated steel liner to
be used in the seismic area. The alternative is called a “composite segmental lining” that
consisted of the following as illustrated in Figure 13.30.

• Embedded continuous steel plates with Nelson studs for anchors

• Expansion steel plate over circumferential joints
• Circumferential dowel connectors that yielded at a predetermined force

The “composite segmental lining” has the same dimensions as the standard segmental lin-
ing since these segments are cast in a standard precast segmental tunnel lining forms. The
embedded steel plates are anchored to the intrados face of the precast segment with Nelson
studs. A seismic event typically induces high tensile and shear stresses in the tunnel final lin-
ing. The anchored steel plate is designed to provide stability to the precast segmental tunnel
lining by binding the concrete together to create a unitized structure after a seismic event.
A movement may occur during a seismic event. The system has steel expansion plates to
cover and contain movement in the circumferential joint. The plates are designed to prevent
the infiltration of soil in the event of the circumferential joints could open during a seismic
event. The revised segment design also incorporates a post-​tensioned rebar reinforcement
system cast into the special seismic section segments and circumferential dowels with special
yielding characteristics.
This system is a real innovation for the tunnel industry as it met the owner’s design
requirement for a “flexible” lining and provides a lower cost solution than an all-​steel fab-
ricated tunnel lining for the seismic areas (del Amo et al., 2021).

13.7.2 Segmental liner design for low external (confining) pressure

tunnels
Precast concrete segments have been occasionally used to confine internal water pressure
[typically less than about 380 kPa (55 psi) in lieu of a two-​pass liner (internal steel or
792 Handbook of Precast Segmental Tunnel Lining Systems

reinforced concrete liner)]. Typical occurrences for such applications are for water or sewer
conveyance.
Various ingenious and special methods (as cited by Szechy, 1966) have been implemented
in the past to lock in prestress in the segments to carry low internal water pressure. These
include the following:

• The Keiser method (1960) incorporating pressure grouting to 150% of the work-
ing load
• A 4.2 and 4.9 m (14 and 16 foot) diameter sewer tunnel under the Seine (Lalande,
1949) incorporating tensioning jacks and an external prestressing steel band around
the segment’s extrados and a reinforced shotcrete internal liner, and
• A sewer 4.5 m (14.7 foot) diameter tunnel under the Rhine (Hochtief Nachrichten,
1960), incorporating prestressing bar between joints.

Modern precast concrete segment technology has incorporated internal connecting steel
between the segment to transfer and carry the internal pressure. These include the following:

• A 3.35 m (11 foot) diameter outfall tunnel in San Diego incorporating internal rebar
hoop steel connected by bearing/​stress transfer plates between segments (Kaneshiro
et al., 1996) (refer to Figure 13.31).

Figure 13.31 Continuous hoop rebar joined by cast iron bearing transfer plates.
Note: Continuous hoop reinforcing steel joined by cast iron bearing transfer plates during
assembly in segment forms.
 Innovative products and applications 793

Figure 13.32 Continuous hoop bar with special C-​ and H-​s hape connectors system.
Note: Continuous hoop reinforcing steel joined by cast iron bearing transfer plates during
assembly in segment forms.
Source: Obayashi, 2023.

• An outfall 10.4m (34-​feet) outside diameter tunnel for the Metropolitan Area Outer
Discharge Channel in Tokyo Bay incorporating internal rebar hoop steel connected
by C-​shaped connectors, which are in turn butted together with H-​shape connectors
(Miyao, et al., 1999, Obayashi Corp., 2023). Refer to Figure 13.32. A description of
the key features of this system is as follows:
• Radial joints incorporate wedges inserted along the tunnel axis by incorporating
C-​shaped connectors that are embedded on the joint surface. When the two seg-
ments are butted together, those create an H-​shaped space. Then an H-​shaped
connector is inserted into the space to fasten the two segments together.
• Gaps in the H-​shaped connectors are filled with quick-​set, non-​shrink cement
paste for corrosion resistance.
• Circumferential joints incorporate push grips (wedge pins) that retain the expansion
force of the seals between rings so that water tightness is assured. Also, the circum-
ferential joint is a tongue and groove configuration to accommodate shear forces.

Four 2.7 to 3.7 m (9 to 12 foot) outside diameter effluent tunnels in Japan where the design
was incorporated:

• Saitou et al., 1999

• Nishikawa, 2003
• Shield Tunneling, 2004
• Hirosawa et al., 2008

The Thun Flood Relief Tunnel, 6.0 m (19.7 foot) outside diameter in Switzerland (Kohler and
Rupp, 2008; Haefliger, 2009), and the Los Angeles County Sanitation District’s (LACSD)
794 Handbook of Precast Segmental Tunnel Lining Systems

Effluent Outfall Tunnel (Van Greunen et al., 2016; Cimiotti et al., 2021) incorporated con-
tinuous post-​tensioned tendons (Figure 13.33).
The LACSD 5.5 m (18 foot) internal diameter tunnel is the first application of a post-​
tensioning system in the USA and the extent is the largest worldwide. The contractor’s
design for this project included a flat duct system to improve constructability (Figure 13.34),
compared to the Japanese and referenced design. Optimization details of the post-​tensioning
system are provided in Cimiotti et al. (2021).
In summary, strategies to carry low internal water pressures when designing precast con-
crete segment liners requires the following considerations:

• Evaluation of risks in terms of probability and consequences

• Ground modulus combined with secondary grouting to lock in thrust and check on
gasket gap vs. water pressure resistance.
• Hoop steel with stress transfer by bearing plates or other special transfer plates or
by use of prestressing tendons. As the diameter of tunnel becomes larger prestressing
tendons are more favored as the bearing transfer plate detailing may require a thicker
segment to accommodate the required stress transfer.
• Anchorage/​bearing transfer detailing, testing and proof-​of-​concept and its constructa-
bility, and corrosion protection.

Figure 13.33 Continuous post-​t ensioned segment lining method ( Japanese round anchor).
Note: Continuous hoop reinforcing steel joined by cast iron bearing transfer plates during
assembly in segment forms.
Source: Nishikawa, 2003
 Innovative products and applications 795

Figure 13.34 LACSD flat anchor duct system.

Note: The flat anchor system improves the internal hydraulics of the lining by eliminating obstruc-
tion that will cause turbulence during flow.
Source: (Courtesy of Dragados USA and LACSD)

Areas for further innovation and evaluation may include the following “embryonic” alter-
native material systems such as

• Consideration of a thin steel or plastic embedded liner and

• Use of alternative materials such as glass fiber reinforced polymer (GFRP) rebar where
innovations are being made in bridge construction because of its corrosion resistance
characteristics and high tensile strength.

13.8 INNOVATIVE APPLICATIONS FOR PRECAST SEGMENTAL TUNNEL

LININGS

13.8.1 Lining systems used in mine construction and development

projects
In the past mine development access entries were typically constructed by the drill and
blast construction method. Any topsoil was typically open-​cut, or a shaft was constructed
to a rock interface. At this point conventional drill blast construction is used to develop an
access entry or shaft to the ore body. The use of TBMs and precast segmental tunnel lining is
a recent development in the worldwide mine construction industry. There are two projects
that have incorporated TBMs and precast segmental tunnel lining that have proven this is a
mine access development method worthy of consideration on future projects.
796 Handbook of Precast Segmental Tunnel Lining Systems

Grosvenor Coal Mine Project in Central Queensland, Australia

The Anglo-​American Company tendered in 2012 the construction of two declines, one at
783 m (2,569 feet) and one at 1,023 m (3,356 feet). The contract was awarded to Redpath
Australia who proposed an EPB TBM/​precast segment lining alternative to the planned con-
ventional drill and blast construction method.
The geology above the coal seam was varying types of sedimentary rock with layers of
sand and clay. It was determined that a shielded TBM with a precast segmental tunnel lining
would be the best way to address the mixed ground geological conditions.

Project Details

• Conveyor Drift Decline

• Length   783 m (2,570 feet) at a 1:6 decline slope
• Diameter 8.0 m (26.25 feet) mined, 7.0 m (23.0 feet) finished
• Transport Drift Decline
• Length    1,003 m (3,290 feet) at a 1:8 decline slope
• Diameter   8.0 m (26.25 feet) mined, 7.0 m (23.0 feet) finished
• Precast Segmental Tunnel Lining
• Outside diameter 8.0 m (26.25 feet)
• Inside diameter 7.0 m (23.0 feet)
• Segment arrangement 5+​1 with flat invert segment
• Joint connectors   Bolts at radial and circumferential joints

The TBM was supplied by Robbins. They designed a special ventilation system to deal with
the presence of methane and included a rescue chamber on the nine-​section back-​up gantry.
The TBM had an innovative disassembly design that enabled the center core of the TBM to
be unbolted and retrieved back through the 7.0 m (23 feet) internal diameter precast seg-
mental lining, this facilitated one TBM to be used for both declines.
Construction of the project was initiated in July, 2012 and both declines were completed
by March, 2015, ahead of the initial desired completion date based on conventional drill
and blast decline construction. All parties involved viewed this innovation solution a suc-
cess as the work was done safely and completed on time.

York Potash Woodsmith Mine Conveyor Tunnel, Whitby, England

The purpose of the project is to transport mined potash (polyhalite) from the 1,500 m
(4,820 feet) deep production shaft to the processing plant and deep-​water port ship loading
facility located 37 km (22.6 mile) from the mine as illustrated below in Figure 13.35. The
typical industry method for transporting mined aggregate long distances would be a large
overland belt conveyor system.
This project presented a significant design challenge in that the area between the mine
production shaft and the processing mill/​deep water load-​out facility was the North York
Moors National Park with a protected forest woodlands area that prevented the placement
of a large surface conveyor system and associated service roads.
The solution was to put the conveyor system in an underground tunnel through which the
conveyor system would be assembled and operated with minimal impact to the protected
woodlands above.
Based on geotechnical investigations it was determined that the most efficient way to con-
struct the 37 km (22.6 mile) long tunnel would be the use a TBM with a precast segmental
 Innovative products and applications 797

Figure 13.35 Woodsmith Project Layout.

Note: Overview of the project from mine heading to the product loadout for shipment to the
end user. (Image Courtesy of AngloAmerican).

tunnel lining as the final lining. The TBM provided mining efficiency in the soft rock Redcar
Mudstone geological conditions. Precast segmental tunnel linings with a gasket sealing
system will isolate the Mineral Transportation System (MTS) from the protected wood-
lands above.

Tunnel lining details

• Tunnel lining outside diameter 5.6 m (18.4 feet)

• Tunnel lining inside diameter 4.9 m (16.0 feet)
• Tunnel lining thickness     350 mm (13.75 inches)
• Ring length          1.5 m (4.92 feet)
• Ring arrangement       5+​ 1 key
• Lining reinforcement      Steel Fiber 500 MPa (72,500 psi)
• Gasket sealing system EPDM anchored with 20 mm (0.75 inches) face
• Hydrostatic pressure design up to 29 bar (420 psi)
• Circumferential joint   Dowel connectors
• Radial joint       Steel bolt connectors

Tunnel construction was started in 2020 by Strabag. Their contract is for the 37
km (23.6 miles) of the MTS. Tunneling is advancing well. This innovative precast seg-
mental tunnel lining concept solved a sensitive environmental issue and has allowed the
798 Handbook of Precast Segmental Tunnel Lining Systems

project to advance forward. Strabag states that they are building a sustainable tunnel
system as they sourced local sand and aggregates for the segment manufacturing and
are using modern diesel-​electric hybrid tunnel locomotives that incorporate a kinetic
energy recovery system while reducing emissions resulting in a very innovative tunnel
project.

13.9 LINING APPLICATIONS IN SHAFT CONSTRUCTION

Accessing the tunnel elevation from the surface has always been a time consuming and
expensive operation as frequently varying geological conditions will be encountered and
thus the excavation and support systems must be designed to address these situations. In
recent years there have been innovations in the excavation and support systems to make this
safer and less costly.

13.9.1 Vertical shaft sinking machine systems using precast

segmental lining
Accessing the tunnel elevation to begin the actual tunnel construction has always been a
complex, expensive and time-​consuming task for tunnel construction companies.
One of the more significant recent innovations in the tunnel industry has been the
development of the vertical shaft sinking machine (VSM) system that incorporates pre-
cast segmental linings as the primary shaft support system. The system as developed by
Herrenknecht AG in Germany, constructs the shaft “top down” using a mechanized exca-
vation machine with a slurry muck excavation system integrated with the precast segmental
tunnel lining support system that is added at the shaft collar area as the shaft is excavated.
The precast segmental tunnel lining acts as a modified sunken caisson system lowering itself
due to the mass weight of the concrete segments as the shaft is excavated as illustrated in
the images in Figure 13.36.
To date shafts have been constructed throughout the world in sizes ranging from 4.5 m to
12.0 m (14.75 to 12.37 feet). The system can be used in varying geological conditions and
the gasketed PSTL provides a watertight shaft lining upon completion.
The linings have standard design parameters based on the diameter of the shaft. Through
bolts are used in the ring-​to-​ring joint to compress the sealing gasket, located near the

Figure 13.36 Typical vertical shaft sinking operations.

Note: Note the confined workspace from which vertical shaft mining operations can be con-
ducted. (Courtesy of Herrenknecht AG.)
 Innovative products and applications 799

intrados face of the segment. The gaskets prevent external ground water from entering
the shaft.

Vertical shaft mining case history –​Ballard Siphon Project, Seattle, WA

The vertical shaft mining (VSM) system with a precast segmental tunnel lining as the pri-
mary lining was used for the first time in North America to construct the Ballard Siphon
Project in Seattle, WA (USA) in 2012. The project had two shafts for access and recovery:

• Forebay structure 27 m (89 feet) deep at north location

• Afterbay structure 41 m (134 feet) deep at south location

The south structure was constructed first using a Herrenknecht vertical shaft sinking
machine. The shaft was lined with specially designed and manufactured precast segmental
shaft lining segments.

Shaft lining details

• Shaft lining outside diameter 9.80 m (32.1 feet)

• Shaft lining inside diameter 9.0 m (27.5 feet)
• Shaft lining thickness 400 mm (15.8 inches)
• Shaft lining width 1.0 m (3.3 feet)
• Ring arrangement 4 piece ring (4 pieces at 90°/​each)
• Ring to ring connection 18 mm (0.7 inches) continuous rods, 8 per ring
• Radial joint connector 27 mm (0.9 inches) straight bolts 8 per ring
• Lining reinforcement: steel fiber 38 kg/​m3 (64 lbs/​yd3)
• Lining concrete design strength 41.4 MPa (6,000 psi)

The system performed as planned and designed. The south structure VSM excavation began
in January 2012 and was completed in June 2012. The VSM/​precast lining system was then
used to construct the north shaft.

13.9.2 Conventional access shaft construction using precast linings

In England and in Toronto, Ontario, Canada where the geology above the tunnel is basic-
ally a stiff self-​supporting clay or shale, tunnel contractors use special designed precast seg-
mental linings to support the shafts as they mine top town with conventional excavation
equipment with men and equipment located in the shaft during construction.
In this construction method the segments are lowered into the shaft then bolted to the pre-
viously installed ring by bolting through the segment connecting to the previously installed
ring. Refer to Figures 13.37 and 13.38.
Each ring is then completed, the leading edge is sealed with excelsior and grout is pumped
behind the liner at low pressure to put the precast segmental tunnel lining in compression
with the surrounding strata. In England and Toronto, Canada, the tunnel constructors have
somewhat standardized on access shaft diameters for tunnel projects so local precast seg-
ment manufacturers have lining casting molds in inventory. This assists in keeping this shaft
construction competitive.
To facilitate installation in the shaft the segments are cast with a conduit through the
segment through which the circumferential bolts are inserted to connect to the ring above.
The radial or vertical joints are typically connected with steel bolts or a special longitudinal
fixing joint that connects rebar hooks cast into the segments.
800 Handbook of Precast Segmental Tunnel Lining Systems

Figure 13.37 Typical shaft PSTL ring-​t o-​r ing connection detail.

Note: The segment being installed is supported from above by the previously installed ring with
special thru-​b olts specifically designed for the application.

Figure 13.38 Vertical shaft segment installation.

Note: The ring segment being installed is supported from the surface until the workers install
the connection system as detailed in Figure 13.37.

13.10 UTILITY CORRIDOR TUNNEL CONSTRUCTION USING

PRECAST LININGS
Throughout the world in major urban areas, it has become common to build dedicated
tunnels in which electric cables, fiber optic cables, gas lines and water lines are mounted in
the tunnel thus exposing the various lines in the tunnel, which has a walkway thus making
the monitoring and replacement/​upgrading of a utility conduit easier. Moscow and London
 Innovative products and applications 801

have extensive utility corridor tunnel systems and other metropolitan areas are currently
implanting or planning such underground structures as a precast segmental tunnel lining
provides an excellent surface for the attachment of the many utilities. These tunnels are typ-
ically in the range of 4.0 to 6.0 m (13.12 to 19.68 feet) diameter.

13.11 GAS AND OIL INDUSTRY PIPELINE TUNNELS

Gas and oil pipelines are typically buried and backfilled in trench construction techniques.
When the pipelines have to go under a lake or through a mountain over which a pipeline
cannot be placed because of governmental regulations. Precast segmental tunnel lined tun-
nels provide many benefits for gas and oil pipelines:

• Pipelines remain exposed in the tunnel for spill monitoring

• Pipelines can be easily repaired or replaced
• If a spill occurs, the segment gaskets will keep the oil contained within the precast
segmental tunnel lining
• Other utilities can be put in this same tunnel
• In a seismic event, the precast segmental tunnel lining will absorb the external forces
protecting the pipes inside
• In certain parts of the world pipelines are put in tunnels for security reasons

Examples of oil and gas energy pipeline projects in North America include the following.

13.11.1 Enbridge Line 5 Tunnel, Traverse City, Michigan, USA

Purpose:

• Replace an existing 50-​year-​old steel pipeline on the bottom of Lake Michigan

Project details:

• 6.4 km (4.0 miles) of tunnel to be built by TBM and using a precast segmental tunnel
lining for support and watertight construction.

Tunnel lining arrangements (projected):

• Tunnel lining outside diameter 7.16 m (23.5 feet)

• Tunnel lining inside diameter 6.40m (21.0 feet)
• Ring arrangements      5+​ 1 key
• Circumferential joints   Dowels
• Radial joints       Bolts
• Segment gaskets       Neoprene gaskets specified

Status:

• In procurement phase     Summer 2022

802 Handbook of Precast Segmental Tunnel Lining Systems

13.11.2 Kinder Morgan Burnaby Mountain Tunnel Project,

Burnaby, BC, Canada
Purpose:

• Contain an oil pipeline from Alberta to the Port of Vancouver, Canada

Project Details:

• 2.6 km (1.64 miles) of tunnel being built with TBM /​PSTL construction

Tunnel Lining Arrangements (projected):

• Tunnel lining outside diameter 4.11 m (13.5 feet)

• Tunnel lining inside diameter   3.65 m (12.0 feet)
• Ring arrangements       6-​ piece universal ring configuration
• Circumferential joints     Dowels
• Radial joints        Bolts
• Segment gaskets        EPDM gaskets

Status:

• Under construction       Summer 2021

13.11.3 Eagle Mountain–​Woodfibre Gas Pipeline Project,

Squamish, BC Canada
Purpose:

• Contain twin gas pipelines running from Coquitlam, BC to Woodfibre LNG Plant

Project Details:

• 3.8 km (2.36 miles) of tunnel to be built with TBM and using a precast segmental
tunnel lining for support and watertight construction.

Tunnel Lining Arrangements (projected):

• Tunnel lining outside diameter 4.0 m (13.1 feet)

• Tunnel lining inside diameter   3.5 m (11.5 feet)
• Ring arrangements       5+​ 1
• Circumferential joints      Dowels
• Radial joints         Bolts
• Segment gaskets        EPDM gaskets

Status:

• In construction phase      Spring 2023

 Innovative products and applications 803

Throughout the world it is common practice to put oil and gas pipelines inside tunnels for
the reasons noted above, there are many of these types of tunnels in operation and being
planned for Eastern Europe/​Western Asia used in the transport of energy to Western Europe
and the Black Sea ports.

13.12 INNOVATIONS IN SEGMENT CASTING PLANT TECHNOLOGY

Steel tubbings had to be manufactured at a steel foundry that could produce and place
molten iron, a very expensive and costly operation. The trend in modern precast segmental
tunnel lining production is to produce the tunnel lining near the project location or within
a manageable shipping distance. This is made possible as major components in segment
manufacturing facility, as detailed in Chapter 8 and Chapter 9 are transportable and rela­­
tively easy to assemble to meet a specific project requirement.
In some major urban areas in North America there has been a continuation of projects
to enable the same segment manufacturing facilities to produce various products from the
same fixed facility, i.e. Los Angeles, California, Cleveland, Ohio in the United States, and
Toronto, Ontario, Canada. This enables the retainage of trained and skilled segment pro-
duction workers.

13.12.1 Major tunnel segment precasting innovations

With greater industry acceptance and use of precast segmental concrete tunnel linings came
many innovations that improved the casting quality of the concrete as improved casting
tolerances and long-​term service life, up to 100 years, were required. Innovations in com-
ponentry used in the manufacture of the precast segmental concrete tunnel linings have
accelerated with the introduction of high production manufacturing systems and segment
handling systems that greatly reduces damage during the segment handling and transpor-
tation operation. Some examples are as follows and are discussed in detail in previous
chapters.

• Carousel style production lines

• Introduction of precision machined steel casting molds
• Innovations in concrete admixtures
• Introduction of fiber (steel or plastic reinforcement)
• Vacuum systems for demolding and stacking
• Controlled steam curing kilns; eliminated water emersion curing

13.13 INNOVATIVE PERMANENT IDENTIFICATION SYSTEMS FOR

EACH PRECAST SEGMENT
Owners and designers wanted to be able to track each segment in the manufacturing process
and be able to identify the date of casting and the quality of the concrete that went into each
segment. With an identification system defective non-​compliance segment can be removed
from inventory, preventing their use in the tunnel, if it was determined post-​production that
a component used in the manufacturing: i.e., steel, concrete, curing procedure, was found
to have failed to meet project specifications.
804 Handbook of Precast Segmental Tunnel Lining Systems

13.13.1 Methods for identifying segments

Various manufacturers worldwide have developed segment identification systems:

• Computer chips or RFID (radio frequency identification) card that are embedded in
the concrete
• Marking plate, face mounted, with a numbering sequence for visual observation
• Marking plate, face mounted, with a bar code system that can be scanned

Marking location is important. Most owner quality control programs want to be able to
identify each segment when it is the stack in storage and then after the segment had been
installed in the tunnel. This requires that two marking plates be placed on the segment at
the time of manufacture. Please refer to Figure 13.39.

• Location 1 –​on the radial joint face so each marker can be read in a stack during
storage
• Location 2 –​on the intrados face so the marker can be read in the completed tunnel

13.14 INNOVATIONS IN PRECAST SEGMENT TUNNEL LINING

MATERIALS
Since the discovery of cementitious concrete, mankind has experimented with different
materials that perform the same function. In modern times new lining materials have been
developed and tested for tunnel applications as cement-​based tunnel linings can be impacted
over time by corrosive water or effluent that can reduce the service life of a cement based
linings. Polymer-​based concrete is one material that is currently being used as a replacement
for precast concrete pipe and segmental linings.

13.14.1 Precast polymer concrete segments

The development and innovations made in producing precast polymer concrete struc-
tures, for the chemical processing, mining and wastewater industries (manholes and sew-
erage conveyance) have been a standard-​of-​practice since 1985 (see Figure 13.40). Interest
in using precast polymer concrete for segmental tunnel linings (see Figure 13.41) has

Figure 13.39 Typical PSTL marker plate.

Note: The above is an example of a typical marker plate in use today.
 Innovative products and applications 805

Figure 13.40 Polymer concrete pipes and manholes.

Note: Polymer concrete pipe and manholes awaiting installation in an open-​c ut application.

Figure 13.41 Demonstration precast polymer segments.

Note: Cast polymer concrete segments stored at a demonstration location. Note the 5+​1 ring
arrangement.

grown significantly in the last 15 years (Cubeta, 2008). They were approved and specified
as an alternative for a sewage project in the USA, but not part of the winning project
because of cost (Klein et al., 2010). Tauber Rohrbau GmbH & Co. has been building
small diameter three-​piece segments from 1.0 to 3.0 m (3.3 to 9.8 feet) in diameter using
polymer concrete conforming to DIN 54815-​1,–​2 (Maguire and Iskander, 2008; Tauber
Rohrbau GmbH & Co, 2021). More recently, the Dubai Strategic Sewerage Tunnels
System is specifying polymer concrete 60 m (197 feet) upstream and downstream (Monks,
2021) of drop shafts and Detroit, Michigan’s Oakland-​Macomb Northeast Interceptor
relining will incorporate a four-​piece, 4.9 m (16 foot) internal diameter polymer segments
(Asadollahi, 2021).
Technological advancements in formulations, manufacturing and design have made
precast polymer concrete tunnel segments considerably more practical and cost-​effective.
Precast polymer concrete has been a consideration for tunnel segments based on its
806 Handbook of Precast Segmental Tunnel Lining Systems

superior corrosion resistance and physical properties compared to thin-​film polymer coated
(embedded fiberglass or spray on epoxy liners) or mechanically or chemically bonded
thermo-​plastic sheets; e.g. PVC or HDPE, lined with conventional Portland cement concrete
segments, which have limitations with respect to external pressure.
In general, polymer concrete is approximately 4 to 6 times stronger than conventional
concrete, such that precast polymer concrete segments could be produced as much as 50%
thinner than conventional concrete segments, and as much as 50% lighter, based on newly
developed and patented core technology. Precast polymer concrete is made from various
thermo-​setting polymer binders, such as polyester, vinyl ester, polyurea and epoxy resins,
combined with inert aggregate fillers. Historically, these traditional binder systems have
had issues. All resin systems are not the same and may contain volatile organic compounds
(VOC) and hazardous air pollutants (HAP), making it difficult to permit and subject workers
to occupational safety and health problems. Also, some resin systems have high shrinkage,
high exothermic reaction, and cannot integrate reinforcing steel because of their differential
thermal properties. While other resin systems, containing VOC’s, HAP’s, and exhibiting
thermal and shrinkage issues are designed such that these are not obstacles to polymer con-
crete production.
Until recently, precast polymer concrete has been produced in small, manually operated,
batch plants, culminating in increased handling, questionable QA/​QC and increased manu-
facturing costs. Polymer concrete producers are adapting to automated batching systems
and RFID microchip technology used in conventional concrete production.
In the case of vehicular tunnels, where corrosion is not an issue, segments can be designed
and manufactured using waffle designs, reducing the overall raw material volumes and cost.
However, waffled designs cannot be utilized in sanitary sewer tunnels due to sanitary sew-
age flow characteristics and turbulence created by waffle designs. A smooth internal surface
is essential. New and innovative polymer concrete segment design allows for a significant
reduction in the volume and weight of polymer concrete segments, while still maintaining a
smooth internal and external surface along with preserving physical and corrosion resistant
properties. By integrating and positioning lightweight cores within the polymer concrete
matrix, the polymer concrete tunnel segment can carry basically the same loads, as if it were
a solid polymer concrete segment (Cubeta and Kaneshiro, 2021). See Figure 13.42. This
innovative technology makes precast polymer concrete tunnel segments a cost-​effective, cor-
rosion resistant alternative to thin-​film coated or sheet-​lined conventional Portland cement
concrete segmental tunnel lining systems.

Figure 13.42 Patented light weight cores for polymer segments.

Note: To lower the cost of the polymer segment an innovative design was developed that incor-
porates light weight center cores that are encapsulated with a polymer matrix designed to meet
the required loads.
 Innovative products and applications 807

13.15 FUTURE INNOVATIONS FOR PRECAST SEGMENTAL TUNNEL

LINING MATERIALS AND DESIGN
In today’s modern society everyone must be conscious of the impact our respective products
may have on the environment. Heavy construction does acknowledge a social responsibility
and are evaluating our construction methods and products to have a lower carbon foot-
print. Concrete is a product in which environmental innovations are occurring.

13.15.1 Green concrete for sustainable construction

The future of concrete has an interesting dilemma. Highlighted as a major contributor to
global warming potential based on cement content brings some to the conclusion that it
should be eliminated from use in construction. However, on a mass-​to-​mass basis concrete
not only has lower embodied carbon than some alternative structural materials but can be
produced with recycled materials, provides excellent resiliency and durability during the use
phase and has the potential to be fully recycled at the end of its lifetime. In the context of
sustainability (balance of environmental, social and economic indicators), concrete may be
an excellent option.
The challenge in marketing concrete as a more sustainable solution is in developing con-
crete formulations that meet or exceed engineering, construction and architectural require-
ments while reducing the environmental impacts associated with its production, placement,
use and end-​of-​life disposition. In addition, any claims on sustainable solutions should be
quantified and validated to prevent inferences of greenwashing.
The quantification and transparent communication of environmental information of con-
crete mixtures/​products should be generated using life cycle assessment (LCA), a science-​
based evaluation of a material or product that follows specific International Standards
Organization (ISO) standards. The evaluation can be from cradle-​to-​gate (raw material
extraction through production including transportation to the producer’s location) or
cradle-​to-​grave (cradle-​to-​gate plus use and end-​of-​life stages). The quantified results are
used to validate any claims on a more sustainable option.
New innovations in concrete mixtures are often viewed with skepticism based on a lack
of decades of performance data and an overarching resistance to change, e.g. “this mix
design has always worked so why introduce any changes?” Some of the resistance is evident
purely based on the concrete specifications for a project. Prescriptive specifications gener-
ally restrict a producer’s opportunity to optimize mixtures that will meet all engineering and
design requirements and contractor expectations in addition to delivering a more sustain-
able solution –​sometimes at an equal cost to a current option. A transition to performance-​
based specifications provides much more latitude to develop innovative concrete mixtures
that meet or exceed all engineering and construction requirements but can also provide a
much more sustainable solution.
Additionally, concrete mixture optimization should be developed and evaluated during
the early design stages of a project allowing for the greatest opportunity to provide the most
sustainable solutions. This approach can ensure that the correct concrete mixture is used in
the project and not altered during the bidding process for economic purposes jeopardizing
the best designed, most sustainable solution.
Evaluating the anticipated changes in the cement and concrete industry over the next
decade to reduce CO2 emissions associated with clinker production will require innovative
solutions and advanced chemistries to meet current and future demands for more durable,
resilient, higher strength and sustainable concrete options.
808 Handbook of Precast Segmental Tunnel Lining Systems

Today, current admixture products such as water-​reducers and strength-​enhancers can

potentially reduce embodied carbon by 5 to 20% (NRMCA, 2021) by concrete mix-
ture optimization using less cement to meet engineering and construction requirements.
Supplementary cementitious materials can also be used for cement replacement with poten-
tial embodied carbon reductions ranging from 20 to 90% (NRMCA, 2021). However, codes
and specifications may limit the higher percent reductions. Moving forward, new cement
blends and cement types, new binder types, additional supplementary cementitious materi-
als and systems introducing waste emissions such as CO2 directly into concrete mixtures,
will require even more advanced chemistries to meet or exceed engineering and construction
requirements while improving on sustainability results.
Advanced concrete admixtures available today are actively supporting more sustainable
solutions in concrete design throughout the life cycle of concrete. In the production stage,
strength enhancing calcium-​silicate-​hydrate (CSH) nanoparticle technology supports the
reduction of cement content and/​or increases the use of supplementary cementitious materi-
als. The addition of this admixture allows for the reduction in embodied carbon by lower-
ing the cement content without affecting engineering properties.
Admixtures such as workability-​retaining admixtures support slump retention in con-
crete mixtures reducing the need for job site adjustments without affecting early-​ age
strength development or time set of concrete. The consistent quality reduces rejections,
which can lower the impact of increased CO2 emissions from re-​producing new concrete
in addition to added costs associated with rejected loads. Supporting the end-​of-​life stages
in concrete products are admixtures that are formulated to treat returned fresh concrete
for subsequent beneficial use such as road base or other valuable applications. This solu-
tion reduces the need for on-​site crushing or potential landfill enhancing the objectives of
a circular economy.
Additional new admixture chemistries will be needed to further support changes in clinker
content and blended cement types as cement producers develop additional process changes
to lower their CO2 impact. New supplementary cementitious materials will likely be intro-
duced requiring additional chemistries to support the required engineering properties of
concrete as well as contractor expectations for construction. Connecting these solutions to
sustainable objectives will be through the transparent quantification and communication of
results using life cycle assessment protocols. Currently, environmental product declarations
(EPD), support this reporting but advancements reducing the time required and associated
costs will play a key role in delivering real-​time information.
The concrete industry is on the precipice of significant change to support the extreme
demands for more sustainable engineered concrete. These changes will require support,
insight and innovation from all elements of the construction value chain to be successful.
A global product that is used more than any other man-​made material in construction not
only carries a great burden of impact but can also play the greatest role in significantly
improving our approach to sustainable construction.

13.16 FUTURE PSTL INNOVATIVE NEEDS

As the need for more tunnels increase to meet the demands of a modern society, the pre-
dominant tunnel lining will be precast concrete segmental tunnel linings as they are versatile
in their application and facilitate tunnel construction by the use of circular tunnel boring
machines (TBMs). Some potential innovative needs are as follows.
 Innovative products and applications 809

13.16.1 Improvements in concrete corrosion resistance

The various corrosion protection methods as noted previously in this chapter are varied,
expensive to implement and many are susceptible to damage during to the tunnel construc-
tion process which is a challenging environment. Extreme care must be taken to protect the
precast segmental tunnel lining’s pre-​applied membranes from TBM trailing gear, rail traf-
fic and worker, making non-​detected penetration or damage. The ideal product would be
an admixture that is added during concrete mixing process that would provide long-​term
corrosion protection, remedial cleaning and reapplication after a specified period of time
would be acceptable. Proper tunnel maintenance is something that needs to be addressed
and improved in the industry. New corrosion resistant materials, i.e. silicates, may need to
be evaluated.

13.16.2 Automation in precast segmental tunnel lining

manufacturing
The major international precast segmental tunnel lining mold manufacturers are all inves-
tigating was to automate various aspects or all of the manufacturing processes. Finding
skilled laborers for a manufacturing program that may last for a relatively short period of
time is a worldwide problem. Automation will also help to maintain consistent product
quality. The major precasters need to continue to implement improved quality control mon-
itoring systems.

13.16.3 Improvements in precast segmental tunnel lining

componentry
The various international componentry manufacturers need to continue to develop new
and innovative materials and products. As detailed above in this chapter the various inno-
vations in connection and sealing componentry have improved precast segmental tunnel
lining quality, improved ring erection time, some tunnel contractors have 6.0 m (19.75 feet)
diameter ring-​build time down to under 15 minutes, while still maintaining ring circularity
to mm tolerances.

13.16.4 Compressible backfill grout that rebounds

Over the past years, the international tunnel industry has investigated and trialed attempts
to develop a “compressible” cementitious based backfill grout that will yield; i.e. “com-
press”, under underground loads from an expansive geology or from seismic activity. In the
1970’s the Germans experimented with a system using hollow glass beads whereby the glass
beads would crush under load thus equalizing the ground pressure around the tunnel lining.
The system worked in principal but was very expensive as glass beads are difficult to handle
and inject into a backfill grouting plant on site.
A more recent industry development, one of many, has been the introduction of a backfill
grout that incorporates engineered polymeric spheres (lightweight synthetic particles) used
in conjunction with a cementitious grout and other proprietary admixtures as part of a pat-
ented system used for backfilling behind the precast segmental tunnel lining for the Region of
Peel West Trunk Sewer, Contract 2 near Toronto, Ontario, Canada. The system performed
as designed but what the industry desires for is a system that has rebound characteristics as
810 Handbook of Precast Segmental Tunnel Lining Systems

many of the loads are cyclical. The industry need is for a system that will rebound after a
seismic event or if the external load relaxes due to a variety of possible reasons.

13.17 INNOVATIVE FUTURE APPLICATIONS FOR PRECAST

SEGMENTAL TUNNEL LINING

13.17.1 Non-​c ircular applications for segmental tunnel linings

During development of precast segmental tunnel linings, both circular and non-​circular
applications have been successfully a completed. While circular (full-​round) tunnel appli-
cations have been the most popular and the global tunneling industry has perfected this
approach with the close integration of tunnel boring machines and segmental tunnel liner
design and in consideration of all anticipated subsurface conditions, many non-​circular
projects deserve notice as historical achievements. As listed below in Table 13.1, several
significant tunnelling projects have made successful use of non-​circular and partially closed
precast concrete segmental lining approaches. Please refer to Figures 13.43 to 13.49 that
illustrate modern examples of the uses of precast segmental tunnel lining in non-​circular
tunnels
Additional discussion on future and applications and materials related to precast concrete
segmentally lined tunnels and shafts, including rehabilitation, multi-​function and repur-
posing of existing facilities is needed. This also includes energy recuperation systems and
use of segments on drill and blast excavated tunnels and caverns.

13.17.2 Retrofitting and rehabilitation of tunnels

The use of precast segmental tunnel liners has recently been considered for specific tunnel
and underground structure rehabilitation projects. In many cases, the existing facility,
once relined with precast elements (many are non-​circular) has begun a new and prolonged
life cycle. Additionally, some underground facilities may, with the relining in place, be

Figure 13.43 Non-​c ircular PSTL in an access tunnel.

Note: Application of non-​c ircular gasketed precast segmental tunnel liners in an arched access
tunnel. (© RhB, Andrea).
 newgenrtpdf
Table 13.1 Non-​c ircular applications for segmental tunnel linings.
Summary of molds types (and materials used) and production casting systems for recently completed tunnels where precast
segmental tunnel lining was selected as the final lining. Note the mold materials changes

Dimensions Tunnel shape Segmental lining features

Item Project name Project location Dia. Length Circ’r Non- ​C ir Gasket Bolts WP Reinf
1 Mount Royal Tunnel 1 CAN 1912 6.1m (20’) 500m (1,650 LF)   Note 1
Montreal, Quebec
2 Parmley System 2 USA >1902 Various Various   Note 2 
Various locations
3 Shanghai Metro 3 China >1990 Double ‘O’ Various      
Shanghai, China Note 3
4 Paris Metro Stations 4 France >1999 20m (66 ft) 180m (591 LF)  
Paris, France Note 4
5 Dual channel sewer Current Various Various  
Location unknown
6 Rail Tunnel Rehab 5 Current Various Various     

Innovative products and applications

Various locations Note 5
Notes:
1
According to contemporaneous construction reports published in 1912, segments were fabricated with large cross-​s ection “tendons” (protrusions on the
radial joints) that interlocked with corresponding depressions on the matching surfaces. These tunnel segments were named per the patent as “Interlocking
Tunnel Blocks” by O’Rourke.
2
According to contemporaneous product catalogue published in 1927, segment molds were fitted with block-​o uts for connection bolts that once installed
with backfilled with mortar to complete and improve the hydraulic surface. It appears that many of these segments were manually erected without powered
equipment.
3
A complex arrangement of non-​c ircular precast concrete elements needed to complete the tunnel lining as the highly specialized TBM advanced (and negotiated
modest curves also) and concurrently placed the Double ‘O’ liner.
4
Installed segments were backfill grouted. PVC waterproofing membrane was installed before installation of precast segments. Due to their size and weight, the
segments were erected using hydraulically powered equipment.
5
Some highly specialized applications of precast segment tunnel lining systems have been adapted for the installation in existing railway tunnel (e.g. Europe)
rehabilitation projects. This system includes multi-​s egment arch and side wall precast elements placed with specialized equipment and suitable backfill. There
is also an application for the use of “flat” precast tunnel invert segments (e.g. Whittier Tunnel in Alaska) and as a base for track systems (Yucca Mount
Exploratory Studies Facility) in Nevada, USA.

811
812 Handbook of Precast Segmental Tunnel Lining Systems

Figure 13.44 Non-​c ircular PSTL in a sewer tunnel.

Note: Application of non-​c ircular gasketed precast segmental tunnel liner in a portioned sewer
tunnel. (Image courtesy of CBE Group).

Figure 13.45 Installation of precast segmental lining after Station Cavern Excavation –​ Paris Metro.
Note: Special machine used for the ring erection consisting of 11 segments 1.2 m long and 0.8 m
thick (4.0 × 2.6 feet).
 Innovative products and applications 813

Figure 13.46 WMATA Metro station –​ train room –​ Picture 1.

Note: Application of non-​c ircular gasketed precast segmental tunnel liners in an arched metro
station.

Figure 13.47 WMATA Metro station –​ train room –​ Picture 2.

Note: Application of non-​c ircular gasketed precast segmental tunnel liners in an arched metro
station.
814 Handbook of Precast Segmental Tunnel Lining Systems

Figure 13.48 Shanghai Metro –​ Double-​O tunnel –​ demonstration rings.

Note: Demonstration rings for the Double-​O tunnel for the Shanghai Metro System. Note
bolted connections.

Figure 13.49 Shanghai Metro –​ Double-​O tunnel –​ completed segmental liner.

Note: Completed precast segmental liner for the Double-​O tunnel for the Shanghai Metro System.

repurposed from their original functions. Please refer to Figure 13.50 below, showing and
tunnel relining operation underway.
The benefits from relining tunnel and underground facilities with precast segment tunnel
lining includes the following for example

• Improved ground support

• Improved groundwater control
• Improved hydraulic conductivity (smoothness); for air and fluids as the case may be
• Improved fire protection
• As needed to support the repurposing requirements
 Innovative products and applications 815

Figure 13.50 Tunnel relining operations.

Note: Application of non-​c ircular precast segmental tunnel liners into an existing tunnel to pro-
long service life.

Figure 13.51 Janbach, Austria Pilot Scheme.

Note: The section there are 54 m between the red dots, is where the geothermal energy
recovery precast segmental tunnel lining was installed in the city of Jenbach, Austria.

13.18 GEOTHERMAL ENERGY RECOVERY USING PRECAST SEGMENTAL

TUNNEL LINING
In Europe it is a common practice to capture geothermal energy to heat and cool private,
commercial and government buildings. Below is an example of where geothermal energy
was captured from a nearby tunnel project to heat and cool a minicipal building. Please
refer to Figure 13.51.

Jenbach, Austria –​Pilot Scheme Project Description

• 54 m (175 feet) long section equipped with energy segment lining

• Supply of municipal building with heat energy
• Absorber pipe connection through escape shaft
• Start of operation; winter 2011

Figures 13.52 to 13.56 show examples of the concepts presented above.

816 Handbook of Precast Segmental Tunnel Lining Systems

Figure 13.52 Geothermal project arrangement and operations concept.

Note: The above figure shows how the system operates in the winter to recover geothermal
energy from the earth, in the summer the water flow is reversed to assist in cooling.

Figure 13.53 Absorber pipes fitted in cage –​ concept.

Note: Integrating energy absorbing pipes into precast segmental tunnel lining manufacturing.
The absorber pipes are installed and cast into the tunnel lining at the precast segment manufac-
turing plant.

13.19 INNOVATION IN MECHANIZED TUNNELING FOR CROSS-​

PASSAGE CONSTRUCTION
For the case of rail and road twin tubes providing traffic in both directions, cross-​passage
adits need to be built in regular distances to satisfy the fire life safety requirements. For
tunnels in competent rock or stiff soil, this can be done using the conventional excavation
methods. However, for tunnels in saturated soft soil under high hydrostatic pressure and
with high permeability of the ground, significant grouting and freezing program should be
implemented to control the risk. The construction cost and time for the application of this
method is significant and it is difficult to entirely mitigate the risk.
 Innovative products and applications 817

Figure 13.54 Absorber pipes fitted in cage –​ actual.

Note: The rebar cages assist in the proper placement of the absorber pipes.

Figure 13.55 Coupling of absorber pipes.

Note: The absorber pipes are coupled segment to segment to create a closed circulation loop.

To address these issues, an innovative method using the mechanized tunneling to build the
cross-​passage adits was used recently by Bouygues for the Tuen Mun Chek Lap Kok Link
(TMCLK) project in Hong Kong (Figure 13.57 and 13.58). In this method, a microtun­­
neling machine is used to bore the cross-​passage adit by launching the MTBM from one
tunnel and retrieving it in the parallel tunnel. The cross-​passages were bored by MTBM
while the main tunnel was still under construction. To be able to have a simultaneous con-
struction of the main tunnels and the cross-​passage tunnel connecting them and allow for
the continuous logistics of the main tunnel excavation, the internal diameter of the main
tunnels should be at least 10 m (32.8 feet). The 4.5 km (2.75 mile) long subsea section of
the TMCLK tunnel consists of two road tubes each with two traffic lanes in one direction.
These tunnels were built using two slurry TBMs [14 m (45.9 foot) excavation diameter]
to overcome the challenging ground conditions and the surrounding water pressure higher
than 5 bar (72 psi) at the tunnel face.
One of these TBMs started its drive with the world’s largest diameter 17.63 m (57.8
feet) along the first 630 m (2,070 feet) approach structure at the portal area before being
818 Handbook of Precast Segmental Tunnel Lining Systems

Figure 13.56 Coupled absorber pipes.

Note: The circulation connections remain exposed for leak detection and future maintenance.

Figure 13.57 TMCLK 17.63m TBM at Herrenknecht.

reconfigured into a 14 m (45.9 feet) diameter TBM (Combe and Schwob, 2021). For cutting
tools replacement, saturation diving and fully automatic robotic arms were developed
to change a total of about 2,000 disc cutters on both TBM cutterheads. For TMCLK in
addition to the new solutions used for TBM tunneling, innovative design concepts were
developed for the construction of the approach structures shafts including a 500 m (1,640
feet) long and 43 m (141 feet) deep “caterpillar shape” cofferdam with 15 cells.
As explained above, several major innovative solutions were used for the construction
of this mega project. However, the focus of this section is on the construction of 46 subsea
cross-​passages between the two parallel subsea road tunnels for which an innovative mecha-
nized tunneling method using a MBTM was applied for the first time. This method can also
be used to connect the tunnel to a shaft by launching the MTBM either from the shaft or
from the tunnel depending on the project construction approach. In transportation tunnels,
 Innovative products and applications 819

Figure 13.58 TMCLK project overview.

Figure 13.59 C ross-​p assage construction using microtunneling –​ an overview cross-​section –​

retrieval chamber.
Note: The system is sealed from outside ground water with retrieval chamber installed in
receiving tunnel.

this shaft can be a ventilation or an emergency egress shaft. Similar type of connection
between a tunnel and a shaft also exists in the case of water and wastewater projects when
a drop shaft or a distribution shaft is connected to the main tunnel.
National Fire Protection Association (NFPA) 502, the fire protection standard for road
tunnels requires a maximum spacing of 300 m (985 feet) between cross-​passages. However,
for TMCLK project, a maximum spacing of 100 m (330 feet) was used to provide a higher
fire safety standard. The initial design was suggesting the use of ground freezing method
which is typically used to build cross-​passages in grounds with high permeability and high
hydrostatic pressure. To reduce the risk and to remain within the tight schedule of the
project, the ground freezing design was changed to the use of microtunneling method and
jacking precast pipes to build the cross-​passage tunnels. Figures 13.59 and 13.60 show the
concept of this solution and how it can allow the simultaneous construction of the main
820 Handbook of Precast Segmental Tunnel Lining Systems

Figure 13.60 C ross-​p assage construction using microtunneling –​ an overview cross-​section –​

jacking side equipment.
Note: On the jacking side, a considerable amount of complex equipment must be installed.

Figure 13.61 Inside the launching side tunnel.

Note: Compact micro-​t unneling equipment and layout at the launching area.

tunnels and the cross-​passage adits when the diameter of the main tunnels is large enough
to allow for the concurrent construction logistics.
As shown below in Figures 13.61 and 13.62, the main challenge for the application of
this method is the layout and the design of the launch and retrieval areas for the MTBM
to allow the continuation of the main tunnels TBM excavation. At the cross-​passage loca-
tions, fiberglass reinforcing bars are used for the main tunnel segmental liner to allow for
the MTBM to bore through the segments. Also, on both sides of the MBTM drive, the
main tunnel liner needs to be reinforced with an end wall to allow for jacking forces to
 Innovative products and applications 821

Figure 13.62 Inside the retrieval side tunnel.

Note: Compact microtunneling equipment and layout on retrieval side micro-​t unneling machine
recovery area.

be adequately supported and distributed through the main tunnel liner and to prevent the
ovalization of the main tunnel rings. In addition, this wall allows the installation of a sealing
system to prevent the leakage during the temporary phase of construction and the per-
manent phase of operation.
For TMCLK project, 46 separate cross-​passages typically about 13 m (42.6 feet) long were
built using two 3.65 m (12.0 foot) diameter slurry MTBM (Herrenknecht AVN 3000) for
excavating the ground and installing the precast concrete pipes for the permanent liner of
the adit. The average time needed for the construction of one cross-​passage was 20 days
including transportation, assembly, microtunneling and disassembly. The specially designed
compact microtunneling equipment was used on both launching and receiving sides to min-
imize the microtunneling construction space required and to allow continuously keeping
one lane open for the main tunnel construction traffic. After the MTBM breaks into the
steel bell cap on the receiving side, the watertightness is temporarily ensured by grouting the
cross-​passage adit and the main tunnel interface, then the pressure inside the bell is lowered
and finally the bell is opened and the MTBM is retrieved.
In summary, the TMCLK project demonstrated the innovative use of mechanized tun-
neling for cross-​passage construction alongside with the main tunnel excavation. This
method can be applied practically in all type of geological and hydrogeological condi-
tions. If the internal diameter of the main tunnel is more than 10 m (32.8 feet) the exca-
vation of the cross-​passages can be performed concurrently with the boring of the main
tunnels. This mechanized cross-​passage construction method will increase the safety and
reduce the overall risk while at the same time it will result in major reduction in the con-
struction time and cost. In addition, the use of this method can be further extended to
the case of adit construction between the main tunnel and a shaft for transportation and
water projects.
822 Handbook of Precast Segmental Tunnel Lining Systems

BIBLIOGRAPHY AND ADDITIONAL READING MATERIALS

PUBLISHED STANDARDS
• ASTM A496 Standard Specification for Steel Wire, Deformed, for Concrete Reinforcement
• ASTM C1116 Standard Specification for Fiber-​Reinforced Concrete
• DIN 14889-​2 Fibres for Concrete–​Part 2: Polymer Fibres–​Definitions, Specifications and Conformity
• DIN 54815-​1 Pipes Made of Filled Polyester Resin Moulding Materials–​Dimensions,

MATERIAL AND MARKING

• DIN 54815-​2 Pipes Made of Filled Polyester Resin Moulding Materials–​Requirements and Testing.
• EN 14889-​ 2 European Standard for Fibres for Concrete–​ Part 2: Polymer Fibres–​Definitions,
Specifications and Conformity.
• ISO 14025 Environmental Labels and Declarations –​Type III Environmental Declarations –​
Principles and Procedures.

PUBLISHED TECHNICAL PAPERS AND DOCUMENTS

Asadollahi, P., 2021, “Detroit’s Oakland-​Macomb Interceptor Rehabilitation”, Building Materials
and Construction Technologies Conference, Dubai, UAE, 07 Apr 2021.
Cimiotti, C., N. Karlin, A. Navarro, E. Velasco, A. Sanz, and P. Navarro, 2021, “Segment Design for
Exceptional Circumstances: Post-​Tensioning and Squeezing Ground”, Rapid Excavation and
Tunneling Conference 2021 Proceedings, Las Vegas, NV.
Combe, B., and A. Schwob, 2021, “Tuen Mun–​Chek Lap Kok Link in Hong Kong –​Design and
Innovative Construction Methods of an Undersea Tunnel”, AFTES International Congress
2021, Paris, 6–​8 September 2021.
Cubeta, R., 2008, “Solid Cast Polymer Tunnel Segments: The Corrosion Resistant Solution for
Sanitary Sewer Tunnels”, North American Tunneling Conference Proceedings, San Francisco,
CA, 2008, pp. 58–​61.
Cubeta, R., and J.Y. Kaneshiro, 2021, “Polymer Concrete in Wastewater, Transportation, Mining and
Chemical Processing Industries and Its Carbon Footprint”, Building Materials and Construction
Technologies Conference, Dubai, UAE, 07 Apr 2021.
Del Amo, A., Y. Binghzi, R. Goodfellow, and G. Roy, 2021, “Composite Segmental Lining to Resist
Rupture”, RETC Conference Proceedings 2021, Las Vegas, NV, pp. 444–​455.
Haefliger, P., 2009, “Thun Flood Relief Tunnel”, Sept 2022 Presentation to D. Klug European Tunnel
Seminar.
Hirosawa, N., M. Nakashima, K. Imafuku, H. Nakayama, M. Miyake, M. Ishida, K. Teshima, and
H. Kinugawa, 2008, Development of Composite Concrete-​Packed Steel Segment, Nippon Steel
Technical Report No. 97, January, pp. 39–​44. www.nipp​onst​eel.com/​en/​tech/​rep​ort/​nsc/​pdf/​
n9707.pdf
Hochtief Nachrichten (Newsletter), 1960, “Rheintunnel Dusseldorf”, Aug/​Sept.
Kaneshiro, J.Y., S.J. Navin, and G.E. Korbin, 1996, “Unique Precast Concrete Segmented Liner for
the South Bay Ocean Outfall Project”, Proceedings of the International Conference on North
American Tunneling, Washington, DC.
Keiser, A., 1960, Druckstollenbau (Pressure Tunnel Construction), Springer-​Verlag, Wien-​Bregenz,
Austria.
Klein, G.S., M. Monaghan, S. Gambino, D. Deutscher, and R. Guizar, 2010, “Lined Concrete
Segments: An Alternative Construction Method for a Large Diameter Sewer”, North American
Tunneling Conference, Portland, OR, June 2010.
Kohler, D., and B. Rupp, 2008, “Thun Flood Relief Tunnel, Part 1–​Challenges Faced by the Hydro-​
Shield Method”. Tunnel, Vol. 7, pp. 27–​37.
Lalande, M., 1949, Diversité de applications du béton précontraint, Travaux, Paris, France, Feb.
Maguire, C., and E. Iskander, 2008, “The Use of Polymer Concrete in Sewage Tunnels for One-​Pass
Segmental Linings”, North American Tunneling Conference, San Francisco, 2008, pp. 146–​151.
 Innovative products and applications 823

Miyao, H., Y. Kamoshita, M. Kanai, K. Fukumoto, H. Watanabe, H. Minakami, M. Ozeki, and

H. Hashimoto, 1999, “Development of a New Lining for Shield Tunnels Bearing Internal
Pressures”, International Tunnelling Association World Tunnel Congress Proceedings, Oslo,
Norway.
Monks, G., “Dubai Strategic Sewerage Tunnel–​Tunnel Design Consideration”, Building Materials
and Construction Technologies Conference, 01 Apr 2021, Tauber. www.min​itun​nel.de/​min​itun​
nel.html
National Ready Mixed Concrete Association (NRMCA), 2021 (January), Net Zero Building
Presentation.
Nishikawa, K., 2003, “Development of a Precast and Prestressed Concrete Segmental Lining”,
Tunnelling and Underground Space, Volume 18, pp. 243–​251.
Obayashi Corporation, 2023, Home Solutions/​Technology OnePass Segment®, Smooth inner surface
type RC segment One Pass Segment® Eliminates secondar lining and achieves high speed con-
struction, website accessed December 12, www.obaya​shi.co.jp/​solu​tion​_​tec​hnol​ogy/​det​ail/​tech_​
d​055.html, Japanese translation courtesy of Obayashi Corporation, January 10, 2024.
Rousselier, M., 1952, “Le revêtement des galeries”, Annales de l’Institut Technique de B.T.P.,
Volume 59.
Saitou, S., M. Kaneko, T. Sagara, M. Sugimoto, and J. Kondou, 1999, Precast Segment Construction,
Tunnel Engineering Thesis, Collection of Reports, Volume 9. (Translation from Proceedings of
Tunnel Engineering, Japan Society of Civil Engineers, released on June 27, 2011, Japanese ver-
sion: www.jst​age.jst.go.jp/​arti​cle/​journa​lte1​991/​9/​0/​9_​0_​325/​_​pdf.)
Shield Tunneling Association of Japan, 2004. P&PC Segment Lining Method, accessed February
2012, www.shi​eld-​met​hod.gr.jp/​engl​ish/​
Szechy, K., 1966, The Art of Tunnelling, Akaemiai Kiado, Budapest, 1097 p.
Tauber Rohrbau GmbH & Co, Mini Tunnel and Tauber Tunneling with Segments Made of Polymer
Concrete. Germany.
Van Greunen, J., Y. Sun, G. Hughes, R.G. Mikola, J.Y. Kaneshiro, D. Haug, and M. Vanderzee, 2016,
“Technical Approach to Lining Design for High Internal Operating Pressure and Seismic Fault
Offsets on the JWPCP Effluent Outfall Tunnel”, International Tunnelling Association World
Tunneling Congress Proceedings, San Francisco.

TECHNICAL CONFERENCES
DRK European Tunnel Seminar No.6, “Segmental Lining Design–​ Design Solutions for Urban
Tunnelling Projects”, Wolf Friedmann and Jan Franzius, 22 Sep 2009.
2021 RETC Paper, “Performance of Tunnel Segments Reinforced with Synthetic Macro Fibers”, Ralf
Winterberg (Lead Author), pp. 837–​883.
2013 RETC Proceedings, “King County Uses New Shaft Technology on the Ballard Siphon Project”,
Kaneshiro et al., 2013, pp. 719–​727.
2016 Seventh International Conference & Exhibition on Mass Mining, Sydney, NSW.
2021 RETC Paper, “Composite Segmental Lining to Resist a Fault Failure”, Angel Del Amo, Bingzhi,
Yang, Robert Goodfellow, Girish Roy , pp. 444–455.

COMMERCIAL CONTACT INFORMATION

Barchip, www.barc​hip.com
Contec Fiber AG, www.cont​ecfi​ber.com
CTS Cordes GmbH & Co., www.cor​des.de
Herrenknecht AG, www.herre​nkne​cht.com
Optimas Solutions, SA, www.opti​mas.com
Sauerisen, Inc., www.sau​erei​sen.com
SolidCast Polymer Technology, www.sol​idca​stte​chno​lgy.com
Index

A+​B grouts, 644, 665 Annular grout, see also Grout equipment; Grout
Absorber pipes injection methods; Grout mixes; Mix
coupled, 818 design development
coupling of, 817 backfilling, 629
fitted in cage, 816, 817 classification
Accelerating admixtures, 271 by amount of cement, 632
Accelerators, 272, 654, 696 grout components, 633–​634, 650
Acid attack filling at crown, 677
cement hydration products, 714–​715 ground loosening, 631
concentrations of aggressive acids, 714 liner ring, 631
ACI 350 recommendations, 255–​256 objectives, 630
Admixtures, 654–​655 quality control, 694–​695
Advanced systems without drilling, 389 ring widths, 631
Aggregates rock tunnels, 631
coarse, 265 support, liner rings, 622–​623
compressible packing models, 267 TBM and segmental lining, 630
compressive strength, 264 TBM tail shield, 641–​643
fine, 265–​266 Anti-​washout admixtures, 655
normal weight steel fiber reinforced concrete, Arrowhead Tunnels Project, 49
265 ASTM C1609–​standard test method, 203, 216,
packing approaches, 266–​267 273, 274, 275, 277, 303–​309, 547, 548
project/​owner-​specific properties, 264 Asymmetrical dowel system, 364, 365
Air-​cooled slag, 263 Automated digital quality control tracking
Air entrainment admixtures, 269–​270 system, 573–​577
Air vibrators, 461, 462, 503 Automatic fiber dosing equipment, 318–​319
Albvorlandtunnel, Germany, 462, 503, 505 Automation, 494, 625–​626, 809
Alkali-​activated cementitious materials Axial force-​bending moment interaction
(AACM), 331, 503–​504 diagram, 151–​152, 203–​204, 211,
Alkali-​aggregate reaction (AAR), 272, 332, 213–​215
715–​716
Alkali-​silica reaction (ASR), 332, 728 Back-​calculation parameters, 274
Alluvial soils, 55 Backfill grouting, 47–​48
American Concrete Institute (ACI) method, 97, Banana bolts, 356
99, 734–​742 Barcelona indirect tensile strength test, 303
American Society for Testing Materials (ASTM), BarChip macro synthetic fiber, 784
93, 264, 584, 706, 763 Bar reinforcement, 33, 38, 229, 312, 324, 327,
The Anacostia River CSO project, 671 331, 339, 340
Anchored gaskets, 433–​436, 554, 781–​782 Barton–​Bandis joint model, 87
Anchored (non-​glued) guide rods, 367–​369 Batching and mixing equipment
Anchored guide rod system, 779 metropolitan area, 655
Anchored plastic dowel with steel center axis, one-​component grouts, 656–​657
774–​775 two-​component grouts, 657–​658
Anchor installation, 390 Batch plant, 470–​471, 665–​666, 673–​674
826 Index

Bay Area Rapid Transit (BART) system tunnels, Carousel segment production, 522–​524
18–​22 Carrousel and stationary production
Beam molds, 54, 547 arguments, 477
Beam-​spring method, 161–​164 with concreting chamber and curing tunnel,
Beam test report and parameters, 550 477
Bedded beam method, 161 crane-​based concrete distribution system, 477
Bentonite, 651, 653 maintenance, 478
Bi-​component grouts, 644 segment production facilities, 478
Bicones and center cones, 775–​776 Cast-​in channel systems
designated concrete socket, 366 anchor installation, 390
range of shapes and sizes, 365, 366 cast-​in-​place systems, 391
segment circumferential joint, 376, 378 curved, 391, 393–​395
shear resistance, 367 curved fastening system, 391, 392
with tie-​rods, 777–​778 embedded fastening systems, 386, 389, 390
Bituminous packers, 370–​371 high-​speed railway tunnel cross-​section, 392
Blacksnake Creek CSO Tunnel, 343, 784, 785 load directions, 391
Bleed test, 689, 690 serrated, 391
Boltless radial (longitudinal) joint connection Cast iron segment forming system, 41
systems, 379–​380, 774 Cast iron tunnel liner segments at St. Clair
Bored hard rock tunnels, 28, 29 Tunnel, 588, 589
Borings and test pits, 70 Cement hydration
Break-​out analysis, 194–​197 concrete gaining strength and durability, 257
The Brundtland Commission, 324 fly ash, 261–​263
Buchan, C. V. (Great Britain), 22 limestone, 264
Bucklock Sof-​Clip couplers, 361 metakaolin, 263–​264
Buckskin Mountains Tunnel, 28, 31 Portland cement, 257–​258
Buc-​Lock, 22 SCM, 258–​260
Building information modeling (BIM) silica fume (microsilica), 260–​261
crucifix joints, 117, 122, 124–​126 slag cement, 263
ring length optimization, 105, 115 tri-​and di-​calcium silicates phases, 257
slight size adjustment of one segment, Zeta potential, 257
117–​122 Cementitious products, 650–​651
Bursting forces, 140 Channel Tunnel Project, 369
Channel Tunnel Rail Link (CTRL) project, 321,
Camlock Radial Joint Connection System, 35 328
Canadian Standards Association (CSA) method, Charcon tunnels (Great Britain), 22–​24
742, 751 Chemical admixtures
Carbonation-​induced corrosion, 736 accelerating, 271
Carbon cure method, 229 air entrainment, 269–​270
Carbon footprint analysis components of, 268
aggregate quantities, 229 fresh state/​post-​hardening properties, 268
carbon cure method, 229 mechanisms, 268
cementitious materials, 228–​229 mixtures proportioning, 271–​272
CO2 emission primary and secondary functions, 268
drilling and blasting method, 232–​233 retarding, 270–​271
from electricity, 231–​232 specialty, 271
emission produced during blasting, 232, WRA, 270
233 Chloride-​induced reinforcement corrosion
fuel consumption, 230–​231 damaged/​corroded tunnels, 710
hydraulic break hammer, 233–​234 degradation in tunnels, 710
loading and transportation, 235 loss of reinforcement, 711
road header, 233 Circular rings, see also Building information
rock waste transportation, 235 modeling (BIM)
TBM, 234–​235 geometry of key segment, 114
fiber reinforcement, 229 internal diameter, 99–​102
global warming and GHG emissions, 228 length, 104–​106
Montreal Blue Line Extension, 236–​238 ring segmentation, 109–​111
Carousel casting system, 583 segmental ring systems, 105–​109
 Index 827

segment shape, 111–​114 accidental removal, TBM thrust jacks, 373,

thickness and outside diameter, 102–​104 374
Circumferential joint, 357, 405 bolting system, 371
Circumferential joints, 553 for gasket pressure, 372–​373
Civil & Building North America Inc. (CBNA), segment lifting and secondary grouting,
401 373, 375, 376
Classic rebar reinforcing, 525–​526 anchored gaskets, 554, 555
Classic steel fibers, 536 circumferential joints, 553
Cleaning equipment, 704 glued gaskets, 554, 555
Cleaning-​out lines, 704 grout/​lifting sockets, 554
Coarse aggregates, 265 radial joints, 553
Cohesive soils, 55 waterproofing gasket seals, 554
Collated fibers, 318 Consolidated drained (CD) triaxial compression
Collated (glued) vs. loose fibers, 318 test, 83
Colloidal grout batch plant, 674 Consolidated undrained (CU) triaxial
Combined Railroad Line and Express Highway compression test, 83
Adler–​Krasnaja Poljana, Sochi, Russia, Construction and Demolition (C&D) debris,
500f Channel Tunnel Rail Link 254
Combisegments®, 494, 498–​499 Construction loads, see also Localized back-​
Combisegment system, 789, 790 grouting; Tail skin back grouting
Comb-​shaped gasket design, 408 pressure; TBM back-​up load; TBM thrust
Commissioning process, 701 jack forces
Composite reinforcement design, 787 description, 131
Composite segmental tunnel lining, 790 dowels, 131–​132
Compressible backfill grout, 809 Contec bi-​component polymer fiber, 786
Compressible packing models, 267 Continuous post-​tensioned segment lining
Compressive strength, 692–​693, 695–​696 method, 794
Computer chips, 804 Control of service life
Concrete alkali-​silica reactions, 283
AACM, 331–​332 corrosion–​chloride diffusion, 282
batching and mixing, 551–​552 diffusion-​based sulfate attack, 282–​283
beams, 546 transport properties, 281
corrosion resistance, 809 Conventional access shaft construction,
cover, 200, 201 799–​800
curing, 552–​553 Conventional degradation mechanisms, see also
EFC, 332–​333 Acid attack; Alkali-​aggregate reaction
global carbon emissions, 331 (AAR); Stray current corrosion; Sulfate
heat conductivity, 183 attack
low-​carbon, 331 bored tunnels, 709
Cone penetration test (CPT) method, 74 carbonation-​induced corrosion, 712
“Conex” wood dowel system, 361 chloride-​induced reinforcement corrosion,
MWRA Effluent Outfall Tunnel Project, 770, 710–​711
771 freezing and thawing attacks, 716
waffle segment bolts, 770, 771 frost attack, 716
Connecting bolts reinforcement corrosion, 710
bolted connection, radial (longitudinal) joint, service life design, 731
359 Coring programs, 696–​697
corrosive applications, 360, 361 Corner design
engineered plastic sockets, 356 with adjustment to distorted angle, 430,
force exerted by gasket, 357, 358 431
HDPE socket creeps, 357 criterias, 424, 426
precast segment bolting materials, 358 deviations drom 90° gasket frame corners,
water conveyance system, 356 427, 429
Connecting dowel, 616 distorted angle on key segment, 430
Connecting dowels frame with an angle of 90°, 426, 427
and bicone innovations, 773–​774 gasket frame corner, 426, 428
Connection systems segmental ring–​rectangular segmental ring
and accessories arrangement, 427, 428
828 Index

segmental ring–​rhomboidal segmental ring Drilling and blasting method, 232–​233

arrangement, 427, 428 Drone surveillance, 53
solid gasket frame corner, 426, 427 Dunnage, 594–​596
universal rhomboidal segmental ring system, Durability parameters, 279–​281
427, 429
Corrosion-​resistant coating, 790 Eagle Mountain–​Woodfibre Gas Pipeline
Coupling multi-​degradation factors, 726, Project, Squamish, BC Canada, 802–​803
729–​731 Early applications
Crack mouth opening displacement (CMOD), Holland Tunnel, New York City–​1919
309, 310 concept proposal, 14–​17
Cracks, 295 Mount Royal Tunnel, Montreal, Quebec–​
Craft training 1912 to 1916, 11–​14
carousel casting system, 583 Earth Friendly Concrete” (EFC®) geopolymer
concepts, 583 concrete, 332
stationary mold system, 583 Earth materials
Creep models, 90 hydrogeology and water pressures, 66–​67
Crown check (contact) grout, 684 IGM (hard soil/​soft rock), 61
Crucifix joints, 113, 117, 124–​126 organic soils, 57, 61
Curved cast-​in channels rock, 61–​66
channel fastening system, 391, 392 seismic considerations, 67–​69
high-​speed railway tunnel, 391, 393, 395 soft ground
segment mold, 391, 393 ASTM D-​2487 and ASTM D-​2488, 57–​59
segments with, 391, 394 classification methods, 56–​57
system, 391, 394 cohesive soils, 55
Cut-​and-​cover tunnels, 67 materials, 55
naturally deposited soils, 55–​56
DAUB:2013 formulas, 140 non-​cohesive soils, 55
Deer Island Effluent Outfall Tunnel Project, 28, Tunnelman’s soil classification and behavior
32 descriptions, 57, 60
Deformed wire products, 526–​528 Earth pressure balance (EPB), 32
Delayed ettringite formation (DEF), 714 Elastic equation method, 156, 159–​161
Demolding, 479 Elastic modulus, 87
Density, grout, 685–​687 Electrical vibrators, 462
Department of Transportation (DOT), 18 Embedded fastening systems, 390
Design memoranda (DM), 90–​91 Enbridge Line 5 Tunnel, Traverse City,
Deutsche Bahn Albvorland Tunnel near Michigan, USA, 801
Stuttgart, Germany, 446–​451 EN 14651–​test method, 308–​312, 549
Digital twin of physical ring, 486 Environmental product declaration (EPD), 339,
Dilatometer (DMT), 82 340
Direct uniaxial tension test, 303 Environmental testing, 71
Discrete element method (DEM), 156, 164–​167 Equipment installation, 700
Discrete macro fibers, 273 Equipment selection, 700
3D non-​linear FEA (NFEA) simulation Ethylene-​propylene-​diene-​monomer (EPDM)
crack width dimensions, 144, 146 elastomer, 407
fiber-​reinforced concrete, 147 aging process, 416
fine mesh, 144, 145 characteristics, 416
transverse tensile stresses, 144, 146 CR/​SBR, 418
Double-​barrier sealing system, 445 physical property material tests, 418, 419
Double punching test (DPT), 303, 320 typical gasket installation, 416, 417
Double ring beam-​spring model, 162 European norm (EN) method, 731–​735
Dowels Explosive spalling, 321–​323
connector, 360–​361 External (confining) pressure tunnels
mechanical, 363–​366 continuous hoop rebar, 792, 793
Paris Line-​12 Project segmental tunnel lining, diameter effluent tunnels in Japan, 793–​794
361, 363 internal water pressure, 791–​792
sof-​clip friction, 361, 362 Keiser method, 792
sof-​fix dowel engaged in socket, 361, 362 External sulfate attack damage process, 713, 714
tunnel projects, 361 External vibrators, 461
 Index 829

Fabricated steel tunnel linings Flexural ASTM C1609 test, 274

BART system, 18–​22 Flood and contact grout injection
WMATA system, 19 crown check (contact) grout, 684
Fastening systems grout delivery system to heading, 682
applications of, 385 grout mix design development, 682
cast-​in channel systems, 390–​395 grout volume measurements, 684
categories, 386 location and sequence of, 682–​684
drilling Florida Department of Transportation (FDOT),
advanced systems without, 389 401
traditional anchor systems with, 386–​389 Flowability, viscosity and stabilization time,
precast segments, 384 687–​689
Fast lock dowel and socket assembly, 36, 37 Fly ash, 261–​263
Fast Lock Dowel System, 35, 361 Force Majeure events, 513
Fastlock steel dowel system, 773 Forklifts, 595–​597
Fault crossing, 178–​181 Freezing and thawing attacks, 716
Federal Railway Administration (FRA), 18 Fresh and hardened grout
Fiber dosing and metering system, 538–​540 bleed and segregation, 647
Fiber reinforced concrete (FRC), 272–​273, defined, 647
see also Full-​scale testing; Material flowability and viscosity, 647
sampling; Steel fiber reinforced fluid and installed properties, 648–​650
concrete (SFRC) gel time, 648
ductility, 273–​275 stabilization time, 647
evolution of, 293–​294 UCS, 648
reduced carbon footprint, 326–​327 Fresh state, rheology tests and workability,
steel fibers, see Steel fibers 277–​279
sustainability assessments, 325–​326 Frost attack, 716
types of, 294–​295 Full-​scale pull-​out test, 375
Fiber reinforcement, 202–​203 Full-​scale tests, 303
Fiber wash-​out test, 545 bending test, 312
Field pressuremeter, 82 cantilever load tests, 219–​221
Final service stages half of circular segmental lining, 219, 221,
long-​term loads, 155 222
Fine aggregates, 265–​266, 654 instrumentation, 217, 218
Finite element method (FEM), 164–​167 instrumentation plan, 217, 218
Fire incident and explosion load testing, 312
actual time-​temperature curve, 180 point load tests, 219, 220
average residual tensile stress decay factor, 191 setup, 217
compressive strength, 188 SFRC precast segments, 312
concrete heat capacity, 184 thrust load, 312–​315
excluding areas, 183 Full-​scale tunnel segment testing, 276
fire loading, FEM model, 182 Fully automated subsystems, 626
first peak tensile stress decay factor, 191
hoop stress variation along lining thickness, Gantry cranes, 469, 472–​475
192 Gas and oil pipelines
hoop stress vs. distance from lining extrados, benefits, 801
193 Eagle Mountain–​Woodfibre Gas Pipeline
internal radial pressure, 194 Project, Squamish, BC Canada, 802–​803
nodal temperatures, 184–​186 Enbridge Line 5 Tunnel, Traverse City,
peak compressive strength, 189 Michigan, USA, 801
post-​crack residual flexural strength, 190 Kinder Morgan Burnaby Mountain Tunnel
standard fire load curves, 180, 181 Project, Burnaby, BC, Canada, 802
temperature history, 187 Gasket systems, see also Anchored gaskets;
temperature profile, 188 Corner design; Glued-​on gaskets; Groove
tunnel and project-​specific conditions, 180 design; Load-​deflection behavior;
tunnel intrados, 181 Segment gaskets; Stress-​relaxation
ultimate compressive strain capacity, 189 cast iron segments, 407, 408
Fire resistance, 430–​433 chloroprene elastomer, 407
Flat corners, 418 comb-​shaped gasket design, 407, 408
830 Index

corrosion protected segments, 445 non-​cohesive soils, 83–​84

design criteria, 411 rock strength, 84–​87
EPDM, 416–​419, 432, 444–​445 soil, 82
extrusion machine, 414, 415 Glacial soils, 56
green soap, 436 Glass fiber-​reinforced polymer (GFRP) rebars,
groove filling (area utilization in groove), 412, 787–​788
413 cage, 342
hollow-​chamber gasket design, 407, 409 reinforced concrete segment, 342
hybrid composite gasket designs, 442, Glued gaskets, 554
444–​445 Glued-​on gaskets, 430, 433, 435
key segment installation and insertion, 436, Glued steel fibers, 538–​539, 541–​543
437 The Grand Paris Express project
lubrication, 438 Forrestfield Airport Link, 337–​338
mono-​compression caskets, 445 Montreal Metro Blue Line Extension,
ordinary lead, 407 335–​337
profile area, 412 Paris Metro Line 16–​1, 333–​335
profile in complete compression, 411, 412 Grand Paris Subway Program, Paris, France,
profile in contact, 411 398–​400
quality control, 441–​443 Greathead, James Henry, 1
repair, 438 Patent GB 1738 of 1874, 4, 5
sealing profile geometries, 410 remnant, 5, 6
surface friction, 436 tunneling shield, 5, 6
typical gasket frame layout, 415, 416 Great Lakes Clean-​Up Program, 33
typical state-​of-​the-​art gasket profile, 411 Green concrete for sustainable construction,
vulcanization press, 414, 415 807–​808
Gauge rail car, 658 Greenhouse gas (GHG) emissions, 324–​325
Gel time, 691–​692 Groene Hart Tunnel, 151, 154
General design procedure, 99 Groove design
Geological strength index (GSI), 85, 86, 164 groove dimensions, 413
Geophysical tests, 82 with sloped edges, 414
Geotechnical baseline report (GBR), 91–​92 width, 412
Geotechnical data report (GDR), 91 Grooved pipe coupling assembly, 663
Geotechnical interpretive report (GIR), 91 Ground granulated blast furnace slag (GGBFS),
Geotechnical reporting 263
DM, 90–​91 Ground investigation process
GBR, 91–​92 construction stages, 71
GDR, 91 CPT and SCPT, 74, 75
GIR, 91 detailed design stages, 71
Geotechnology, see also Earth materials; horizontal borings, 73
Geotechnical reporting; Ground planning stages, 70
investigation process remote sensing and geophysical methods,
computer programs 74–​75
creep models, 90 rock coring methods, 75–​77
design of segmental tunnel linings, 87 in situ testing in soil and rock, 76–​81
non-​linear elastic modulus constitutive subsurface, 70
models, 89 vertical boreholes, 71–​73
rock models, 88 Ground structure interaction (GSI) model, 69
soil hardening models, 89–​90 Grout equipment, see also Batching and mixing
soil models, 87 equipment; Pipeline
spring constants, 88–​89 phases, 655
defined, 53 transporting backfill grout, 658–​659
design and factors critical to construction, 54 Grouting, 369–​370
designer's focus, 53 Grout injection methods
desktop data study, 69–​70 close-​up of, 640, 641
shear strength parameters development of method, 643–​644
cohesive/​granular soils, 82–​83 precast segments, 639–​641
elastic modulus, 87 pressure readings, 638
field evaluation tools, 82 selection of factors, 644–​645
 Index 831

tail seal system, 638 History, see also Early applications

TBM grout ports and excluder plates, 638, early precast, 3, 4
639 fabricated steel tunnel linings, 18–​20
TBMs, 637 generalized timeline, 2, 3
Grout-​lifting socket pull-​out test, 376 O’Rourke, John Francis, 7–​8, 11, 42
Grout/​lifting sockets, 554 Parmley, Walter C., 4–​7
Grout ‘Mix’ design development 699, see also patent of Mattson, A. F., 1, 2
Fresh and hardened grout tunnel shield design and patents
admixtures, 654–​655 Barlow, Peter W., 4
bentonite, 651, 653 Greathead, James Henry, 4–​7
cementitious products, 650–​652 Hoek–​Brown failure criterion, 85, 164
fine aggregates, 654 Hollow-​chamber gasket design, 407, 409
iterative process, 650, 651 Horizontal borings, 73
performance requirements, 646–​647 Hose bursts, 702
Grout mixes Hoses, 704
one-​component active grout, 634 Hybrid composite gasket designs
one-​component inert grout, 634–​635 AFTES, 444–​445
one-​component semi-​inert, 634–​635 co-​extrusion process, 442
pea gravel and flood grout, 637 hydrophilic cord, 444
two-​component active, 635–​637 segment gasket, 442, 444
Grout pressures, 676 self-​healing-​effect, 444
Grout spills, 704 Hybrid solutions
Guide rods, 401, 617 circumferential post-​tensioning of ring, 343
anchored (non-​glued) guide rods, 367–​369 GFRP bars, 341–​343
circumferential joint, 367 reinforcing steel bar and welded wire mesh,
radial (longitudinal) joint, 367 340–​341
radial joint alignment, 778–​779 Hydration control additives, 654
radial (longitudinal) joint detail, 404 Hydration of cement, see Cement hydration
segment guide rod and groove, 367 Hydraulic break hammer, 233–​234
Hydraulic operated piston pumps, 661–​662
Hamster cages, 778 Hydrophilic (swelling) seals, 780
Handling devices
customized, 479, 480 Index testing, 70
demolding, 479 Indirect “special” tension test, 303
equipment, 481 Individual segment geometry
with gantry cranes, 482, 483 hexagonal segment system, 111
outside storage yard transport, 482 issues, 112
rotation/​turning, 481 rectangular segment system, 112
transport in hall, 481 rhomboidal segment system, 111, 113–​114
transport out of hall, 481–​482 trapezoidal segment system, 113
Hardened concrete testing of fibers, 547 Inert mortar grouts, 632
Hardened state testing, 320 Inner Doha Resewage Implementation Strategy
HDPE membrane system, 383, 789, 790 (IDRIS) Project, 384
Health and safety provisions In situ physical testing, 71
backfill grouting, 47–​48 Intact rock, 84–​85
precast segmental tunnel linings, 43, 48 Interlocking tunnel blocks, 7–​8, 11
Herrenknecht VSM Technology, 380, 383 Intermediate geomaterials (IGM), 61
Hexagonal segment system, 111 Internal sulfate attack, 714
High-​range water reducers (HRWR), 654–​655 Iyengar Diagram, 140, 141, 144, 169
High strength welded wire fabric (WWF),
782–​783 Janbach, Austria Pilot Scheme, 815
High-​tech 3D measurement systems Janssen joint model, 162, 163
baseline mold tolerance set, 558 Japanese steel radial (longitudinal) joint
mold fabrication templates and jigs, 558 connector system, 380
mold manufacturer, 558 Joint bars, 341
molds and segments, 557 Joint connection innovations, 769–​770
proofing molds, 558–​564 Joint gap measuring tool, 614
segment tolerances, 559, 565 Joint packing materials, 370–​371
832 Index

Junk liners, 514 Lovat tunnelling equipment and segment erector

“Junk segments” for tunnels, 27, 28 (Canada), 25–​27
Low-​point pump station (LPPS), 101
Keiser method, 792
Kinder Morgan Burnaby Mountain Tunnel Macro synthetic fiber (MSF), 295, 784–​785
Project, Burnaby, BC, Canada, 802 Macro-​synthetic fiber reinforced concrete
Kinnear Moodie & Company (Scotland and (MSFRC) segmental lining, 295
Great Britain), 20–​22 Macro synthetic/​polymers, 343–​344
Koralmtunnel KAT 2, Austria, 504, 506 Master ring dimension tolerances, 485
Material sampling
Laboratory testing, 71–​72, 757 ACI, 319
LACSD flat anchor duct system, 795 fresh state sampling and testing, 319–​320
Lacustrine deposits, 55–​56 hardened state testing, 320
Life cycle assessment (LCA) software, 338, 807 Materials tests, 216–​217
Lifting and hoisting of segments Material testing levels, 275
forklifts, 595–​597 Mattson, A. F. (1906 patent), 515
Nylon slings, 596 Maximum deformation method, 175–​176
pickers, 596–​598 Mechanical dowels
Lignosulfonates, 654 installation of segment, 363
Limestone, 264, 337 pull-​out and shear resistance, 363
Limit state design (LSD), 97 pull-​out resistance, 364, 365
Linear elastic finite element analysis, 141, segment alignment, 363
143–​145 shear resistance, 364
Lining systems, 795–​798 speed of installation and safety, 365, 366
Load and resistance factor design (LRFD) Mechanical erector, 369
method, 97, 223 Mechanical interlock principle, see Cast-​in
Load cases and factors channel systems
ACI design codes, 99 Mechanical lifting, 603–​605
construction stages, 98 Metakaolin, 263–​264
final service stages, 98 Micro-​PP fibers, see Passive fire protection (PFP)
other loads, 98 Microtunneling, 820
production and transient stages, 98 Mid-​century assessments and tunnel planning
required strength and resistance, 98–​99 reports
ultimate strength and resistance, 97, 98 tunnel lining research activities, 17–​18
Load-​deflection behavior Milwaukee Metropolitan Sewage District
diagram, 423, 425 (MMSD), 32
pressure-​path diagram, 423 Minimum recommended concrete cover,
short-​term relaxation, 423, 426 200–​201, 734, 747
STUVA recommendations, 423 Mitigation methods (stray current)
TBM thrust jacks, 424 connection plate, reinforcement cages, 721,
test, 441 722
testing machine, 423, 425 current field and equipotential surfaces,
Localized back-​grouting 718–​720
defined, 150 degradation mechanisms, tunnel linings,
design check, 152, 154 726–​728
modeling, 152, 153 examples, 718
results, 152, 153 general arrangement, 722, 724
segmental tunnel linings, 150–​152 short steel bars, 722, 725
Location and sequence, gravel injection, steel bars, 722, 724
678–​681 steel fibers, 722, 724
Lock-​out/​tag-​out station, 582, 700, 705 stray current mitigation, 718, 720, 721
Logistics of supply, 577–​581 stray current paths, 725
Longitudinal bars, 198–​199, 341 Mixed-​face tunneling, 12, 72
Longitudinal joint, 9, 112–​114, 117, 167–​170, Mixtures proportioning (for concrete), 271–​272
356, 357, 359, 617; see also Radial Model validation, 256
Joints Modern anchored gasket profiles, 782
Longitudinal joint bursting load, 167–​171 Modern gasket profiles with molded corner,
Loose steel fibers, 543 781, 782
 Index 833

Modern precast concrete segment, 511, 512 Operator-​assist systems, 626

Mold manufacture Organic soils, 57, 61
assembly, 464–​465 O'Rourke, John Francis, 1, 7–​8, 11
inspection, 466 interlocking tunnel block casting system, 42
mechanical processing, 463, 464 segment casting form, 41, 42
painting, 467 Outside storage yard, 472–​475
steel fabrication, 463 Overcut parameter, 104
test ring construction, 465–​466 Overhead cranes, 45–​46, 469, 475, 482, 583
transport, 467 Overhead precast segment handling, 48
Mold QR code tag, 574 Owners and tunnel design engineers, 767
Molds and production systems, 40
Mono-​compression caskets, 445 Pan type planetary mixer, 657
Mono-​EPDM-​compression-​gaskets, 445 Parametric study
Montreal Blue Line Extension reinforcement ratio, 212
CIP section-​baseline case, 236–​238 steel yield strength, 213
CIP section-​low carbon case, 237, 239, 240 yield strain, 212
PLB project, 236 Paris Line-​12, 363
TBM tunnel, fabricated segmental lining, Parmley system, 1, 7–​11, 40, 811
237–​238, 241, 242 Passive fire protection (PFP)
Multi-​service vehicles (MSV), 482, 600, 601, explosive spalling, 322
658 fiber and dosage recommendations, 323
MWRA Effluent Outfall Tunnel Project, 770, history on, 321–​322
771 Patented light weight cores, 806
Patented radial (longitudinal) joint connector,
The National Environmental Policy Act (NEPA), 382
324 Pea gravel and flood grout system, 637
National Fire Protection Association (NFPA), Pea gravel angle of repose, 681
100, 819 Pea gravel delivery system, 678–​679
National Ready Mix Concrete Association Pea gravel placing system, 679
(NRMCA), 657 Peak ground accelerations (PGAs), 67
Naturally deposited soils, 55–​56 Peak ground velocities (PGVs), 67
Non-​circular applications, 810 Penetration test (CPT) method, 82
Non-​cohesive soils, 55, 83–​84 Penetrometer, 690, 691
Non-​corrosive radial (longitudinal) joint Peristaltic pumps, 659–​660
connector system, 381 Permanent bolting system, 360
Non-​linear elastic modulus constitutive models, Personnel training, 698–​699
89 Physical mastering build, 486
Northeast Ohio Regional Sewer District Pickers, 594, 596–​598, 600
(NEORSD), 33–​34 Pipe couplings, 702, 703
Norwegian Geotechnical Institute (NGI), 63 Pipeline
Nylon slings, 596 cleaning pig, 664
hydration controlling admixture, 659
Old school paper trail method, 573, 576 hydraulic operated piston pumps,
“One Click LCA” platform, 338 661–​662
One-​component active grout, 634 PCP, 660–​661
One-​component grouts, 656–​657 peristaltic pumps, 659–​660
batch plant, 665–​667 tunnel grout pipeline, 662–​664
delivery to heading, 666, 667 Plasticizers, 147, 654
grout injection, 668–​670 Plywood packers, 370
grout issues and resolution, 670–​672 Pneumatic vibrator, 462
trailing gear equipment, 666, 668 Point load strength index (PLSI) testing, 85
One-​component inert grout, 635–​636 Polycarboxylate superplasticizers, 655
One-​component semi-​inert, 634–​635 Polymer concrete pipes and manholes, 805
One-​part grouts, 644 Polymer rubber seals, 780–​781
One-​pass Polypropylene fiber, 39, 316, 322, 332, 352
tunnel lining, 34–​5, 97, 197, 200, 227, 514, fire resistance benefits, 785, 787
767 typical matrix, 786, 787
tunnel lining mock-​up assembly, 35 Pop-​up stores, 513
834 Index

Portland cement, 236–​239, 257–​259, 262–​264, PVC/​HDPE membrane, corrosion resistance,

331–​332, 714, 806 383–​386
Portland cement industry, 254
Port of Miami Highway Tunnel, Miami, Florida, QR identifier tag, 575
USA, 398–​405 Q-​system, 63, 65, 66
Post-​installed anchors, 386, 388 Quality assurance and quality control (QA/​QC)
Post-​installed assembly channels procedures
adjustability and flexibility, 393 automatic fiber dosing equipment, 318–​319
bolted in-​place channel fastening system, 394, collated fibers, 318
396, 397 documents, 315
tension rod suspension system, 397, 398 “Made in the USA” certificate of compliance,
Post-​tunnel construction, 789 317
Pot life, 647, 682, 692 micro-​synthetic PP fiber material compliance
Pre-​bored pressure meter test (PMT), 76 requirements, 316
Precast plant labor crews, see also Craft training placing and finishing segments, 319
plant safety considerations, 582 precast plant, 315
precast shifts, 582 steel fiber material compliance requirements,
skills/​skillsets, 582 316
Precast polymer concrete segments, 804–​806 Quality documentation system
Preconstruction testing, 285–​287 documentation and process monitoring, 490
ASTM C1609–​standard test method, production and logistics management,
306–​308 490–​491
ASTM standard, 304, 305 segment labeling, 489–​490
beam testing, 303, 304
conventional concrete tests, 302 Radial joint alignment, 769, 778–​779
deflection hardening SFRC, 305 Radial (longitudinal) joint connector, 381
direct uniaxial tension test, 303 Radial joints, 41, 341, 553, 793, 797, 801–​802;
EN 14651–​test method, 304–​312 see also longitudinal joints
full compaction, 306 Radial/​longitudinal joint, 357
full-​scale testing, 303 Radio frequency identification (RFID), 804
indirect “special” tension test, 303 Rad-​link radial joint connection system,
serviceability, 304–​305 776–​777
small-​scale flexural testing of beams, 303 Rail borne set-​ups, 599, 600
Primary curing period, 552 Railroad and subway tunnels, 99
Production and transient loadings Rebar cage 528, 529, 730, 787, 817
description, 122 Rebar cage design vs. segment mold design, 528,
design checks and factors, 130–​131 529
segment handling, 129–​131 Rebar chair systems, 529–​530
segment storage, 123–124, 127, 481–​483, Rebar management and assembly shop,
502, 556, 591, 593 532–​536
segment stripping, 98, 122, 126, 556 Rebar reinforced concrete segment, 341
segment transportation, 127, 128 Rebar-​type reinforcing, 524–​525
Programmable logic controller (PLC) controlled Rectangular segment system, 112, 117
grouting system, 696 Reduced carbon footprint, 293, 326–​327, 785;
Progressive cavity pumps (PCPs), 660–​661, 664 see also carbon footprint analysis
Project-​specific gasket performance testing Reduced flexural rigidity, 156
load-​deflection-​behavior test Reinforcement corrosion, 710
gasket frame corners, 441 Reinforcement spacing, 201–​202
on gasket profile, 439–​440 Reinforcing design, 528
proof of suitability, 439 Remote sensing, 53
short-​term waterproofing tests, 439 and geophysical methods, 74–​75
stress-​relaxation tests, 441 Repair/​replacement of segments, 759–​762
Proof/​secondary grouting, 370, 639; see also Residual soils, 56
localized back-​grouting (secondary Retarding admixtures, 270–​271
grouting) Retrofitting and rehabilitation of tunnels,
PSTL marker plate, 804 810–​815
Public-​private-​partnership (PPP) construction Rhomboidal segment system, 113–​114
model, 401 Riachuelo Shaft Site and Grout Batch Plant, 666
 Index 835

Ribbon mixer, 659 upward and downward turns/​curves, 107

Ring assembly aids, 616, 617 Segment bolting/​dowel systems, 377–​379
Ring builder’s assistant, 618, 620 Segment demolding, 556
Ring damage, 621–​622 Segment erector, 7, 12, 14, 25, 603, 618–​619,
Ring designs 768–​769
defined, 457 Segment fabrication plant layout
tunneling method, 457 carousel, 518–​520
types, 458, 459 collection of systems, 517
Ring erection cycle, 613–​616 “Segment Factory 4.0” (automated segment
Ring joint, 162 production)
Ring segmentation application of release agent, 495
5+​1 arrangement, 110 cleaning of mold, 494, 496, 497
handling and erection process perspectives, cost-​intensive personnel, 494
109 injection sockets, 495, 497
hexagonal, developed plan view, 110, 111 opening and closing of mold, 494, 496
and segmentation, 109, 111 robot-​supported working line, 494, 495
segment slenderness, 109 safety installations, 498
Ring types, 589 Segment factory/​field factory, 469
Riyadh Metro Line 1, Saudi Arabia, 501–​503 Segment feeder, 607
Road header, 233 Segment formwork with membrane, 385
Road tunnels, 99, 101–​102 Segment gaskets
Robotic arm, segment unloading station, 626 defined, 407
Rock groove with gasket profile, 407, 410
coring methods, 75–​76 typical concrete tunnel segment, 407, 409
defined, 61 watertightness, 407
mass parameters, 85–​87 Segment handling, 47–​48, 129–​131, 556, 591,
micro-​ and macro-​behavior, 61 593, 602–​603, 624–​626, 728, 787
models, 88 Segment inspection and receiving report, 591,
Q-​system, 63, 65, 66 592
RMR, 63–​65 Segment joints, 357
RQD, 62–​63 Segment lifting, 373, 375, 376, 595
waste transportation, 235 Segment mold
Rock mass rating (RMR) system with concrete base and wood sidewalls, 516
classification parameters and ratings, 63, 64 with rebar cage, 517
mass strength and support requirements, 63 with QR identifier tag, 575
parameters, 63 steel segment forms, 516, 517
rating adjustments, joint orientations, 63, 65 Segment mold design, see also Carrousel
rock classes determination, 63, 65 and stationary production; Handling
rock mass classes, 63, 65 devices; Mold manufacture; Quality
Rock quality designation index (RQD), 62–​63 documentation system
2D/​3D model, 458
Sample segmental ring build report, 615 defined, 458
Sanborn map collection, 70 dimensional stability, 462–​463
San Francisco Municipal Transit Authority external vibrators, 461
(SFMTA) Central Subway Tunnel Project, and formwork dimensional tolerances, 488,
787 489
Schedule affects, 512–​513 implementation of, 460–​461
Secondary curing period, 552–​553 manufacturer, 458–​460
Secondary grouting, 373, 375, 376 quality control, 491–​493
pressure, see Localized back-​grouting; Proof ring designs, 457–​459
grouting serviceability, 462
Securing rebar to mold–​spear bolt, 531, 532 types of ring geometry, 457
Securing sensing equipment, 530, 531 Segment precast plant, 14, 393, 540, 582
Segmental lining bending moments, 167 Segment producers, 461
Segmental ring systems Segment production location, 588–​589
and outer diameter, 105, 106 Segment production plant
schematics of, 106–​108, 112–​114 factory/​field factory, 469
universal rings, 108–​109 factory with storage yard, 468
836 Index

infrastructure, 475 flexural crack width in segments, 225–​226

mobile batching plants, 470–​471 maximum allowable crack width, 226–​227
outside storage with gantry cranes, 472–​475 segmental tunnel lining systems, 223
parameters, 468 stress verification, 224
production quantities, 468 verification, tunnel segments, 223
quality control testing, 471 Service life design
redundancy, 476 conventional degradation, 731
reinforcement production factory, 471 Green Heart Tunnel, 756–​759
safety and rescue concept, 475 reinforcement corrosion, 750–​755
utility systems, 475 Set retarders, 272, 654
Segment quality control, 570, 572, 573 Shaft linings, precast segmental, 27, 28, 32,
Segment sealing, 14 799
Segment stacking for storage, 123–124, 127 Shanghai Metro–​Double-​O tunnel, 814
Segment stack unloading, 608–​609 Shear recovery systems, 194–​197
Segment storage, 123–124, 127, 481–​483, Short-​term relaxation, 372, 423, 426
502, 556, 591, 593 Short-​term waterproofing tests, 439
Segment stripping, 27, 98, 122, 126, 556 Shotcrete pot under hopper, 679
Segment tag, 574 Silica fume (microsilica), 237–​240, 260–​261
Segment tolerances, 567–​572 Site layout, 699–​700
Segment transportation, 127, 128, 556 Slag cement, 263
Sehulster fast lock dowel system, 773 Small-​scale flexural testing of beams, 303
Sehulster Tunnels Sof-​Clip friction dowel, 362, 774
one-​pass precast tunnel linings, 34 Sof-​Fast dowel system, 776
segment reinforcing, 33 Sof-​Fix anix dowel system, 775
Seismic cone penetration test (SCPT) method, 74 Sofrasar Sof-​Shear 375 Miami Bicone, 403
Seismic considerations Soft ground tunnels
design earthquake motions, 68 ASTM D-​2487 and ASTM D-​2488, 57
fault rupture, 69 classification methods, 56–​57
prediction of response, 68 cohesive soils, 55
site characterization, 67–​68 materials, 55
Seismic loads naturally deposited soils, 55
axial and curvature deformations, 177–​178 non-​cohesive soils, 55
bored tunnels, 172 Tunnelman’s soil classification and behavior
closed-​form solutions descriptions, 57
ground-​tunnel interaction, 174 Soil hardening models, 89–​90
lining demand, 174–​175 Soil models, 87
lining ovality demands, 172 Solid plastic friction lock dowel system, 774
maximum free-​field ground deformation, Spalling mechanisms, 284–​285
173–​174 Special robust vibration tables, 546
fault crossing, 178–​181 Specialty admixtures, 271
numerical analysis Specific laboratory testing of samples, 71
dynamic time-​history analysis, 177 Specific water well testing, 71
maximum deformation method, 175–​176 Spring constants, 88–​89
pseudo-​dynamic time-​history analysis, 176 Stacking finished segments, 594
typical FEM, 175, 176 Standard penetration tests (SPTs), 82
tunnel deformation types during earthquake, Start-​up process, 701
172, 173 Stationary mold production, 521–​523
Self-​healing-​effect, 444 Statistical process control procedures, 277
Semi-​inert mortar grouts, 632 Steam curing, 200–​202, 522–​523, 553, 803
Serviceability approach, 255 Steel bolt connectors, 772–​773
Serviceability limit state (SLS) design, 99, Steel fiber balling
223–​226 during casting, 541
cracking verification, 225 example, 541
current crack width criteria, verification of, large ball, 542
227–​228 multi-​balls, 542
deformation verification, 224–​225 Steel Fiber Reinforced Concrete (SFRC), 18, 38,
design checks and limiting values, 223, 224 43, 256, 293–​308, 311, 312, 314, 320,
design flowchart for, 223 323, 324, 326–​331, 333–​337, 340–​344,
 Index 837

399, 514, 537, 547–​551, 721, 725, 726, deformations, 152, 157
see also Preconstruction testing design codes, 154
characteristics of, 298–​302 Groene Hart Tunnel, 151, 154
durability benefits, 328–​331 shear forces, 152, 158
economic benefits, 328 TBM shield, 151, 152
Steel fibers TBM Tunnel Project, 152, 155, 156
appropriate fiber usages, 297 TBM launch, 623–​624
attributes effecting performance, 298 TBM tail shield
dimensional tolerances, 298, 299 differential thrust cylinder extension, 612
dispersion of fresh concrete, 543–​545 liner ring, 610–​611
glued, 538–​539, 541–​543 ring positions, 612–​613
logistics and storage of, 537 steering, 609–​610
loose steel fibers, 543 tunnel alignment ahead, 611–​612
vs. macro-​synthetic fibers, 295–​297 TBM technology
metering of, 538–​543 design improvements, 768
precast segments, 296–​298 guidance systems, 768
reception of, 537–​538 innovations, 768
recommendations, 298, 537 segment erector technology, 768
reinforced concrete, 551 vacuum lifting, 768–​769
reinforcement, 783–​784 TBM thrust jack forces, 373
Storage yard accidental removal, 132
layout, 591, 593, 594 3D NFEA program, 144
segment stack, 590–​591 eccentricities and load factors, 139
Stray current corrosion, see also Mitigation Iyengar diagram, 141, 142
methods linear elastic finite element analysis, 141,
corrosive effect, 718 143–​145
defined, 717 nominal machines thrust, 136
electric trains, 716–​717 radial section cut through circumferential
rail tunnels, 716 joint, 138
train catenary stray currents, 717 rendered 3D view, 137
Stress-​relaxation tests, 441 rock thrust, 135
defined, 422 simplified equations, 140–​141
diagram, 423, 424 SLS of cracking, 136–​137
STUVA recommendations, 422, 423, 447 stress distribution under, 137
test rail assembly, 423 transverse view, circumferential joint, 139
tests rails for gaskets, 422 TBM trailing gear
Sulfate attack dependency, 602–​603
external sulfate attack damage process, 713, individual segment unloading, 606–​608
714 mechanical lifting, 603–​605
water-​soluble sulfate ions, 713 segment stack unloading, 608–​609
Sulfonated naphthalene formaldehyde segment supply and unloading, 602
condensates (BNS), 654 vacuum lifting, 603, 605, 606
Supplemental geotechnical borings, 71 Technical developments in modern era, see also
Supplementary cementitious materials (SCM), Health and safety provisions
258–​260 concrete mix designs, 38
Sustainability assessments, 325–​326 logistics of supply, 43–​46
reinforcing materials, 38–​39
Tail skin back grouting pressure segment molds and production systems,
AASHTO DCRT-​1-​2010, 148 39–​42
backfilling, tail skin void, 148 Temporary “full-​moon” support frames, 195
design check, 151 Temporary “half-​moon” support frames, 196
grout pressure, 147–​148 Temporary steel bracing, 194
model, 149 Tension rod suspension system, 397, 398
results, 150 Test batch preparation, 685, 686
tail void back-​grouting, 149 Testing equipment and apparatus, 685
TBM back-​up load Test rail assembly, 423
axial forces, 152, 157 Test rings
bending moments, 152, 158 construction, 465–​466
838 Index

digital 3D segment models, 567 steam curing, 200–​201

precaster’s yard, 566 strain and stress distributions, 203
slop, 566 Tunnel diameter and lining thickness, 103
types, 565 Tunnel grout line plugs and clean-​out, 704
3D segment measurement Tunnel grout pipeline, 662–​664
instrumentation and analysis, 483–​484 Tunnel heading, 674–​675
regime, 484–​485 Tunnel lining, 26–​27
Through segmental tunnel linings, 170, 172 references, 344–​352
Thrust cylinders, 618, 619 Tunnel relining operations, 815
Tie-​rods, 777–​778 Two-​component grouts, 657–​658
and bicones, 377
socket with color coded, 378 U-​bolt connection, 772
tunnel segmental ring installation sequence, Ultimate limit state (ULS) design, 98, 141, 308
379 Unconfined compressive strength (UCS), 164,
Time of set, 689–​691 648
T-​joint test devices Unconfined compressive strength testing
authorities, 418 apparatus, 693
flat corners, 418 Unconsolidated undrained (UU) triaxial
German STUVA recommendation, 419 compression test, 83
waterproofing test, 420 The Unified Soil Classification System (USCS),
“T-​lock” (Ameron) system, 383 82
TMCLK project, 819, 821 United States Environmental Protection Agency
Traditional anchor systems with drilling (US EPA), 254
anchor placement alignment quality, 387 Universal rhomboidal segmental ring system,
disadvantages, 386 428, 429
embedded fastening systems, 386, 389 Universal rings, 105, 115, 108–​109, 458
fixing anchor failure in tunnel crown, 388 US Environmental Protection Agency (EPA), 324
post-​installed anchors, 388 Utility corridor tunnel construction, 800–​801
trial locations, 386, 387 Utility systems, 475
tunnel workers, 386 Utility tunnels, 99
Traditional bar reinforcing cage, precast
segment, 39 Vacuum lift guide pins, 369–​370
Trailing gear equipment, 666, 675 Vacuum lifting, 603, 605, 606, 768–​769
Transporting segments Vane shear test (VST), 76, 79
load preparation, 597, 599 Vertical boreholes, 71–​73
mobile equipment, 598 Vertical shaft sinking machine systems,
MSVs, 600 798–​799
rail borne set-​ups, 599, 600 Volume measurements, 681
Transverse bars, 340
Trapezoidal segment system, 113 Waffle segment bolts, 771
Tunnel boring machine (TBM), 1, see also Washington Metropolitan Area Transit
Production and transient loadings Authority (WMATA) system, 19–​20
axial force-​bending moment interaction Washington State Department of Transportation
diagrams, 204 (WASDOT), 453
circular rings, see Circular rings Water, 267–​268
concrete cover, 200 and wastewater tunnels, 99
concrete strength and reinforcement, 197–​199 Waterproofing gasket seals, 554
cost-​effective approach, 254 Water-​reducing admixtures (WRA), 270
defined, 97 Watertightness testing, see also T-​joint test
early age strength for stripping, 199–​200 devices
fiber reinforcement, 202–​203 project-​specific approval test, 420
fibers and reinforcing bars, 204–​211 tunnel segment gasket, 420, 421
FRC post-​cracking nominal residual strengths, Water wells, 70
203 Weather protection, 595
manufacturing company, 26 Welding station, watertight attachments, 385
post-​crack flexural strength parameters, 203 Well built rings, 614
reinforcement spacing, 201–​202 Wind-​blown soils, 56
soft ground, 255 Wireless ring erector controls, 620
 Index 839

WMATA Metro station, 813 WSDOT SR 99 Alaska Way Tunnel in Seattle,

Woodsmith Project Layout, 797 Washington, 451–​454
World Commission on Environment and
Development (WCED), 324 York Potash Woodsmith Mine Conveyor
Worst-​case-​scenario-​tests, 439 Tunnel, Whitby, England, 796–​797