0% found this document useful (0 votes)
2K views429 pages

Historic Concrete - Background To Appraisal

Uploaded by

sonychu999
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
2K views429 pages

Historic Concrete - Background To Appraisal

Uploaded by

sonychu999
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 429

00.

qxd 16/11/2001 16:16 Page 1

Historic Concrete
backround to appraisal

IP: 130.194.20.173
On: Tue, 17 May 2011 06:10:52
00.qxd 16/11/2001 16:16 Page 3

Historic Concrete
backround to appraisal

Edited by
James Sutherland, Dawn Humm and Mike Chrimes

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:10:52
00.qxd 16/11/2001 16:16 Page 4

Published by Thomas Telford Publishing, Thomas Telford Ltd, 1 Heron Quay,


London E14 4JD. URL: http://www.thomastelford.com

Distributors for Thomas Telford books are


USA: ASCE Press, 1801 Alexander Bell Drive, Reston, VA 20191-4400, USA
Japan: Maruzen Co. Ltd, Book Department, 3–10 Nihonbashi 2-chome, Chuo-ku, Tokyo
103
Australia: DA Books and Journals, 648 Whitehorse Road, Mitcham 3132, Victoria

First published 2001

Also available from Thomas Telford Books


Titles to be inserted under the heading Also Available from Thomas Telford Books

Manual of numerical methods in concrete. MYH Bangash. ISBN 07277 2942 6


Owen Williams (The Engineer’s Contribution to Contemporary Architecture series).
D Yeomans and D Cottam. ISBN 07277 3018 5
Innovations in concrete. David Bennett. ISBN 07277 2005 8
Structural detailing in steel. MYH Bangash. ISBN 07277 2850 4

A catalogue record for this book is available from the British Library

ISBN: 0 7277 2875 X

© Authors and Thomas Telford Limited 2001

All rights, including translation, reserved. Except as permitted by the Copyright,


Designs and Patents Act 1988, no part of this publication may be reproduced, stored in
a retrieval system or transmitted in any form or by any means, electronic, mechanical,
photocopying or otherwise, without the prior written permission of the Publishing
Director, Thomas Telford Publishing, Thomas Telford Ltd, 1 Heron Quay, London E14
4JD.

This book is published on the understanding that the authors are solely responsible for
the statements made and opinions expressed in it and that its publication does not nec-
essarily imply that such statements and/or opinions are or reflect the views or opinions
of the publishers. While every effort has been made to ensure that the statements made
and the opinions expressed in this publication provide a safe and accurate guide, no lia-
bility or responsibility can be accepted in this respect by the authors or publishers.

Typeset by Gray Publishing, Tunbridge Wells


Printed and bound in Great Britain by MPG Books, Bodmin

IP: 130.194.20.173
On: Tue, 17 May 2011 06:10:52
00.qxd 16/11/2001 16:16 Page 5

Contents
1 Introduction 1
James Sutherland
2 The innovative uses of concrete by engineers and architects 11
Frank Newby
3 Concrete and the structural use of cements in England before 1890 45
Lawrance Hurst
4 The era of proprietary systems 68
Michael Bussell
5 The development of reinforced concrete design and practice 83
Michael Bussell
6 Cement and concrete as materials: changes in properties,
production and performance 105
George Somerville
7 Concrete foundations and substructures: a historical review 117
Mike Chrimes
8 The early development of reinforced concrete shells 165
Peter Morice and Hugh Tottenham
9 Concrete shell roofs, 1945–65 177
Robert Anchor
10 Prestressing 191
Francis Walley
11 The development of concrete bridges in the British Isles
prior to 1940 211
Mike Chrimes
12 UK concrete bridges since 1940 251
William Smyth
13 Reinforced and prestressed concrete in maritime structures 275
Brian Sharp
14 The Concrete Institute 1908–23, precursor of the Institution
of Structural Engineers 303
Anita Witten
15 Concrete in tunnels 315
Alan Muir Wood
16 Water-retaining structures in Britain before 1920 323
Michael Gould
17 Historic concrete in dams 343
D.A. Bruggemann, K.J. Hollock, G.P. Sims
18 Concrete roads 359
Graham West
19 Military 371
John Weiler
Appendix I: sources of furtherinformation on historic concrete 383
Appendix II: 19th century proprietary floor systems 401
Appendix III: proprietary reinforcing and flooring systems 407
Appendix IV: assessing ancient mortars by John Pickering 425
Index 429

IP: 130.194.20.173
On: Tue, 17 May 2011 06:10:52
00.qxd 16/11/2001 16:16 Page 6

Frank Newby
It was with great sadness that the Editors learnt of the death of one of the con-
tributors, Frank Newby, on 10 May 2001, before this book to which he con-
tributed so much was published.
Although Frank wrote only one chapter, he contributed far more than that,
both through sharing his knowledge of concrete design, and through the time
he dedicated to making the 1996 exhibition ‘Revolution or Evolution’, which
partly inspired this book, a reality.
Frank was one of the leading structural engineers of his generation, a Fel-
low of the Royal Academy of Engineering, Fellow and Gold Medallist of the
Institution of Structural Engineers, Hon FRIBA and for 32 years senior part-
ner in F.J. Samuely and Partners.
These bald statistics give no idea of the character of this friendly, creative,
very generous but in many ways self-effacing man. Of his many interests offi-
cial status certainly was not one. He did not wish to be a president or chair-
man of anything but give him something to design or an intricate problem to
solve and he would blossom.

IP: 130.194.20.173
On: Tue, 17 May 2011 06:10:52
01.qxd 15/11/2001 15:42 Page 1

1 Introduction

James Sutherland
It is common knowledge that the dome of the Pantheon in Rome is made of con-
crete but few realize that the Liver Building (Figure 1.1), which has dominated
the Liverpool waterfront since about 1909 is a pioneer example of reinforced con-
crete framing. Even fewer people know of the Eldon Street flats, also in Liverpool
and even more notable as pioneers, being built in large panel precast concrete in
1905.1 Here the structure (Figure 1.2) is remarkably similar to that ‘invented’ as
system building in the 1950s although perhaps more robust. The first flat slab
concrete floors in Britain are normally attributed to Owen Williams in the 1930s
more than 20 years after the independent invention of flat slab construction in
Switzerland and America. Recently an example (Figure 1.3) has been found in
Britain dated 1919.2 Could there be an even earlier one? These are just cases of

Figure 1.1 The Royal Liver


Building, Liverpool completed
in 1909. Early high-rise
concrete framing.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:11:58
01.qxd 15/11/2001 15:42 Page 2

2 Historic concrete

information lost or ignored. Concrete, it seems, is not seen as a suitable subject


for historical study, yet in the last 100 years it has probably had a greater impact
on our surroundings, and indirectly on our way of life, than any other material.
One wonders why this influence is not more widely recognized. Apart from any cul-
tural interest, there are very practical reasons for studying the history of concrete.
In Britain we have a vast stock of concrete structures of varying ages. Frequently
these need to be appraised, repaired or altered, but relevant drawings or other
records seldom exist and concrete is probably the most totally opaque of all struc-
tural materials. With steel, timber and masonry one can generally see, and even
measure, what is there but with concrete it is often hard to judge from outside
even whether it is mass concrete or reinforced or prestressed. It might not be struc-
tural at all but just protecting a steel frame. To find out more one can use ultra-
sonic testers and covermeters, drill holes or chip away surfaces but, even so, the
results may be puzzling. The best starting point for appraisal is some knowledge

Figure 1.2 Concrete flats


in Eldon Street, Liverpool.
Pioneer large panel
precast concrete
construction of 1905.
(a) Transporting units
from the precasting yard.
(b) Erecting wall units,
with floor units to follow.
(c) The completed
building. Reproduced
with the permission of
the Liverpool Record
Office, Liverpool Libraries
and Information Services. b

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:11:58
01.qxd 15/11/2001 15:42 Page 3

Introduction 3

Figure 1.2c

Figure 1.3 Bryant &


May’s factory of 1919
with possibly the earliest
flat-slab concrete
construction in Britain.

of what is likely to be found at different dates and with different types of struc-
ture. This is where history — the theme of this book — comes in.
The book arose out of a set of papers on different aspects of the history of
reinforced and prestressed concrete published by the Institution of Civil Engineers
in 1996.3 Concurrently there was also an exhibition called ‘Revolution or Evolution’
and a half-day meeting. All feedback pointed strongly to the desirability of pub-
lishing an edited and expanded version of the papers in book form, the thinking
being that engineers who may be asked to make appraisals are more likely to remem-
ber, find and refer to a book than to look out back numbers of a journal. Likewise
architects, surveyors, historians of construction and general readers could well find

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:11:58
01.qxd 15/11/2001 15:42 Page 4

4 Historic concrete

a book but be less likely to search for papers in a journal which, at first, might not
seem relevant to their interests. Time will show whether this thinking was correct.

The format of For the most part the book deals with concrete in Britain, but with references to
discoveries and works in other countries where these have influenced practice in
the book Britain. The first 14 chapters are edited versions of the original papers with much
of the statistical and tabulated material transferred to appendices for easy refer-
ence. A further five chapters have been added covering military applications of
concrete and its use in tunnels, in roads and pavements, in water-retaining struc-
tures and in dams, thus making good some major omissions on civil engineering.
No rigid starting date has been taken in the different chapters, the authors decid-
ing on significant periods in each case. For foundations, military works and water-
retaining structures the treatment goes back to the middle of the 19th century, or
marginally before, to include mass concrete in fortifications, reservoirs and the bases
of gas holders. Also for the development of cements and for patent flooring systems
the effective starting time is around 1850 while for reinforced concrete in Britain
it is seen as about 1900 and for prestressing the middle of the Second World War.
Effectively there are no finishing dates and generally no attempt has been made
to cover present practice. Reinforced concrete continues to find favour for most
types of structure but here the concentration is on the ‘working stress’ era up to
around 1970, with a strong emphasis on out-dated reinforcing details and only a
glancing reference to limit state thinking and present codes. There is little on water-
retaining structures like water towers and swimming pools after 1920 because by
this time reinforced concrete was well established and techniques were similar to
those for other concrete structures. With prestressed concrete the principles have
remained constant since its introduction in the 1940s but the size of tendons and
anchorages has increased and the detailing has developed. As with reinforcing bars,
patent piling systems and early prestressing anchorages and ducts are recorded
in some detail in the appendices.
It would never be possible in one book to cover every historical aspect of the struc-
tural use of concrete. Even with the number of chapters expanded beyond the range
of the original papers there must still be gaps. For instance, silos, bunkers, pipelines,
sewers, masts and fencing are not specifically referred to. Also the architectural treat-
ment of concrete, the development of precasting (especially in relation to quality
of finish), system building, semi-structural concrete cladding and artificial stone are
all very relevant both to the historian and to engineers carrying out appraisals
but are hardly touched upon in the main chapters. Likewise the problems with
concrete, which have received so much publicity in the last 40 years are only
covered incidentally. Some notes on these ‘neglected’ topics are given below.

Visual aspects of In Britain concrete was seen at first as a purely structural material, perhaps at its
most impressive on a large scale in bridges and also in utilitarian structures where
concrete appearance was not thought to matter, as in the case of most jetties and industrial
plants. Except for artificial or reconstructed stone, which is really an ersatz mater-
ial mimicking natural masonry, the really creative approach to the appearance of
concrete did not come until the 1930s when rough-boarded shuttering and
various forms of exposed aggregate finish were introduced. Maxwell Ayrton’s archi-
tectural detailing to Twickenham Bridge (1928–33) is a good example.4 It included
bush-hammering, brushing off the surface cement when green and, in the case
of the breakwaters, a hammered ribbed finish later dubbed as ‘Elephant House’
after the London Zoo building of 1962–65.
It was after the Second World War that carefully designed exposed concrete sur-
faces became really popular, especially in buildings. The positions of construction
joints and tie bolts were controlled, with the latter often emphasized, and, in the

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:11:58
01.qxd 15/11/2001 15:42 Page 5

Introduction 5

Figure 1.4 Visually dominant


concrete of the 1960s (Essex
University Library). (a) Precast
and in-situ concrete all designed
to be exposed (rough-boarded
and fractured rib finishes).
(b) Detail of ribbed finish to
fascia beams during ‘fracturing’.

case of rough-boarded formwork, the width, surface finish and irregularity of the
boards were all carefully set out. This care may be seen as a softening of the ‘hon-
est’ or ‘brutal’ architectural functionalism of the period. The Hayward Gallery and
the National Theatre on the South Bank of the Thames are well-known examples,
the finishes at the National Theatre being visually particularly successful. With such
care what may have been seen at first as a method of saving money rapidly became
just the reverse, the extra cost often greatly exceeding that of all but the most
expensive applied finishes.
Apart from rough-boarded finishes, exposed aggregate surfaces also became
popular in the 1950s and 1960s and were exploited with varying success, the best
again being planned with great care and with joints emphasized.
Also after 1950 precasting of concrete was recognized not just as a means of
saving time but of improving the appearance of structural framing and of concrete
cladding. Quality is easier to control with precast than with in-situ concrete and, at
worst, individual units can be rejected without too much disruption. Here the visual
problem lies largely in controlling the widths of joints and their alignment.
The Library at Essex University (Figure 1.4) is virtually a dictionary of the var-
ieties of the techniques, which were used in the 1960s to achieve fine integral fin-
ishes and the virtual elimination of applied coatings on buildings of fundamentally
dominant form. Structural ‘honesty’ had by then become near to a religion.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:11:58
01.qxd 15/11/2001 15:42 Page 6

6 Historic concrete

This is not the place to argue the aesthetic principles behind exposed concrete
surfaces. However it must be admitted that much of the public dislike of concrete,
which became strong in the 1970s and 1980s, has been due to poor and uneven
weathering with the occasional crack or rusty streak due to exposed reinforcement
or iron in the aggregate. The big misconception has been that there are some types
of concrete finish, which, even if expensive, are permanent. In many cases sim-
ple cleaning on a regular basis may be all that is needed for good maintenance
but, where some actual defects like spalling have occurred, it is very difficult to
make repairs to concrete which are visually acceptable in the long term. Clever
craftsmen can hide patching initially but there is a tendency for evidence of it to
reappear and become more and more obvious year after year. Also after some years
sharp edges may be blurred by acid rain and aggregate exposed irregularly. In
many cases owners of buildings with exposed concrete finishes have opted for
painting — as with stucco — but this does undoubtedly alter the original image
and of course needs to be repeated regularly.
The misconception that there are permanent finishes to concrete which need
no maintenance is only matched by the misunderstanding over the durability of
reinforced concrete as a structural material.

The durability of Writing in a pioneer textbook published in 1904, C.F. Marsh stated that the dur-
ability of reinforced concrete was ‘well established’ and that the cost of maintaining
concrete it was ‘nil’.5 This myth has continued for a surprisingly long time. It may not be quite
dead yet. If applied to mass concrete or to reinforced concrete wholly protected from
damp, Marsh’s statement would be near to the truth but much reinforced concrete
is exposed to the weather or to internal leaks in buildings or bridge decks.
Engineers always seem to have felt a strong urge to reinforce concrete sections
even where this is not necessary to resist calculated forces and often both design-
ers and those carrying out the construction have taken far too cavalier an attitude
to cover to reinforcement. In Chapter 5, M.N. Bussell enlarges on the lack of
adequate guidance on cover and on durability (at least in the period up to the
1930s). With precast units such as columns and mullions much of this reinforce-
ment was doubtless put in to prevent damage during handling and in the case of
in-situ concrete to control shrinkage. Figure 1.5 shows two sections through
precast reinforced mullions and sills as illustrated in a publication first issued in
1918.6 Here the cover seems impossibly small by any standards and it is doubtful
whether the reinforcement was needed at all. If actually used, one must wonder

Figure 1.5 Example of very


small concrete cover to
reinforcement as advocated
in a book first published in
1918.6

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:11:58
01.qxd 15/11/2001 15:42 Page 7

Introduction 7

if this detail led to rust and splitting? It is hard to be sure. Fence posts are a good
example of the variable performance of reinforced concrete with low covers.
Lack of cover to reinforcement has not been the only cause of deterioration in
concrete. In Chapter 6, Somerville discusses several aspects of durability with use-
ful references, emphasizing in particular our blindness to the reduced protection
of steel as carbonation takes place. This was not recognized in practice until the
1970s although scientifically it was known half a century earlier.
In spite of many shortcomings with past practice much excellent reinforced and
prestressed concrete exists today but we should not assume that it will not need
maintenance in the future. Appraising engineers would do well to emphasize this
continuing need when reporting to clients, even where visually there appear to
be no defects at present. Everyone expects timber, iron and steel to need main-
tenance. Concrete cannot be considered immune.

System building in There was a strong movement towards concrete building systems in the 1950s
and early 1960s. This was partly ideological and partly because of a real need to
concrete build quickly after the War using the minimum amount of steel. The systems all
depended on prefabrication. Intergrid and Laingspan, initially intended for schools
and both based on prestressing, were used quite extensively for office blocks as
well. Individual system-built houses in precast concrete such as the Airey house
date mainly from 1945 but some were built in the 1930s.7 In the 1950s and 1960s
several large-panel precast systems were developed in Britain, or imported from
the Continent, for use in housing blocks up to 25 storeys high.8 The fashion for
high-rise housing was short lived on social grounds and, structurally, large-panel
precast systems went out of favour following the partial collapse of the Ronan Point
tower in 1968.9 Nevertheless many such buildings remain, all the taller ones
having been strengthened. Apart from full system building, precast wall panels
with insulation sandwiched between concrete skins have been commonly used on
all types of building from the 1950s onwards.
It is not always easy to understand from visual inspection how these system build-
ings were designed and especially how the concrete units were joined. Here
engineers may find it useful to consult the archive of trade literature referred to
at the end of this introduction.

Postscript on It is interesting to speculate on the impact of reinforced concrete on those work-


ing on site around 1900 and soon after. The most uninitiated site worker would
detailing have had a feeling for how a beam or column of timber or steel should behave,
reinforcement even if without understanding the distribution of the stresses. However with rein-
forced concrete it must at first have been far from obvious what the bars were doing
and what was important about the way in which they were placed. New conven-
tions for detailing had to be developed. Figure 1.6 shows the evolution in the rep-
resentation of reinforcement from a highly pictorial form for a complex detail early
in the century to a typical stylized CAD output, which must be incomprehensible
to anyone outside the industry, but presents no problem to the steelfixer today.

Further detailed Some useful references to many aspects of structural concrete, especially those not
covered in individual chapters of this book, can be found in the Appendix.
information on Journals such as Concrete and Constructional Engineering and Concrete are a mine of
historic concrete technical information as are the books in the Concrete Series by Concrete Publi-
cations Ltd, which are listed in this Appendix. These are just examples. There
are also the journals or proceedings of the engineering institutions and general

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:11:58
01.qxd 15/11/2001 15:42 Page 8

8 Historic concrete

Figure 1.6 The development


of methods of defining
reinforcement on drawings.
(a) Pictorial detailing of c.
1900 (Hennebique).
(b) Formalized pictorial
detailing of c. 1910.
(c) Stylized CAD detailing of
the 1990s.

engineering and architectural journals which are listed with their periods of
publication.
As well as published accounts of structures and structural forms, there is
a useful archive of ‘real’ reinforced concrete drawings dating back to 1903.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:11:58
01.qxd 15/11/2001 15:42 Page 9

Introduction 9

Figure 1.6c

This was built up by the Institution of Civil Engineers together with the Concrete
Society and is held at the Institution, partly as a historical record but also to help
appraising engineers to know what they might expect or to understand better what
they have found.
Thanks are due to all those who have already provided this historical material
and, in particular, to L.G. Mouchel & Partners for the gift of many drawings
and especially for a number of complete dossiers on their early projects with
calculations and other information as well as working drawings. Further additions
to this archive would be welcome. Finally trade brochures are being gathered
on early reinforcing methods, on prestressing anchorages, on precast concrete
building systems and on patent piling systems and held by the Institution of
Civil Engineers while the Science Museum has a growing collection of related
hardware.

References 1. Report of City Engineer, City of Liverpool. Concrete Dwellings, Eldon Street: Liverpool
28 April 1905; Patent Specification No. 6115 A.D.1901 (John Alexander Brodie); Moore,
R., An early system of large-panel building. RIBA J., 1969, 383–86.
2. Yeomans, D., Construction since 1900: Materials. Batsford, 1997: 121–23.
3. Historic Concrete. Proc. Instn Civ. Engrs Structs & Bldgs, 1996, 116, 255–480.
4. Gray, W.S., Childe, H.L., Concrete Surface Finishes, Renderings and Terrazzo. London,
Concrete Publications Ltd., 1935.
5. Marsh, C.F., Reinforced Concrete, Archibald Constable, 1904.
6. Lakeman, A., Concrete Houses and Small Garages, 4th edn. London, Concrete Publi-
cations Ltd., 1918, Rewritten 1949.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:11:58
01.qxd 15/11/2001 15:42 Page 10

10 Historic concrete

7. Post-War Building Studies No. 1. House Construction. HMSO, 1944; Post-War Build-
ing Studies No. 23. House Construction Second Report. HMSO, 1946.
8. Diamant, R.M.E. (in collaboration with the Architect and Building News). Industrialised
Building 50 International Methods. London, Illiffe Books Ltd., 1964 & 1965.
9. Ministry of Housing and Local Government. Report on the Inquiry into the
Collapse of Flats at Ronan Point, Canning Town. HMSO, 1968.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:11:58
02.qxd 15/11/2001 15:43 Page 11

2 The innovative uses of concrete by


engineers and architects

Frank Newby
Synopsis Concrete as originally exploited has little tensile strength, but with the develop-
ment of reinforced concrete at the end of the 19th century, a composite material
capable of taking tensile stresses was available to engineers. Taking an international
view, this chapter looks at the way in which engineers and architects chose to exploit
this and the later developments such as prestressed concrete, illustrating innova-
tive structural and architectural uses in a historical context.

Introduction In general terms concrete is an agglomerate of coarse material such as crushed


stone, gravel, or broken bricks or tiles, and sand mixed with a binder and water
which in time forms a hard artificial stone. Mixed with sand, binders, such as
lime, have been used from the earliest times to form mortars for brickwork.
Similarly the first concrete was lime concrete, and as improved binders were
developed concretes were classified by the type of binder (such as, pozzolana,
natural cement, artificial or Portland cement and epoxy). Each type of concrete
has its own properties and has been exploited in many different ways by archi-
tects and engineers.
Concrete as defined above has little tensile strength compared with its
compressive strength. Towards the end of the 19th century, reinforcement of
concrete to form a composite material capable of taking tensile stresses became
feasible, and the age of reinforced concrete began. The original wrought iron
reinforcement was replaced by steel in its different forms, not to mention copper
and bamboo, and today the use of glass and carbon fibre reinforcement is in its
infancy.
An alternative method of making concrete capable of taking tensile loads came
in the 1930s with the advent of prestressing, where the concrete is precompressed
so that tensile stresses do not occur under working conditions. This represented
a fundamental change in structural thinking, and prestressed concrete can be
considered as a new material with its own technology and uses.
The chief characteristic of concrete is that when freshly made it forms a plastic
mass which requires containment and support until it has sufficiently hardened.
Its form and appearance are therefore defined by the shape and surface texture
of the mould, which are specified by the engineer or architect, and by the sequence
of building, as determined by the contractor.
The three main structural requirements are first that the concrete construc-
tion will satisfactorily carry all loads into the ground or onto other supports,
secondly that in underwater construction it will harden, and thirdly that it
will retain its stability in a fire. Today concrete is also used as external weather-
proof cladding, for limiting acoustic transmission and for providing thermal
mass.
In order to illustrate the general development of concrete construction against
which works in the UK can be assessed, examples showing innovative structural
and architectural use are described in mainly chronological order.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:12:21
02.qxd 15/11/2001 15:43 Page 12

12 Historic concrete

Plain concrete Although mortars and even weak concretes were developed in Egypt and earlier
civilizations, it was in the Roman period that the first major use of concrete
was made. Vitruvius in his Ten books on architecture describes ‘a kind of powder
from natural causes produces astonishing results from the neighbourhood of
Vesuvius. This substance mixed with lime and small stones not only lends strength
to buildings but when piers of it are constructed in the sea they set hard under
water’. He was referring to pozzolana, as we know it today, a natural cement of
volcanic origin. Whereas lime obtains its hardness from atmospheric exposure,
pozzolana has silica and alumina inclusions which react with the lime even under
water.
The earliest extant domes which used pozzolana concrete appeared in the roofs
to baths in Pompeii in the second to first century BC. Being plastic, concrete can
be poured into three-dimensional shapes and is much cheaper than cutting
stonework. The surface of the domes reflects the rough timber shuttering.
In the reconstruction of Rome in the first century AD, much use was made of
pozzolana concrete. When Hadrian took over construction, one of his first pro-
jects was that of constructing a civic temple, the Pantheon (Figure 2.1). This had
a dome 43.3 m in diameter, some three times larger than any other built, and was
completed in 128 AD. He appreciated the potential of pozzolana concrete and had

Figure 2.1 The Pantheon in


Rome, 128 AD.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:12:21
02.qxd 15/11/2001 15:43 Page 13

The innovative uses of concrete by engineers and architects 13

the courage and financial power to proceed with his revolutionary design. To min-
imize the weight of the dome he introduced five rows of coffers shaped to be seen
from the ground. He also reduced the density of the pozzolana concrete as he built
up to the eye of the dome. The external surface of the dome was stepped to
facilitate casting against vertical shuttering. In describing the Pantheon, Rowland
Mainstone1 states, ‘The Roman concrete dome … permitted, for the first time, an
architecture of large unencumbered interior spaces that could be experienced only
from within’.
Although the properties of pozzolana had been known and exploited for cen-
turies, it was not until John Smeaton’s experiments in the mid-1750s that the rea-
sons for these properties were understood. Smeaton’s first civil engineering project,
undertaken at the age of 32, was to rebuild the Eddystone Lighthouse, for which
he needed hydraulic mortar which would set under water. He solved his imme-
diate problem by using pozzolana, which he knew about from Vitruvius and from
Bélidor’s Architecture hydraulique, but his interest was roused. He therefore carried
out a series of tests to discover what it was that made some limes, tarras and poz-
zolana hydraulic and established that it was the clay content alone that imparted
this property. Details of these tests, which eventually led to the further widespread
research into the production of hydraulic limes and cements, are given by Smeaton
in his 1791 book on the lighthouse, and the history of these developments are
detailed by Pasley2 and by Vicat.3
‘Roman cement’, patented by Parker in 1796, which was a natural hydraulic
cement made from calcining cement stones found in the chalk, set too quickly
for use in foundations but was used extensively for watertight mortar. The devel-
opment of ‘Portland cement’, the artificial cement patented by Aspdin in 1824,
did not seriously advance until about 1859 when John Grant, engineer to the
Metropolitan Board of Works, carried out a long series of tests before using it on
a major civil engineering project, the London main drainage works.4
Lime concrete was specified in 1817 by the architect, Robert Smirke, for the foun-
dations of the Millbank Penitentiary, London. Quicklime and gravel were mixed
and, for no logical reason, dropped into a trench from a height of 6 ft and water
added. He later used the same technique when underpinning and rebuilding the
Customs House in 1827. In 1840, Charles Barry also specified a lime concrete for
the foundations of his new Palace of Westminster, which was also dropped into
place, this time from a height of 10 ft.
In France, François Coignet, who was active in the chemical industry, attempted
in 1852 to build an exposed lime concrete walled factory following the ancient pisé
system of construction. After overcoming difficulties with mixes and water con-
tent on this, his first encounter with concrete, he commissioned an architect to
design a sophisticated four-storey house in St Denis, Paris, which had exposed con-
crete replicating a typical stone building. The intermediate floors had fireproof
floors of timber beams encased in lime concrete, while the roof slab had iron joists
encased in concrete.5
Because of its success he took out a British patent in 1855 (No. 2659) entitled
‘Emploi de Béton’ which gave details of his method of construction and contained
an addendum emphasizing the complete sufficiency of concrete.
Concrete walls need no facing materials such as stone etc. The hollow part
of the mould in which the concrete is poured should have the form to be
given to the mass whether the walls be plain or with projections such as
cornices, string-courses … or any kind of ornament.
In a submission to the organizing committee of the Universal Exhibition of 1855,
where he wished to build a concrete house, he makes a telling remark. ‘The reign
of stone in building construction seems to have come to an end. Cement, concrete
and iron are destined to replace it.’

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:12:21
02.qxd 15/11/2001 15:43 Page 14

14 Historic concrete

Figure 2.2 Church at


Le Vesinet, 1862.

Later he also wrote,6


‘building cheap vaults of unusual spans for great public spaces … will
not be a servile imitation of Roman works but will far surpass them in
daring, elegance and economy’.
Following the success of his early works, Coignet became a building contractor,
showing to the world by his many designs, particularly the church at Le Vesinet
(Figure 2.2) in 1862, that mass concrete was an acceptable material for construc-
tion. His work was illustrated in the UK where Joseph Tall patented a shuttering
system in 1865 (No. 822) which came to be widely used in the building of houses.
In the second half of the 19th century concrete was the new material that excited
architects, particularly for fireproof construction, and there are a surprising
number of buildings still extant with exposed concrete facades.
The concrete finish frequently left a great deal to be desired, sometimes to the
point where the concrete had to be covered with a concrete render or designed
using coloured precast concrete tiles as permanent shuttering. Despite these prob-
lems, its fireproof property meant that many theatres were built in mass concrete
with fireproof floors, notably the Royal English Opera House, London (now
the Palace Theatre), which was begun in 1888.7 The embedding of iron joists in
concrete to provide fireproof flooring systems had flourished in the UK from 1844

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:12:21
02.qxd 15/11/2001 15:43 Page 15

The innovative uses of concrete by engineers and architects 15

Figure 2.3 Glenfinnan


Viaduct, Scotland, 1898.

and these are described by Lawrance Hurst.34 In the field of civil engineering there
are no better examples than the Borrodale Bridge, with a single span of 127 ft
6 in and the 21-arch Glenfinnan Viaduct, which is 416 yd in length and up to 100 ft
in height (Figure 2.3). Both these structures were designed in mass concrete by
engineers Simpson and Wilson of Glasgow and built by ‘Concrete Bob’ McAlpine
in 1898 for the West Highland Railway.

Development Although Wilkinson8 took out a patent for reinforced concrete in 1854 in the UK,
the first practical appearance of the material was in France in 1855 with Lambot’s
of reinforced ‘ferciment’ concrete boat. However, a far more important development took place
concrete in Europe in 1867 when Monier produced his reinforced concrete flower pots which led to
his patent for structural reinforced concrete in 1877. At the same time Hyatt,9 the
inventor of glass prisms for pavement lights, had organized a number of tests on
concrete beams reinforced with iron ties which were carried out by David Kirkaldy
and found to be satisfactory. He also carried out fire tests to show that the steel
and concrete acted in unison under heat.
In 1879 the astute contractor, Wayss, bought the German rights to Monier’s
patent, and in 1887 he published Das System Monier (Eisengerippe met Cementumhülling)
in seiner Anwendung aug das gesammte Bauwesen. As often happened, the new
system was first used in civil engineering, in this case for arched bridges, some of
which have the elegance of Maillart. One of these, built in 1890, spanned 37.2 m
(Figure 2.4).
François Hennebique,10 a provincial building contractor working in Belgium,
experimented with concrete reinforced with iron from about 1879, and in 1892
patented his system, giving his method of calculation and typical reinforcement
details for the bending of beams and slabs. He accepted many commissions before
moving to Paris in 1894, when he gave up contracting and supervised the
training of concrete contractors instead. At first his engineers carried out the
calculations in Paris, but later this was done by his agents in their own offices and
sent to Paris for checking.
The building system was essentially a frame with infill floor slabs and external
cladding in brick, stone, concrete or glass. Continuity of reinforcement through

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:12:21
02.qxd 15/11/2001 15:43 Page 16

16 Historic concrete

Figure 2.4 Monier Bridge,


1890.

Figure 2.5 Hennebique’s


Tourcoing factory, 1895.

the haunched column and beam joints provided rigidity naturally and made can-
tilevers easy to construct. Architects accepted these characteristics and developed
their designs accordingly. As early as 1895 the spinning mill at Tourcoing, one of
Hennebique’s earliest projects, already clearly illustrates the new concrete archi-
tecture (Figure 2.5). By the time he had produced his first building in the UK
(the 1897 Weaver flour mill in Swansea with agent, L.G. Mouchel), several 100
projects had already been constructed in Europe.
The Paris Exhibition of 1900 revealed for the first time to a wide international
audience the great potential of reinforced concrete both in architecture and in
civil engineering, since it was used for a number of pavilions and other installa-
tions such as bridges and sewage works. For the Petit Palais, Hennebique together
with the architect Charles Girault, built a free-standing spiral slab staircase,

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:12:21
02.qxd 15/11/2001 15:43 Page 17

The innovative uses of concrete by engineers and architects 17

a three-dimensional form which would be repeated many times. In 1904


Hennebique built a house for himself in Paris, the flamboyant design of which
again demonstrated the role of concrete in domestic architecture. For its
construction thin precast concrete units 14 in high were used as permanent shut-
tering. Hennebique published a house magazine, Le béton armé, which was circu-
lated to all his regional offices giving details of the latest projects and information
on new developments. In 1899 one of his engineers, Paul Christophe published
the first textbook on reinforced concrete, Le béton armé et ses applications.
Many other systems were patented, mainly on the geometry and nature of the
reinforcement, but only a few survived and were developed into the 20th century.
These included those of Coignet and Considère, both of whom followed Hennebique
in opening offices in the UK. These are described by Michael Bussell in Chapter 4.

Development Natural cements had been produced in the USA since the early 1820s and
Portland cement since 1871. As in France and the UK, unreinforced concrete walls
of reinforced and fireproof filler-joist floors were used, but in the USA concrete blocks with
concrete in the different finishes and colours became popular. Mass concrete received a boost
USA when it was used for the foundations of the Statue of Liberty.
Ernest Ransome,11 who first used reinforced concrete in America, began by
running his father’s San Francisco factory which was producing artificial stone.
In 1844 he patented the use of cold twisted iron bars of square section as
reinforcement, carrying out extensive tests to confirm his patent. In 1888 he used
reinforced concrete for the main floor structure, which was supported on internal
cast iron columns and external stone walls, but the following year the whole struc-
ture was in concrete. On another contract in 1889 he introduced and patented the
ribbed floor construction still used today (Figure 2.6). He precast the beams and
together with in-situ concrete slabs formed a composite construction. He went on
to build many notable multi-storey factories and warehouses. The 16-storey Ingalls
building in Cincinnati of 1902–1903 (Figure 2.7), then the world’s tallest building,
made use of Ransome’s system. His system and that of the Trussed Concrete
(Truscon) Co., which used the Kahn bar, were the most widely used in this early
period, although the Monier and Hennebique systems had also been introduced.

Figure 2.6 Ransome rib


floor, 1889.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:12:21
02.qxd 15/11/2001 15:43 Page 18

18 Historic concrete

Figure 2.7 Ingalls building,


1903.

Earlier, in 1902, Orland W. Norcross, an engineer from Boston, patented a flat


slab system of radial reinforcement from columns so as to eliminate connecting
beams between columns for economy, having successfully built a structure on this
principle. The first mature method of flat slab construction was the inclusion of
an enlarged mushroom head first used by CAP. Turner in 1905–1906 and patented
in 1908. He used four layers of reinforcement orthogonally and diagonally between
columns. The omission of beams gave extra headroom or allowed the floor-to-
floor height to be reduced, and it allowed services or partitions at a constant height
as well as providing more light internally. Turner became as busy an engineer as
Ransome and Truscon.

The work of Robert Maillart12 was trained at ETH in Zurich under Professor Ritter. He gradu-
ated in 1894 and started work with a contractor who was building Hennebique
Robert Maillart, structures. He then moved to the Department of Works of Zurich, where he
1872–1940 designed and built a mass concrete arched bridge which the city architect clad in
stone. In 1901, with another contractor, he designed his first reinforced concrete
bridge, a three-hinged arched structure spanning 38 m at Zuoz, in which he showed
his appreciation of the character of reinforced concrete by joining together the

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:12:21
02.qxd 15/11/2001 15:43 Page 19

The innovative uses of concrete by engineers and architects 19

Figure 2.8 Tavanasa Bridge,


1905.

slabs and walls to form a monolithic box construction. The road slab cantilevered
beyond the spandrel to form a marked shadow line, a feature which was to become
typical of Maillart’s own designs and of many bridges today. His professor, Ritter,
checked the design and supervised a load test.
In 1902, at the age of 30, he set up his own design and contracting firm and
successfully tendered for the bridge at Bilwil. Like Zuoz, it had two spans and was
of similar construction. Maillart’s moulding of the concrete at joints, a feature which
also became a hallmark of his work, is of particular interest. As with Zuoz, small
cracks occurred in the solid spandrels, so in 1905 for the 51 m span Tavanasa
Bridge (Figure 2.8) he reduced the spandrel concrete to a minimum, expressing
the three-hinged structure with elegance and minimum cost.
Maillart also studied the nature and economics of floor construction. In 1909
he patented his ‘beamless deck’ system after testing prototypes, and for a few
years built many buildings, mainly warehouses, using the system in Switzerland
(Figure 2.9), Russia and other parts of Europe. In contrast to Turner’s system of
1908 in the USA which used four layers of mesh, Maillart only used two layers in
each face of the slab with widely splayed column heads and octagonal or square
columns, a more rational engineering solution and quite different from the frames
of Hennebique.
During the 1914–18 war Maillart was marooned in Russia, but on his return he
set up office again but as a consultant. His next development in bridge design was
in 1923 when he built a deck-stiffened arch at Flienglibach. Here a thin arched
slab was loaded through walls from the road deck, which had deep parapet beams
of sufficient stiffness to minimize the deformation of the arch under unequal load-
ing. Two years later he built the Valtschielbach Bridge (Figure 2.10) to a similar
design.
His greatest bridge, also a deck-stiffened arch, was at Schwandbach in 1933,
where the elliptical ground plan of the road deck is supported on a varying width
thin arch by vertical walls to give a truly three-dimensional form for the whole
(Figure 2.11). His structural virtuosity and architectural flair are also revealed in
his unusual but logical roof to the train shed at Chiasso13 with its organic shape
and in his spectacular Cement Hall of 1939 built for the National Fair in Zurich.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:12:21
02.qxd 15/11/2001 15:43 Page 20

20 Historic concrete

Figure 2.9 Maillart’s flat


slab, 1909.

Figure 2.10 Valtschielbach


Bridge, 1923.

The work of Born 7 years later than Maillart, Freyssinet14 studied reinforced concrete under
Rabat in Paris. In 1905, at the age of 26, he took up his first professional job (with
Eugène Freyssinet, a provincial government department) at the period when Maillart was building
1879–1962 the Tavanasa Bridge. In 1910 Freyssinet designed the Verdre Bridge with three
spans: 225, 240 and 225 ft. Each span was a three-pinned arch with decentring
jacks positioned horizontally at the crown. Due to creep of the concrete the crowns
deflected 5 in, so he reinstated the jacks to relevel. For later bridges he turned to
two-pin arches to reduce deflections and carried out further studies into creep
and into concrete mix design to produce higher strengths.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:12:21
02.qxd 15/11/2001 15:43 Page 21

The innovative uses of concrete by engineers and architects 21

Figure 2.11 Schwandbach


Bridge, 1933.

Figure 2.12 Montluçon,


1915.

In 1914 Freyssinet left government service and became technical manager of a


construction firm, Mercier Limousin et Cie, but had to spend four years in the army
constructing many industrial projects in reinforced concrete which used a minimum
weight of steel which was in short supply. Innovation and invention were readily
accepted and Freyssinet introduced some of the first concrete shells ever built.
In 1915 he covered the glass works at Montluçon with a series of arches with
concrete vaults in between (Figure 2.12). The following year, for the Schneider

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:12:21
02.qxd 15/11/2001 15:43 Page 22

22 Historic concrete

Figure 2.13 Orly hangars,


1921.

Figure 2.14 Conoid shells,


1928.

factory at Le Creusot, he produced a continuous roof vault with external stiffen-


ers and internal ties with the weight distributed onto external arches. For aircraft
hangars he turned to concrete vaults, and for the two airship sheds at Orly in 1921
he produced thin corrugated vaults spanning 70 m, rising 50 m, and 200 m in length.
These hangars were later seen by architects as great works of art (Figure 2.13). In
1928 he began to use conoid shells (Figure 2.14) for a number of projects such as
the Austerlitz station in Paris, while his Plougastel Bridge, completed in 1930, broke
all records with its three spans of 180 m (Figure 2.15). He described this project
at the Liège conference of 1930, by which time, aged 50, he had retired from the
firm to devote his time to the development of prestressed concrete, a system he
had patented in 1928. However, it must be said that for the previous 40 years a
number of engineers15 had tried to invent a workable system.
During the Plougastel Bridge project, Freyssinet began to understand the nature
and magnitude of creep, concluding that the losses in prestress would not be exces-
sive if high strength concrete and high tensile steel were used. He succeeded in
making hollow pretensioned, prestressed concrete lighting poles on which he spent
much of his fortune, but to no avail as the market collapsed. In 1934 his luck
changed when, at the last minute, he saved the Marine Terminal at Le Havre from
collapse (it was settling at the rate of 1 in per month) by adding extra concrete

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:12:21
02.qxd 15/11/2001 15:43 Page 23

The innovative uses of concrete by engineers and architects 23

Figure 2.15 Plougastel


Bridge, 1930.

between existing foundations and stressing them together to form a grillage which
he supported on prestressed concrete piles. This was his first attempt to exploit
his ideas on a large scale. Prestressing was accepted, and in 1936 he built his first
bridge, of 19 m span and 4.9 m wide, over the dam at Portes-de-Feu. Very soon
he was building bridges from prefabricated units and stressing them together. His
most telling remark was that ‘prestressing is a state of the mind’ and singly it was
the most far-reaching engineering invention of this century. Francis Walley has
given an extensive review of the subject in Chpater 10.36

The architect and The first architect in France to develop concrete construction was Auguste
Perret,16 who in 1903 chose a concrete frame for a 10-storey block of flats in 25
the new material — Rue Franklin in Paris (Figure 2.16). Externally he emphasized the structure
reinforced by cladding the frame with a plain ceramic tile and the infill with a decorative
concrete tile with a subtle difference in colour. Glass blocks, recently invented, also became
part of the design. Although he worked in his father’s building firm, the flats
were constructed by others, but for a car showroom two years later he also acted
as the contractor and exposed the concrete frame to produce a functional classical
facade (Figure 2.17).
After his father’s death in 1905, he and his brother renamed the firm Perret
Frères and within a short time became specialists in reinforced concrete. They were
appointed to construct the theatre in the Champs Elysées in 1911 to a design by
Henri van de Welde. Perret criticized the proposed structure and ended up
redesigning it with plans signed by the official architect. Because of the confined
site on which to build two auditoria (the larger seating 2000 people), the struc-
ture was brilliantly rationalized. He went on to explore structural expression in
his industrial buildings and the interior of the Esden clothing factory in Paris in
1919 gave a new dimension to the possibilities of reinforced concrete. Three years
later in 1922 he took on the design and building of a minimum cost church at
Le Raincy (Figure 2.18) where his use of pierced precast concrete panels for
external walls together with exposed concrete barrel vault roofing expressed a true
concrete aesthetic.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:12:21
02.qxd 15/11/2001 15:43 Page 24

24 Historic concrete

Figure 2.16 Rue Franklin,


Paris, 1903.

Figure 2.17 Car showroom


Paris, 1905.

In the UK the earliest buildings in which architects exploited the nature of con-
crete were the 1907 Lion Chambers in Hope Street, Glasgow (Figure 2.19) and
the 1909 Liver Building in Liverpool, which was 15 storeys high, both built by the
Hennebique Company. In the USA Frank Lloyd Wright thought that concrete
would bring a new architecture and it is interesting to see how, in 1906, he designed

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:12:21
02.qxd 15/11/2001 15:43 Page 25

The innovative uses of concrete by engineers and architects 25

Figure 2.18 Church at Le


Raincy, 1922.

Figure 2.19 Lion Chambers,


Glasgow, 1907.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:12:21
02.qxd 15/11/2001 15:43 Page 26

26 Historic concrete

his Unity Church in Oak Park, Chicago, where he used monolithic walls with elab-
orate mouldings (Figure 2.20). He turned to using concrete blocks as permanent
shuttering, the blocks, some of which used white cement, being made in plaster
moulds. Many houses in California were to be built with this technique.
In the early 20th century in Germany the engineer–contractors, Wayss &
Freytag, and Dyckerhoff & Wildmann, were busy constructing many industrial
buildings with a variety of roof forms. In 1913 Dyckerhoff & Wildmann, with
the architect Max Berg, built an impressive exposed concrete Centennial Hall in
Breslau (Figure 2.21) with a dome 213 ft in diameter supported on a series of arches
and clearly exhibiting the plasticity of concrete. At the same time a new movement

Figure 2.20 Unity Temple,


Oak Park, Chicago, 1906.

Figure 2.21 Centennial Hall,


Breslau, 1913.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:12:21
02.qxd 15/11/2001 15:43 Page 27

The innovative uses of concrete by engineers and architects 27

in architecture emerged, influenced by the huge new exposed concrete industrial


structures such as warehouses and silos which were the result of scientific method
and economy. Gropius, Le Corbusier and Mies van der Rohe were the major expo-
nents. In 1911 Walter Gropius led the way with his Fagus factory in Anfeld
a.d.Leine, which has a reinforced concrete frame with stub cantilevers on which
the steel and glass skin was supported. He used a similar construction in 1925 for
the Bauhaus in Dessau. Le Corbusier spent a short time in Perret’s office in
1909–10 and in 1915 published a sketch of his ferro-concrete skeleton for dwellings
which formed the basis of later designs by his contemporaries. A good survey of
architectural design in concrete was given by T.P. Bennett in 1927.17
At the first international conference on reinforced concrete in Liège in 1930,18
Dischinger of Dyckerhoff & Wildman gave details of the development of shells from
his firm’s dome at Jena in 1924 to their cylindrical shell of 1926. Predicted future
forms included northlight shells and square domes similar to those at Reval (now
Tallinn), built in 1916,19 and at Brynmar, constructed much later in 1951. In the
discussion Freyssinet talked about his experience with concrete shells.
One engineer who was absent from the Liège conference was Sir Owen
Williams,20 who had started work in 1911 with the British office of the US Indented
Bar-Engineering Co. He quickly moved to the Trussed Concrete Steel Co. and
became chief estimating engineer at the age of 23. During the 1914–18 war he
designed aircraft and concrete ships. In 1920 Owen Williams set up as a consult-
ant and was appointed structural engineer for the Wembley Exhibition of 1924,
and he later built a number of bridges with Maxwell Ayrton as architect. Williams’s
flair for concrete design is readily seen in his Findhorn, Spey, Crubenmore and
Loch Alvie Bridges on the A9 in Scotland (Figure 2.22). In 1929 at the age of 39
he registered as an architect and took on the responsibility for the design of a whole

Figure 2.22 A9 bridges by


Owen Williams, 1925–26.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:12:21
02.qxd 15/11/2001 15:43 Page 28

28 Historic concrete

building, not just the structure. His Boots ‘wets’ building in Nottingham of 1930–32
with its large-scale economic flat slab structure and its standard industrial curtain
wall on the perimeter was admired by Modern Movement architects. Although
he followed this with various minimum structure buildings for the Daily Express,
Peckham Health Centre and others, the Boots ‘wets’ is still his best achievement.
Like Maillart, he took especial care in moulding the concrete at structural joints
such as the abutments of his bridges and the column heads of his flat slabs.
The International Congress of Modern Architecture (CIAM), set up in 1928,
brought together modern architects such as Gropius, Le Corbusier and like-
minded architects in the UK such as Wells Coates and Joseph Emberton.
Conferences discussed the emerging society and how architects could best
serve its needs. In 1931 at the age of 30 Berthold Lubetkin21 came to England
from Russia via Paris and formed the architectural practice, Tecton, which was
given the commission to design the Gorilla House at London Zoo. Lubetkin had
studied reinforced concrete in Berlin and Paris, where he met Le Corbusier and
Mies van der Rohe, and had attended Perret’s Atelier. While he was designing the
Gorilla House, the Paris office of Christiani & Neilsen, engineering contractors
who specialized in reinforced concrete works, suggested that he should meet
Ove Arup, their chief designer in London. This was the beginning of a long and
fruitful collaboration between Lubetkin, considered the leading modern architect,
and Arup, the engineer–contractor. Arup moved to the contractors J.L. Kier in
1933 as director and chief designer so that, after 11 years of civil engineering
experience, he could also tender for the construction of buildings and thus
associate with Modern Movement architects.
Lubetkin’s earliest buildings were his best. For the block of flats, Highpoint, Arup
persuaded Lubetkin to use external load-bearing walls and floor slabs supported
on spine beams so that there was less structural obstruction in the flats than with
the typical two-way frame. However, some structural gymnastics were necessary
to maintain a standard architectural grid of columns at ground-floor level.
Sliding formwork, originally developed for silo construction, was innovatively
used here by Arup, the contractor, for the exposed concrete external walls of a
domestic building, although later there were problems with weathering owing to
their 5 in thickness.
The Penguin Pool at London Zoo was Lubetkin’s second project and came to
symbolize modern reinforced concrete architecture. With structural daring
Lubetkin elegantly intertwined two spiral ramps, first hinted at by the spiral stair-
cases in Paris. Arup worked closely on the project and approved the structural
form of the final proposal.
The calculations for the ramps were complex and were carried out by Felix
Samuely,22 who also suggested that the section should be trapezoidal, a change which
was agreed to. Samuely had just arrived from Berlin where he had been a
co-partner in the consulting firm of Samuely & Berger, set up by him in 1929 at
the age of 27. Samuely worked with Kier for nine months before again setting up
a partnership, this time with Helsby and Hannam. They were to be the structural
consultants of Bexhill Pavilion, with the architects Mendelssohn (with whom Samuely
had worked in Berlin) and Chermayeff. Being an independent consultant with
experience in concrete and, more importantly, in the new technology of welded
steelwork, he attracted the Modern Movement architects from 1934 to the outbreak
of war. Of particular note were the Wells Coates flats in Brighton and Kensington
(Figure 2.23) and the shallow 100 ft diameter concrete dome at Folkestone for
Pleydell Bouverie. Like Arup, Samuely was dedicated to the integration of structure
in architecture and appreciated that the role of the engineer was to give the archi-
tect the confidence to create a structure for his building. He was senior lecturer in
structures at the Architectural Association, a member of the MARS Group and joint
editor of their Master Plan of London, published in the Architectural Review 1942.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:12:21
02.qxd 15/11/2001 15:43 Page 29

The innovative uses of concrete by engineers and architects 29

Figure 2.23 Palace Gate


flats, Kensington, 1938.

Arup was to some extent impeded by staying with Kier because as a contractor
he had to tender for work rather than having the freedom to offer his services to
architects as an independent consultant. Nevertheless, he joined the Architectural
Association and the MARS Group, where he followed his intellectual pursuits and
formed close friendships with a wide circle of architects. The mathematician and
structural analyst Ronald Jenkins joined Kier in 1935, 2 years after the design of
the Penguin Pool, and followed Arup when he left to set up his own contracting
firm, Arup & Arup, in 1938.

Eduardo Torroja, Torroja was one of the most creative engineers of this centry. His contribution to
the understanding of building structures and their development is comparable to
1889–1961 that of Maillart with his bridges. Frank Lloyd Wright considered him the creator
of organic structures and architecture.
Torroja’s three-dimensional concept of structures can be clearly seen in the roof
of a hospital clinic built in 1926, one of his first projects as a consulting engineer.
The structure is similar to a top hat on an octagonal plan, with a compression ring
on top and a wide brim in tension. His search for economic and pure structures
is evident in his design, also in 1926, for a portal-framed bridge in a cutting, where
the interconnecting slabs between the frames were located in positions where they
acted as compression flanges. Nervi was to use the same idea in his design for the
sophisticated lecture theatre in the Unesco building in Paris in 1952.
Torroja, the son of a geometer and architect, was brought up in the Spanish
atmosphere of brick or tiled vaults which became even more extravagant after the
introduction of Portland cement. Working with an architect, his first notable shell
in 1933 was a 31⁄2 in thick 156 ft span dome at Algeciras supported on eight columns
interconnected by pretensioned ties (Figure 2.24). A more interesting shaped shell
with a span of 180 ft was used for a pelota court (Fronton Recoletos) in Madrid
in 1935. Here he combined latticing in the glazed areas with a solid 31⁄4 in thick
slab, elegantly solving the problem set by the architect Zuaso. The structural design
was confirmed by a model test carried out at the University of Madrid, where
Torroja was professor.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:12:21
02.qxd 15/11/2001 15:43 Page 30

30 Historic concrete

Figure 2.24 Dome at


Algeciras, 1933.

Figure 2.25 Aqueduct at


Alloz, 1940.

Many examples of his work could be given, but one in particular exemplifies
the graceful simplicity of his designs. This is the prestressed concrete aqueduct
at Alloz built in 1940 (Figure 2.25), where the concrete is maintained in constant
compression for watertightness. In his book Philosophy of Structures,23 which should
be essential reading for all engineering and architectural students, he makes a
telling statement: ‘The process of visualizing or conceiving a structure is an art —
it is motivated by an inner experience, by intuition.’

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:12:21
02.qxd 15/11/2001 15:43 Page 31

The innovative uses of concrete by engineers and architects 31

Pier Luigi Nervi, Having had a good academic training and excellent experience with concrete
construction, Nervi set up his own contracting firm at the age of 30. It enabled
1891–1979 him to conceive, design, develop and build his own structures. In 1926 he built a
cinema in Naples which had a three-dimensional top-hat structure, 30 m in diam-
eter, similar to Torroja’s hospital roof construction of the same year.
Built in 1930, the staircase tower at his Florence Stadium, with its spiral flights
stiffened by a reverse spiral beam, is not only economic but it is also concrete sculp-
ture (Figure 2.26). His first large ribbed shell structure was the hangar built
for the Italian air force in 1935 (Figure 2.27). This majestic building of in-situ
concrete covers a column-free area of 350 ft by 132 ft. A model test was used in
the design. A similar hangar, built in 1940, shows a bolder and simpler support
system. The cost and time of construction were reduced by using precast
concrete latticed units which had projecting reinforcement subsequently welded
for continuity.
For long-span roofs, Nervi appreciated that a lightweight structure was benefi-
cial. He first proposed the use of thin slabs of ferro-cement, cement mortar rein-
forced with superimposed layers of mesh and small diameter bars, for an exhibition
building at the Rome World Fair of 1939. Ferro-cement, which is in fact a devel-
opment of Lambot’s ferciment of 1847, has a high degree of elasticity and resis-
tance to cracking. No formwork is required and architectural or organic shapes
are easy to form. During the war, Nervi was commissioned to build boats using
ferro-cement and tests showed that such construction was possible. However, its
first actual use only came in 1945 when Nervi built a 165 t yacht with a hull thick-
ness of 3.5 cm. Its first use in building construction followed in 1946 when he built
a small warehouse for his firm. After much testing by Professor Oberti, he used it
on some large-scale roof structures, including the 94 m span arched Exhibition
Hall in Turin of 1948–49 (Figure 2.28).
This roof was made up from ferro-cement inverted trough units 2.5 m long
and 1.45 m high and with a wall thickness of under 5 cm sitting end to end and

Figure 2.26 Florence


Stadium staircase, 1930.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:12:21
02.qxd 15/11/2001 15:43 Page 32

32 Historic concrete

Figure 2.27 Air force hangar,


1935.

Figure 2.28 Exhibition Hall,


Turin, 1949.

connected with in-situ concrete chords. The system worked well and Nervi went
on to use ferro-cement for all his later roof structures and for permanent
formwork on his next innovative structural form, isostatic ribbed floor slabs
(Figure 2.29), where each rib carries load to the columns according to its relative
stiffness. Their positioning is determined by the designer and therein lies the art

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:12:21
02.qxd 15/11/2001 15:43 Page 33

The innovative uses of concrete by engineers and architects 33

Figure 2.29 Isostatic slab,


Gatti wool factory, Rome,
1953.

of the engineer. Nervi’s art is to be seen in all his structures from the Turin Exhi-
bition Hall to the 1957 Sports Palace in Rome and has been well documented
by him.24 What is of interest is the work he carried out as a consulting engineer
with architects on buildings. Although parts of the structures are exposed and are
recognizably Nervi’s, he seems to have had little influence on the overall archi-
tectural form, which may confirm that a brilliant structure does not necessarily
produce good architecture. However, Nervi has left us with a rich selection of struc-
tures and writings to study.25

Postwar In the continent innovative use was made during the war of prestressed con-
crete units for the construction of bombproof U-boat pens and for motorway
construction in bridges, while in the UK the Mulberry harbours were outstanding.26 The main
Europe problem for the building industry of postwar Europe was the rapid reconstruc-
tion of houses, schools and factories, while for the civil engineers it was
the construction of bridges. Skilled craftsmen and structural materials — steel
and timber — were in short supply, so that industrialized system building of
large, precast concrete floor units and wall panels was used extensively for multi-
storey housing. Precast and prestressed concrete floor and long-span roof systems,
reinforced concrete shells and prestressed concrete bridges all made their appear-
ance, together with new, larger capacity mechanical handling equipment such as
tower and mobile cranes.
Architects had to learn how to use the new structural forms and to express the
aesthetics of large panel construction. They exposed a wide range of aggregates,
tooled the concrete or used special plastic linings to the moulds, and it was found
that white cement and sand became an acceptable colour when wet. Prestressed
concrete floor units spanned up to 50 ft and gave architects a greater flexibility in
the planning of buildings, while the introduction of 30–40 ft span in-situ concrete
continuous coffered flat slabs cast onto plastic pans allowed the concrete to be left
exposed or simply painted.
In 1946, after 24 years of contracting, Ove Arup set up as a consulting engi-
neer, forming his partnership in 1949. Because of his concern for the integration

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:12:21
02.qxd 15/11/2001 15:43 Page 34

34 Historic concrete

Figure 2.30 Hatfield


Technical College, 1950.

of structure in architecture, on which he had widely written and lectured, and


particularly because of his intellect and personality, he attracted architect clients
and outstanding engineers.
Ove Arup & Partners’ answer to multi-storey housing was the in-situ concrete
box-frame construction with industrialized shuttering systems, though they turned
to prefabrication for other structures. At the 1951 Festival of Britain, a showcase
for structural ingenuity, they built a continuous prestressed concrete footbridge
and a prestressed concrete two-way grid roof. The partnership have been con-
sultants on many prestigious concrete-framed buildings such as Coventry Cathe-
dral and the Sydney Opera House, on which Peter Rice gained his practical
experience before becoming the leading engineer-architect of the last 20 years.
In 1950 Felix Samuely built his first and perhaps his most innovative prestressed
concrete structure. For the heavily loaded floors of a printing factory in Bristol,
he designed precast concrete three-hinged frames with prestressed concrete ties.
For ancillary buildings he used prestressed concrete planks as reinforcement for
continuous composite floors. At the same time his structure for Hatfield Techni-
cal College had site-cast large three-hinged roof frames and trussed beams for
the composite floor construction (Figure 2.30). He instigated further research on
composite construction27 and ended up with a series of standard designs which
were checked by full-scale tests similar to those carried out by Ransome in 1889
to confirm his composite ribbed construction.
For roofs, prestressed concrete beams and trusses reached unprecedented spans
in industrial buildings. In the UK A.J. Harris, who had worked with Freyssinet
and who ran the Prestressed Concrete Co., was responsible for the innovative
design of the BEA hangar at London Airport in 1951 (Figure 2.31). Secondary
roof units, made up from precast units post-tensioned together, span 110 ft and
are supported on 150 ft span primary post-tensioned in-situ concrete box beams.28

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:12:21
02.qxd 15/11/2001 15:43 Page 35

The innovative uses of concrete by engineers and architects 35

Figure 2.31 BEA hangar,


London Airport, 1951.

Figure 2.32 BOAC hangar,


London Airport, 1955.

At the same time Owen Williams was designing a hangar for BOAC (Figure 2.32)
with a 336 ft span external in-situ concrete arch with counterbalances supporting
a 140 ft roof structure. On the adjacent site he won a limited competition to build
one of the first stayed roofs in the UK. This was a hangar with double cantilevers
of 110 ft with raking steel ties.
That the development of concrete shell structures across Europe was spectacu-
lar may be seen in the many projects illustrated in the proceedings of the first and
second symposia on concrete shell roof construction in London in 1952 and Oslo
in 1957. Peter Morice and Hugh Tottenham have written on early shell develop-
ment, and Robert Anchor describes Twisteel’s extensive construction in the UK
of north-light shells for warehouses and industrial buildings in the 1950s. British
Reinforced Concrete Ltd also built standard shells, but it was to the independent
consulting engineers, such as Ronald Jenkins of Ove Arup & Partners, that archi-
tects went to discuss their roofing proposals.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:12:21
02.qxd 15/11/2001 15:43 Page 36

36 Historic concrete

Figure 2.33 Assembly hall,


Wigan, 1949.

Figure 2.34 Clifton


Cathedral, Bristol, 1965.

The type and size of shells evolved with notable structures by Heinz Isler in
Switzerland and by Esquillan in Paris. Esquillan’s immense CNIT exhibition hall
in Paris, completed in 1958, is a double shell dome covering a triangular area and
supported at the corners 218 m apart.
In the UK Samuely built his only barrel-vault roof for a factory in Bristol in 1948.
For economy and to provide a wider range of architectural forms, he turned to

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:12:21
02.qxd 15/11/2001 15:43 Page 37

The innovative uses of concrete by engineers and architects 37

Figure 2.35 MODA building,


Jeddah, 1978.

folded plate roofs (or ‘Faltwerke’ as it was known in Germany). Here the shell roof
is made up from a series of planes rather than a curve. These could be made either
of in-situ concrete or of a composite construction of precast units with the
shell reinforcement in the in-situ concrete screed (Figure 2.33). To eliminate top
shuttering for steep slopes he prestressed solid units together.
For a school in London in 1949 he not only used a concrete folded plate for
the balcony structure of the assembly hall, but supported it on a star beam29 which
distributed the load to three columns. This innovative system was used to advan-
tage by F.J. Samuely & Partners for Clifton Cathedral (Figure 2.34) in 1965, six
years after Samuely’s death at the early age of 56. Their roof construction in
Jeddah, intended to carry helicopters (Figure 2.35), clearly shows the influence
of Nervi’s isostatic floor design.

The work of Felix Any history of shell construction would be incomplete without a description of the
Candela (b. 1910) work of Felix Candela,30 an architect by training but with an interest in mathe-
matics and structures. In 1936 he was about to leave Spain on a travelling schol-
arship to visit Dischinger & Finsterwalder and study concrete shells when the Civil
War broke out. He joined the Republicans and in 1939 was deported to Mexico.
After a few years of varied experience with contractors, he built for himself his
first shell in 1949, an experimental ctesiphon shell where drooped hessian between
arches was sprayed with concrete.
Two years later, at the age of 41, Candela set up a contracting organization with
his brother to build shell roofs in which he did the architectural and engineering
design. In 1951 the firm began with a startling building, the Cosmic Rays Pavil-
ion (Figure 2.36), where the brief called for a concrete slab with a maximum thick-
ness of 5/8 in. Candela re-read Aimond’s paper on hypars,31 written in 1936, and
did relatively simple calculations for its design. The building was instantly acclaimed
by architects and he went on to build cylindrical shells, folded plates and conoids
before concentrating on hypars. Because he was responsible for the construction,
he could take risks and experiment. He did not agree with elastic design of
shells and developed his own method of calculation, based on ultimate load design.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:12:21
02.qxd 15/11/2001 15:43 Page 38

38 Historic concrete

Figure 2.36 Cosmic Rays


Pavilion, Mexico City, 1951.

Figure 2.37 Iglesia de la


Virgen Milagrosa, 1955.

His output was prodigious and each structure was a development of the last and
showed a complete mastery of the hypar form. Of particular note is the Iglesia de
la Virgen Milagrosa built in 1955 (Figure 2.37).

Long-span bridges The need for new bridges was more acute in Europe than in Britain. Freyssinet
first built prestressed concrete bridges in France in 1942 and after the war
and tall buildings continued with many over the River Marne using prefabricated units, by which
time prestressing was beginning to be widely accepted by engineers. Many

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:12:21
02.qxd 15/11/2001 15:43 Page 39

The innovative uses of concrete by engineers and architects 39

Figure 2.38 Lake Maracaibo


Bridge, 1957–62.

systems of post-tensioning, following Freyssinet’s cone anchorage, were developed


and described at the various world conferences on prestressed concrete, as were
the varied uses of the system. Fritz Leonhardt was particularly prominent in the
development of prestressed concrete bridges and has written on the aesthetics of
bridges. In Italy, Riccardo Morandi built his first prestressed concrete bridge in
1950. Twelve years later he crossed Lake Maracaibo, Venezuela,32 with a 51⁄2 mi long
bridge with maximum spans of 771 ft which uses a double cantilever system with
prestressed concrete inclined ties (Figure 2.38).
Cable-stayed bridges increased the economic spans of concrete and each year
greater distances are being bridged as reliable concrete strengths increase. Amongst
these, the Ganter Bridge in Switzerland (Figure 2.39) built in 1980 by Christian
Menn, a pupil of Maillart, is much acclaimed. The outer spans are curved on plan
and the diagonal ties were stressed after the concrete had been poured. The main
span is 174 m and the piers rise to 150 m from the valley floor.
The development of tall reinforced concrete buildings, following Ransome’s
16-storey Ingalls building of 1903–1904, is best seen in the USA. Concrete service
cores, enclosing lifts and staircases, provided the stiffness to withstand wind loads
transmitted to them by the floor slabs, which left the external framework to carry
vertical loads only. In 1953 in Chicago, the Marina City apartment block, circu-
lar on plan, reached 60 storeys using this system. At the same time Myron Gold-
smith33 was preparing his thesis on the appearance of tall concrete buildings
(Figure 2.40). He envisaged a system whereby windloads would be carried on a
massive external framework, with wide-centred columns and storey-height beams
which would also carry intermediate floors. Some buildings of this form appeared
later in Europe. Goldsmith was working with Skidmore, Owings & Merrill (SOM)
in Chicago and teaching at IIT where Mies van der Rohe of the Bauhaus was pro-
fessor. Goldsmith and Fazlur Kahn, a structural engineer also at SOM, carried on

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:12:21
02.qxd 15/11/2001 15:44 Page 40

40 Historic concrete

Figure 2.39 Ganter Bridge,


Switzerland, 1980.

Figure 2.40 Myron


Goldsmith’s thesis on the
appearance of tall concrete
buildings, 1953.

with research at IIT on tall buildings from which the idea of shell and core con-
struction evolved. Their Brunswick Building in Chicago (Figure 2.41) is a good
example. As the core is relatively stiffer than the external frame near the ground,
wind loads are gradually transferred to the core so that the external frame only
carries vertical loads. While he was working with Bruce Graham at SOM on the
design of the Two Shell Plaza in Houston (Figure 2.42), Kahn studied the distri-
bution of load in a regular external framework of a shell and core system sitting
on wide centred columns. He used a powerful computer to determine the sizes
of columns so that they were all equally stressed. This indicated that the visual
appearance of a shell core system could be exploited by the architect and that it
would be possible to analyse many different configurations of external columns.
Also in Chicago, which has excellent sand and aggregates, high-strength con-
cretes are being developed and used. Thus the maximum possible height of build-
ings is continually increasing and external latticed concrete frameworks are
beginning to compete with tall steel construction. However, the tallest concrete
construction is not used for buildings but for the latest North Sea oil rigs. This
new problem which has been facing civil engineers is comparable with that faced
by Brunel and Stephenson in the 19th century when designing long-span bridges

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:12:21
02.qxd 15/11/2001 15:44 Page 41

The innovative uses of concrete by engineers and architects 41

Figure 2.41 Brunswick


Building, Chicago, 1966.

Figure 2.42 Two Shell Plaza,


Houston, 1972.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:12:21
02.qxd 15/11/2001 15:44 Page 42

42 Historic concrete

Figure 2.43 Brent Field oil


rig, 1991.

to carry heavy locomotives. Varied designs for rigs in steel and in concrete, each
breaking new ground and advancing technology, have been built, but their design-
ers go unnoticed by society. For the Brent Field, Norwegian contractors have just
installed a 1500 ft high rig in reinforced concrete (Figure 2.43). Further historic
details of the construction is given in Chpater 13.37

Conclusions This short survey of the development of concrete highlights the roles of the engi-
neer and architect. Construction under water and fireproofing of buildings were
their respective technical problems from the Roman period until the discovery of
natural cement and the important invention of Portland cement. Portland cement
concrete was able to carry substantial compressive loads, which made it suitable
for foundations, load-bearing walls and fireproof floors. Reinforcing Portland
cement concrete at the end of the 19th century produced a revolutionary new struc-
tural material which brought about the scale of construction we see today.
Its innovative use by engineers came very early with Maillart and his Tavanasa
Bridge of 1905. His mathematical and structural prowess led him to envisage intu-
itively a concrete box construction for a three-hinged arch bridge. He had his own
contracting firm and so was able to carry the risk of both design and construc-
tion. In 1909 he patented a fully tested flat slab building construction. His breadth
of vision and inventiveness may also be seen in his deck-stiffened arch bridge
system and his Chiasso roof truss, while his Cement Hall in Zurich shows his
artistic virtuosity.
Eugène Freyssinet was also an engineer and contractor whose innovative uses
of concrete in early shells and bridges led him to his momentous concept of pre-
stressing. Nervi and Candela followed in the footsteps of Maillart and Freyssinet
in that they were contractors which allowed them the freedom to develop their
ideas. Nervi invented ferro-cement for lightweight long-span roof construction,
and his artistic flair and writings increased the perspective of design in concrete.
Candela, an architect by training, illustrated the many possible forms of hyper-
bolic paraboloid shells.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:12:21
02.qxd 15/11/2001 15:44 Page 43

The innovative uses of concrete by engineers and architects 43

Eduardo Torroja and Fazlur Kahn were both structural consultants who made
significant innovative uses of concrete but also contributed to a better under-
standing of the nature of structures. Kahn made use of a powerful computer, the
evolving tool for structural analysis, to develop his various schemes for tall build-
ings, with encouragement from architects at SOM in Chicago.
From Hadrian onwards, architects have always been interested in the latest tech-
nology. During the nineteenth century they used cements for renders and plain
concrete to imitate stonework and also to solve the problem of fire resistance.
Perret in Paris pioneered the exposed external framed building, while Frank Lloyd
Wright in 1906 built his first concrete-walled Unity Temple in Chicago. In
Germany in 1913 the 64 m Breslau dome surpassed that of the Pantheon in size.
The Modern Movement architects made use of concrete as a framework onto which
they hung cladding or as a perforated wall, void of decoration. As reinforced con-
crete was a new material without a tradition of knowledge or experience, archi-
tects had to turn to specialist engineers for advice. When shells and then prestressed
concrete appeared, the advances in their use for building were made by close part-
nerships of mutually appreciative architects and engineers.
Architect–engineer collaboration is essential for concrete architecture to
progress, but for civil engineering the architect-engineer Santiago Calatrava, who
spans both disciplines, may have the answer. This short study shows that histori-
cally those civil engineers with artistic flair have been responsible for many of the
major innovations in the use of concrete. Surely then, engineering students should
be encouraged to develop their artistic ability and a study of history would given
them a sense of presence in the ever-changing construction industry.

References 1. Mainstone, R.J., Developments in Structural Form. MIT Press: London, 1975.
2. Pasley, C.W., Observations on Limes, Calcareous Cements, Mortars (etc.). John Weale:
London, 1938.
3. Smith, J.T., A Practical and Scientific Treatise on Calcareous Mortars and Cements
(translated from L.J. Vicat’s Résumé des connaissances positives). John Weale:
London, 1837.
4. Grant, J., Experiments on the Strength of Cement, Chiefly in Reference to
the Portland Cement Used in the Southern Main Drainage Works. Spon: London,
1875.
5. Collins, P., Concrete. The Vision of a New Architectury. Faber & Faber: London, 1958,
27–28.
6. Coignet, F., Bétons Agglomérés Appliqués à l’art de Contruire. E. Lacroix: Paris,
1861: 196
7. Collins, P., Concrete. The Vision of a New Architecture. Faber & Faber: London,
1958: 54.
8. Brown, J.M., W.B. Wilkinson (1819–1902) and his place in the history of reinforced
concrete. Trans. Newcomen Soc., 1966–97, XXXIX, 129–142.
9. Hyatt, T., An Account of Some Experiments with Portland-Cement-Concrete Combined
with Iron, as a Building Material. Chiswick Press: London, 1877.
10. Cusack, P., François Hennebique: The specialist organisation and the success of ferro
concrete 1892–1909. Trans. Newcomen Soc., 1984–85. LVI, 71–86.
11. Banham, R., A Concrete Atlantis. MIT Press: London, 1986.
12. Billington, D., Robert Maillart’s Bridges. The Art of Engineering. Princeton Univer-
sity Press: Princeton, 1979.
13. Billington, D., Robert Maillart and the Art of Reinforced Concrete. MIT Press: Cam-
bridge, Massachusetts, 1990.
14. Ordonez, J.A.F., Eugène Freyssinet, 2C edn., Barcelona, 1979.
15. Abeles, P. W., The Principles and Practice of Prestressed Concrete. Crosby Lockwood:
London, 1949.
16. Collins, P., Concrete. The Vision of a New Architecture. Faber & Faber: London, 1958.
17. Bennett, T. P., Architectural Design in Concrete. Ernest Benn: London, 1927.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:12:21
02.qxd 15/11/2001 15:44 Page 44

44 Historic concrete

18. 1930 Premier Congrès International du Béton et du Béton Armé (Mémoires, résumés,
comptes rendus et discussions). La Technique Des Travaux: Liège, 1932.
19. Christiani, Neilsen. Twentyfive Years of Civil Engineering 1904–1929. Copenhagen,
1929.
20. Cottam, D., Owen Williams. Architectural Association (Works III), London, 1986.
21. Allen, J., Berthold Lubetkin. RIBA Publications: London, 1992.
22. Higgs, M., Felix Samuely 1902–59. Archit. Assn J., 1960, LXXVI, No. 843, 2–31.
23. Torroja, E., Philosophy of Structures. University of California Press, 1967.
24. Nervi, P.L., Structures — the Works of Pier Luigi Nervi. F.W. Dodge: New York, 1957.
25. Nervi, P.L., Aesthetics and Technology in Building. Harvard University Press, 1966.
26. The Civil Engineer in War. Institution of Civil Engineers: London, 1948.
27. Samuely, F.J., Composite construction. J. Instn Civ. Engrs, 1952, Feb., 222–59.
28. Harris, A.J., Hangars at London Airport. J. Instn Struct. Engrs, 1952, XXX(10), 226–35.
29. Samuely, F.J., Space frames and stressed skin construction. RIBA J., 1952, 166–73.
30. Faber, C., Candela. The Shell Builder. The Architectural Press: London, 1963.
31. Aimond, F., Etude Statique des Voiles Minces en Paraboloide Hyperbolique Travail-
lant Sans Flexion. Publications IABSE: Zurich, Vol. IV, 1936.
32. Boaga, G., Boni, B., The Concrete Architecture of Riccardo Morandi. Tiranti: Lon-
don, 1965.
33. Goldsmith, M., Buildings and Concepts. Rizzoli: New York, 1987.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:12:21
03.qxd 15/11/2001 15:45 Page 45

3 Concrete and the structural use of


cements in England before 1890

Lawrance Hurst
Synopsis The years between the invention of Roman cement in 1796 and 1890, by when
Portland cement had effectively replaced all its competitors as the binder for mor-
tar and concrete, also saw the development of concrete and its increasing use for
foundations and for fireproof flooring. This chapter reviews that development
and the use of cements in buildings by reference particularly to the Royal
Institute of British Architects’ Transactions which were the forum for discussion
because building design was the preserve of architects, generally unassisted by
engineers.
The next chapter takes over the story with the advent of reinforced concrete
building frames by the specialists who had crossed the Channel and the Atlantic
to take out English patents.

Introduction During the 19th century most buildings were designed and drawn in an archi-
tect’s or surveyor’s office, and their drawings included all necessary details of the
structure. The 1909 London County Council (General Powers Act)1 was the first
statutory document to lay down floor loadings, permissible stresses and a basis of
design and hence it was not until then that consulting engineers started to be con-
cerned with building structures. During the period covered by this chapter, the
Institution of Civil Engineers is not the place to seek guidance to common prac-
tice, but rather the Royal Institute of British Architects. In those days the Insti-
tution of Civil Engineers was concerned with engineering works, leaving all aspects
of buildings to the RIBA.
Certain types of buildings were designed by engineers, such as railway stations,
but they were in those cases generally acting as architects, and parts of building
were designed by specialist engineers — R.M. Ordish and J.W. Grover designed
the Royal Albert Hall dome for Major General Scott, himself an engineer, but act-
ing in that case as an architect. The remainder of the ironwork, the masonry, the
foundations and the fireproof flooring were however designed and detailed in
Scott’s own office. Architects undoubtedly had assistance with their structural
designs but that assistance is seldom identified, unless it was obtained from the
supplier of the iron beams or the fireproof flooring or the reinforced concrete com-
ponents. Foundations, masonry, unreinforced concrete and similar non-specialist
components were detailed by the architect, any design necessary being either based
on experience of not unsatisfactory behaviour or by reference to published for-
mulae, which were generally derived empirically from tests.
It is notable that the first paper in Volume 1 of the Transactions of the Institute
of British Architects is George Godwin’s prize essay upon the Nature and properties
of concrete, and its application to construction up to the present period,2 which is followed
by notes on the concrete foundations at Westminster New Bridewell3 and on con-
crete underpinning at Chatham Dockyard4 and later on M.I. Brunel’s experiments
on reinforced brickwork.5 These papers and others from the Transactions of the
RIBA are referred to later in this chapter.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:14:11
03.qxd 15/11/2001 15:45 Page 46

46 Historic concrete

Binders and Cements


aggregates Today we know what is meant by the terms concrete, mortar and cement: unless
qualified by additional words, cement means Portland cement, mortar means a
mixture of Portland cement and sand, and concrete is made with coarse and fine
aggregates and Portland cement. Before and during most of the 19th century such
certainty did not exist, and until 1790 the distinction between cements which set
or did not set under water, termed hydraulic or non-hydraulic, was not known,
or if known was not really appreciated. Any limes burnt from limestone with a
small clay content that would have had some hydraulic properties were unlikely
to be appreciated because limes were invariably slaked and that would have killed
those hydraulic properties at birth. Mortars and renders which needed hydraulic
properties, perhaps for laying bricks in frosty weather or particularly exposed walls,
incorporated terras or crushed brick to give a quicker set.
John Smeaton’s experiments to discover a suitable mortar for use in the
construction of the Eddystone Lighthouse were reported in his Narrative, first
published in 1791,6 where he set out the properties of ‘water cements’ and
distinguished between cements which set by chemical action, under water, and
those which merely hardened, by carbonation and evaporation, and which
dissolved when placed in water. He deduced that ground volcanic rock — either
puzzolana (now usually spelt pozzolana) from Italy or terras (also known as trass
and terrace) from Germany via Holland — were needed as a constituent of a water
cement. His recipes never came into general use, because in 1796, James Parker
patented his Roman cement.7,8 Parker’s Roman cement, also known as Frost’s, was
a quick setting artificial hydraulic cement, ‘artificial’ meaning manufactured,
not ‘natural’ as were the mined volcanic hydraulic cements. It was used for water-
proof renders and for mortar for laying bricks in wet conditions during the first
half of the 19th century, until such time as true Portland cement became
available in 1845–50. Other patent cements were also used during the same
period principally for exterior renders designed to imitate stone and for water-
proofing.9,10
‘Cement mortar’, as used by Sir Robert Smirke in the brickwork above the lime
concrete underpinning to Custom House in 1826 and as used by Mark Isambard
Brunel in the Thames Tunnel was made with Roman cement.11 The remarkable
watertightness of the Thames Tunnel is a tribute not only to the workmanship but
also to the use of Roman cement, without which it would not have been possible.
Robert Stephenson made a rule never to use lime mortar in the arches of tun-
nels, but to build them with (Roman) cement exclusively12 (Figure 3.1). Robert
Smirke also used three or four courses of brickwork laid in Roman cement as chain
bond in rubble walling at Maidstone County Court House.13
Mortar or render made with Roman cement is generally a brownish colour, and
is much harder than the white or dirty grey lime mortar of the period. It was used
for precast embellishments to be incorporated with stucco and for mortars where
its quick setting or strength or hydraulic properties were needed. It was however
seldom mixed with fine and coarse aggregates to make concrete, because it set too
quickly. References to Roman cement are to be found up to 1880, when it was still
significantly cheaper than Portland cement, but not much evidence of its use is to
be found in those later years.
Portland cement was the term used by Joseph Aspdin in his patent of 1824, but
it was not until about 1845 that I.C. Johnson, the manager of the cement works
at Swanscombe, Kent, burnt the new materials at a high enough temperature to
produce what we now know as Portland cement,14 which only slowly replaced
Roman cement in mortars and renders, because it was much more expensive, but
with a few exceptions it was not mixed with aggregates to make concrete for use
in buildings until about 1865.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:14:11
03.qxd 15/11/2001 15:45 Page 47

Concrete and the structural use of cements in England before 1890 47

Figure 3.1 A method of testing and comparing cements used by Pasley (p. 79 in
Ref. 11) and others . Neat (Roman) cement was used and each brick held up by press-
ing the point of a trowel firmly against it for 2 or 3 min. The next brick was similarly
set 6 or 7 min later. Francis White & Co., the cement manufacturers, repeatedly struck
out 29 bricks in this way before the joints gave way.

In 1850 Dobson in his Rudimentary Treatise on Foundations and Concrete Works,15


which is basically on engineering works, not buildings, is quite clear — ‘concrete
is made of gravel, sand and ground lime, mixed together with water’ and ‘beton’
is concrete made with hydraulic lime. Neither Parker nor Frost nor Roman cement
appear in the index. Others define concrete as a combination of aggregates, cement
and water mixed at the same time, but beton meant mixing the sand, cement and
water together first before adding it to the coarse aggregates. Concrete figures
largely in the discussion following a paper read to the RIBA in 1875 on New mate-
rials and recent inventions connected with building16 but in many instances it is not
apparent if Portland cement or lime was used as a binder, such was the confusion
of terminology in general use at that time. It is therefore necessary to be partic-
ularly cautious in making deductions from descriptions of ‘concrete’ or ‘cement’
in 19th century references. In some particular uses gypsum was used as the binder
for mortars in concretes — for flooring in the Nottingham area and for concrete
in Dennett’s floors.

Aggregates

Whilst sand was initially mixed with lime or cement, Roman or Portland, for mor-
tar, a number of materials were used for coarse aggregate to make concrete.
Lime concrete for footings and underpinning was generally made with gravel
or ballast, but the fire proof flooring systems variously specified coke breeze
(obtained from gas works), cinders, broken brick, burnt clay, limestone and even
clean sieved rubbish. Rubbish was the term used to describe the arisings from tak-
ing down old buildings and was therefore a mixture of broken brick, plaster and
mortar.
There was a theory that concrete made with aggregates that had gone through
fire performed better in a fire than that made with those which had not, hence
the use of breeze, brick and burnt clay as aggregate for concrete for fireproof floor-
ing. Crushed brick is a good aggregate for concrete but breeze, which was first
introduced by Matthew Allen in 1862,17 is much more commonly found. Breeze
is liable to be incompletely burnt and to contain a proportion of coal, giving a com-
bustible content to ‘fireproof ’ floors. Breeze also tends to be acidic, because of the
sulphates and sulphides originating in coal, and hence can actively promote rust-
ing of embedded iron or steel in damp conditions.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:14:11
03.qxd 15/11/2001 15:45 Page 48

48 Historic concrete

A solution to The history of concrete foundations is covered in Chapter 7 by Chrimes, but it is


foundation necessary to mention its early use here, if only because of the use of cement mor-
tar in the brickwork. There is little doubt that Smirke’s was the first effective use
problems since Roman times of concrete in foundations, when he used lime concrete in his
remedial works for the sinking foundations at Millbank Penitentiary (1817–22)18
and for the underpinning following the partial collapse of the London Custom
House (1825–27)19,20 and elsewhere.
At Custom House he excavated 12 ft wide and 12–15 ft deep, down to a bed of
gravel, and filled the trenches with lime concrete, on top of which he bedded large
York stone slabs or landings, and then pinned up to the old footings with slate
above 12 courses of hard stock bricks laid in (Roman) cement mortar (Figure 3.2).
The lack of any sign of subsequent movement in the superstructure is a testament
to the success of his underpinning work.
In following years, Smirke continued to use lime concrete as an artificial foun-
dation when building on bad ground, sometimes as a raft, as in the rebuilt cen-
tre of Custom House, and sometimes as a footing, doubtless in conjunction with
brick footings laid in Roman cement mortar.
Occasionally a lime concrete footing is found in buildings on a smaller scale,
but that must be regarded as a exception and would only have been considered
in particularly soft ground; normally the brick footing courses were laid directly
on the earth, perhaps incorporating chain bond timbers to help bridge over soft
spots.
Portland cement concrete footings started to be use for large and important
buildings in the 1860s but did not come into general use until the 1880s and even
later for domestic buildings.

Figure 3.2 Section through


Robert Smirke’s underpinning
of Custom House in 1825.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:14:11
03.qxd 15/11/2001 15:45 Page 49

Concrete and the structural use of cements in England before 1890 49

The early use Plaster floors for upstairs rooms were used in the east Midlands from Elizabethan
times up to the 19th century, with straw or more commonly reed laid over the joists
of mortar and and then covered with 2 in or more of plaster or mortar, made with gypsum or
concrete in occasionally lime, mixed with burnt clay or broken brick or ashes21–23 (Figure 3.3).
flooring Plaster floors were also to be found in the Cotswolds but survivals are more to be
expected in the Nottingham area, where they continue to serve their purposes,
but are known to be susceptible to point loads from bed legs, baths and pianos.
Recipes for plaster floors vary from neat coarse gypsum, to the use of the waste
mixture of lime and ashes from the bottom of a lime kiln, to one-third lime, one-
third coal ashes with one-third loamy clay and horse dung.
Lime ash concrete was shown by John Foulston as a finish to the stone slabs car-
ried on cast iron inverted tees for the suspended floors of Bodmin Asylum (1818)24
and it appears probable that this is the germ of the idea developed by Henry Hawes
Fox for the fireproof flooring he invented for the private asylum he built at
Northwoods, Nr. Bristol, in 1833.
Fox’s Patent of 184425 for the floor, which was developed and marketed by James
Barrett and hence became known as Fox and Barrett, and is found in very many
buildings up to 1885 or even 1890, consists of cast iron inverted tee joists at about
1 ft 6 in centres with stout timber laths spanning between the outstanding flanges
to provide formwork for 8 : 1 : 1 (sieved rubbish or road grit : coal ashes : lime) con-
crete or pugging (Figure 3.4). The surviving floors of Northwoods show that this
is a far stronger material than would be expected from the weak mix specified.
The concrete was laid on 3⁄4 in of 1 : 2 lime mortar pushed through the gaps between
the stout laths to provide a key for the plaster on the soffit which completed the
encasing to the cast iron tees and provided the first truly fire resisting floor, since
previously the iron joists supporting brick arches or stone slabs had been exposed
on the soffit and thus vulnerable to fire. Barrett was not only one of the first to
appreciate the need to provide protection against fire to the undersides of the iron

Figure 3.3 Lime ash on reed


flooring (by courtesy of Philip
Hartley of SPAB).

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:14:11
03.qxd 15/11/2001 15:45 Page 50

50 Historic concrete

Figure 3.4 Fox & Barrett


flooring with cast iron joists,
before 1851 or 1852 (p. 163 in
Ref. 76).

joists, but he was also the first to appreciate the value of composite action, when
he wrote, in his paper to the ICE in 1853:
‘… the force of compression acts upon the joists, only through the medium
of the concrete, and this material is well known to be of the best for
resisting that force.’
He also appreciated the ‘considerable accession of strength’ by building the ends of
the joists firmly into the walls.26
After the patent had expired in 1859, the Fox and Barrett system came into gen-
eral use because its components — iron joists, timber laths and concrete — were
readily available to anyone, and the above principles having not been clearly enun-
ciated were not perpetuated. An ordinary lath and plaster ceiling was used, with
battens interposed between the ordinary laths and the stout laths to accommo-
date the bottom flanges of the wrought iron joists, and the finish was timber boards
on small joists resting on the top flanges, left exposed above the concrete, which
was then little more than pugging. It is in this form that Fox & Barrett flooring
is more likely to be found (Figures 3.5 and 3.6).
A builder from Nottingham, Charles Colton Dennett also developed a concrete
fireproof flooring system using tried material when in 1857 he patented ‘arches com-
posed of sulphate of lime and an artificial puzzolana of burnt clay or Porous cin-
ders’ effectively gypsum concrete, as was already in use for plaster floors in his home
town.27 His arches however sprung off cast or wrought iron inverted tees with bot-
tom flanges directly exposed to fire and hence were no more fire proof than a sim-
ilar floor with brick arches, but they were marketed as being less costly (Figure 3.7).
As the 19th century progressed losses of buildings in fires produced greater and
greater pressures to make buildings fireproof and indeed a series of Acts of Par-
liament sought to enforce the requirement, initially for parts of buildings needed
as means of escape and latterly for certain types of buildings. However, there was
considerable discussion and confusion about what constituted fireproof or fire
resisting construction, with a school of thought which regarded incombustible as

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:14:11
03.qxd 15/11/2001 15:45 Page 51

Concrete and the structural use of cements in England before 1890 51

Figure 3.5 Fox & Barrett


flooring with wrought iron
joists, after 1851 or 1852
(p. 249 in Ref. 26).

Figure 3.6 Fox & Barrett


flooring with battens and
plaster on ordinary laths, at
Finsbury Barracks (1857).

synonymous with ‘fireproof ’, generally citing satisfactory behaviour probably in


small fires which were insufficiently hot to cause failure of the exposed structural
metal. This led to the Fire Brigades being unwilling to enter burning ‘fireproof ’
buildings because of the risk of abrupt and catastrophic collapse when the exposed
cast or wrought iron columns and beams succumbed to the heat.28

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:14:11
03.qxd 15/11/2001 15:45 Page 52

52 Historic concrete

Figure 3.7 Dennett’s arch


floor.

There was nevertheless an increasing market for fireproof flooring as the num-
ber of patents for novel ways of solving the problem economically displays. Almost
all of the patents used Portland cement for screeds and bedding and many incor-
porated Portland cement concrete. Further details of many of the systems can be
found in Sutcliffe,29 Potter,30 Webster’s paper to the ICE,31 Farrow’s BFPC Red
Book,32 and Lawford’s paper to the Society of Engineers.33

Hoop iron in Hoop iron — a thin strip of wrought iron from 11⁄2 in  15 gauge (1.8 mm) to
3
⁄4 in  20 gauge (1.0 mm)34 termed hoop because of its use by coopers for bind-
brickwork and ing barrels — was in general use as bond in brickwork, from about 1830 to the
concrete end of the 19th century, as a replacement for bond timbers, particularly chain bond
timbers, in the centre of walls, which suffered from rot whatever timber was used.35
Advocates of in-situ concrete walling adopted it to control cracking and one rec-
ommended ‘a prodigal expenditure of hoop-iron bond’.36
It was that material which Brunel and Francis adopted for their experimental
brick beams in 1835–38. Francis and Sons at their Vauxhall (Roman) cement works
and Brunel at the Thames Tunnel site both built beams of brickwork laid in neat
Roman cement with hoop iron bond laid in the bed joints and tested them to
failure. Pasley made his own tests at Chatham and also tested an unreinforced beam
and a reinforced beam built with lime mortar for comparison. When his four course
high brick and cement beam incorporating five strips of hoop iron was examined
after gradual collapse the centre strip was found to be bent, the two lower strips
had fractured and the two upper strips were buckled and the bricks crushed.
The unreinforced beam failed abruptly at a small load and the iron strips in the
lime mortar beam were ‘drawn inwards’, i.e. the bond with the mortar failed
(Figure 3.8).
This clear understanding of the behaviour of reinforced masonry, as reported
by Pasley in his ‘Observations’ published in 183837 was not adopted and applied
practically by the designers of buildings and does not seem even to be mentioned
in the Transactions of the RIBA, the forum for discussion about building
construction.
This is perhaps surprising considering the publicity that surrounded Brunel’s
works at the Thames Tunnel and where he had also built ‘two extraordinary semi-
arches for experiment’ of brickwork in neat (Roman) cement with hoop iron laid
in the courses cantilevering either side of a central pier and rising 10 ft, one 60 ft
and the other over 37 ft long and loaded with 62,700 lb38 (Figure 3.9).
The hoop iron that engineers of today are likely to encounter will be as bond
in brickwork, as could be seen in 1994 in quantity on the Mansion House Square
site at No. 1 Poultry in the City of London, hanging down from the back of the
Cheapside elevation where the party walls had been demolished (Figure 3.10). In
Thomas & Frank Verity’s specification for the new building to be erected at 96
Piccadilly in 1890, the contractor was to allow for providing five tons of hoop iron
in the brickwork, which equates to something between 6 and 20 miles of the mate-
rial, and incidentally he was required to use lime concrete footings for the walls
but cement concrete foundations for the cast iron stanchions.39

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:14:11
03.qxd 15/11/2001 15:45 Page 53

Concrete and the structural use of cements in England before 1890 53

Figure 3.8 Pasley’s 10 ft


span brick beams reinforced
with hoop iron (pp. 234, 236,
238 in Ref. 11). (a) Reinforced
Roman cement beam failed
with 4532 lb. (b) Reinforce-
ment lime mortar beam failed
with 742 lb (unreinforced
Roman cement beam failed
with 498 lb).

Figure 3.9 M.I. Brunel’s


reinforced brick cantilevers
(p. 38 in Ref. 77).

Flat floors without This quotation is taken from the heading of the chapter in George Sutcliffe’s book
on Concrete its nature and uses40 published in 1893 on flat concrete floors without
iron — ‘Bold; be iron, of the sort that engineers today would not contemplate building and,
not too bold’ if they know of them, would dread encountering in a survey or appraisal of a
Victorian building for fear of knowing what to say. Is over 100 years of not unsat-
isfactory use a reason to recommend or permit retention of a floor which cannot
be shown by calculation to be structurally sound and which would collapse abruptly
and catastrophically if anything occurred to disturb its abutments?
There were at that time a number of architects and engineers who held that
steel or iron joists or reinforcement in floors constituted unnecessary expense and
indeed some who held that they were a positive disadvantage.41 There are reports
of plain concrete floors 10 ft  10 ft  4 in thick and even 26 ft  20 ft  7 in thick
which behaved satisfactorily and stood up to tests. Others built shallow unrein-
forced concrete arches and even cantilevers. Col. Seddon’s reported tests included
an unreinforced slab 14 ft 6 in  13 ft 6 in clear span  6 in thick which broke
suddenly under a weight of 10 t or 120 lb/ft2.42

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:14:11
03.qxd 15/11/2001 15:45 Page 54

54 Historic concrete

Figure 3.10 Hoop iron bond


in party wall at Mansion
House Square site in 1994.

Most of these slabs had good bearings on thick solid walls and many were of
concrete made with broken brick aggregate, thus ensuring a tenacious material.
Sutcliffe goes on to say:
‘They are, however, economical, and there is no reason why they should
not be used in houses and other buildings where they will not be subjected
to intense heat or heavy impingent loads, up to spans of 10, 12 or even
(with care) 14 feet.’43
It is probable that some of these unreinforced floors are still extant, serving their
purpose without revealing that they cannot be justified to modern standards —
to engineers appraising Victorian buildings Sutcliffe’s adage could be changed to
‘Beware’! (examples are listed in Table 3.11).
Another form of unreinforced construction which does certainly exist in some
quantity is tile creasing — flat floors and arches formed with up to three layers of
ordinary flat clay roof tiles bedded in neat cement or strong cement mortar and
laid to break bond (Figures 3.11 and 3.12).
Tile creasing is generally held to have been invented by Charles Fowler in about
1835 where it was used for the floors and roofs of Hungerford Market, spanning

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:14:11
03.qxd 15/11/2001 15:45 Page 55

Concrete and the structural use of cements in England before 1890 55

Table 3.1 Examples of concrete floors without reinforcement14

No. Length Breadth Thickness Where used Authority Remarks


(ft) (ft) (in)

Flat slabs
1 20 20 14 Atrium Vestae, Rome Prof. Middleton Ancient Roman work
2 6 6 3 Footway of bridge F. Caws
3 14.5 7 6 Brigade depots Col. Seddon 1 P.c. to 4 breeze, slag, brick or
burnt clay, to pass 3⁄4 in mesh
4 10 10 4 – F. Caws –
5 12.5 11.5 6 – C.A. Adams –
6 21 12.5 13 Warehouse F. Caws 1 P.c. to 4 brick
7 26.5 19.5 7 – F. Caws Thicker for 9 in around margin,
1 P.c. to 4 brick
Arched slabs
8 – – 9 12 to 5 Stables W.B. Wilkinson
9 – 9.5 11– 4 Corridor W.B. Wilkinson Granite-concrete
10 50 12 11– 3 Warehouse Broughton Carrying ‘immense weight of
machinery and men’
11 – 12 15– 6 Pantechnicon Lockwood –
12 19 13 11– 41⁄2 House J. Tall –
13 40 16 11– 41⁄2 Drawing room J. Tall –
14 70 16.5 7– 31⁄2 Roof of barn Trench Intrados rising 1 in ft of
span — 1 P.c. to 5 gravel
15 90 20 15 Roof of warehouse W.C. Street Segmental arch with 3 ft rise —
P.c. to 5 Thames ballast
Cantilevers
16 – 4 8– 3 Balcony Potter 1 P.c. to 5 brick
17 50 4 11– 3 Balcony Broughton 1 P.c. to 4 ballast (part, 1 to 6)

P.c.  Portland cement.

Figure 3.11 Flat roof of plain


tiles and cement known
as tile creasing (p. 175 in
Ref. 77).

between iron beams 41⁄2 ft apart to form the terrace,44 but Pasley in 1838 says it
was first proposed by ‘the late ingenious Mr. Smart’ (George Smart, d. 1834).
Brunel used it for supporting the garden above his drawing office where cast iron
beams at 5 ft centres satisfactorily supported tile creasing spanning 5 ft.45
That form of construction, flat and arched, can be found today in entrance hall
floors, flat roofs, porch roofs, balconies and vaults of early and mid-Victorian houses
built in London. It was also used and still exists as an arched non-combustible fire
break in the House of Lords’ roof, in ceilings to timber floors on cast iron beams
in the Palace of Westminster and in the Waterloo Building at the Tower of Lon-
don, and no doubt elsewhere.

Practical William Boutland Wilkinson’s patent of 185446 is generally agreed to be the begin-
application of ning of reinforced concrete in England. His coffered concrete floor reinforced with
old colliery ropes in the ribs in the reinforced concrete cottage he built in Newcas-
reinforcement to tle in about 1865 (Figure 3.13) as described in detail in the careful records made
floors during its demolition in 1954,47,48 is an example of his work. Wilkinson’s patent also
cites the use of hoop iron as reinforcement. The papers and discussion at RIBA meet-
ings in the 1860s and 1870s, when the use of concrete for building was discussed
at length49–56 indicate however that neither Wilkinson’s patent, nor Lambot’s nor
Coignet’s patents of the following year were adopted by the building industry.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:14:11
03.qxd 15/11/2001 15:45 Page 56

56 Historic concrete

Figure 3.12 Underside of


back addition roof formed in
tile creasing, in c.1840
terrace house in Paddington
area.

Figure 3.13 Wilkinson’s


reinforced concrete cottage.

Reinforced concrete floors of the 1860s and 1870s were not distinguished from
the other forms of fireproof flooring which were being developed to satisfy the
demand arising out of public concern about the increasing cost of damage and
the loss of life caused by building fires and enable compliance with the resulting

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:14:11
03.qxd 15/11/2001 15:45 Page 57

Concrete and the structural use of cements in England before 1890 57

Figure 3.14 Ordinary filler


joist floor.

Figure 3.15 King floor, with


fireclay tubes, usually 3 ft
long  4 to 8 in deep, but the
tubes have thinner walls than
indicated in this plate from
King’s 1935 brochure.

legislation. The majority of fireproof flooring systems incorporated rolled or fab-


ricated iron joists at 1 ft 6 in to 3 ft centres supporting various forms of cellular
clay tiles or concrete usually made with breeze aggregate, sometimes containing
a particular form of reinforcement.
These floors lead in the 1890s to the ordinary filler joist floor — with iron or
later steel joists at about 3 ft centres embedded in a slab of breeze concrete — which
still survives in quantity and will be frequently encountered by engineers working
on buildings up to the First World War (Figure 3.14).
The other type of fireproof floor, which continued in use during the 1920s and
can indeed be found up to the Second World War, is formed with hollow clay tubes
or lintels spanning between steel joists, with the ends notched to embrace the
bottom flange and finished with concrete, usually with breeze aggregate, or with
screed; examples of this were by J.A. King (Figure 3.15), Homan & Rogers
(Figure 3.16), and Fawcett (Figure 3.17).
Thaddeus Hyatt, who crossed the Atlantic to England in the early 1870s and
was responsible for over 40 patents concerned with pavement lights and concrete
flooring, had a series of tests carried out by Kirkaldy, and published in 1877 a
pamphlet setting out why reinforcement was only needed in the bottom of a Port-
land cement concrete slab and why all iron needed to be protected by concrete
cover from fire.57 The need to provide cover to protect iron was stated as early
as 1838 by Pasley, who suggested a coating of tiles laid in cement,58 and in 1853
by Barrett who referred to the plaster to protect the iron joists from the action of
fire from below,59 but it was not generally understood until the 1870s and 1880s
that it was not sufficient just to provide a non-combustible structure to achieve fire
resistance. Hyatt and the tests he had carried out certainly helped this understanding.
The need for an appreciation of construction that could really be shown to

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:14:11
03.qxd 15/11/2001 15:45 Page 58

58 Historic concrete

Figure 3.16 Homan &


Rogers floor as sketched by
the late B.L. Hurst in the late
1890s.

Figure 3.17 Fawcett’s floor


as sketched by the late
B.L. Hurst in the late 1890s.

survive a fire was one of the reasons for the formation of the British Fire Prevention
Committee in 1897 by Edwin Sachs. The BFPC’s first testing station was in use less
than a year later to carry out instrumented documented fire tests on a variety of
building materials.60 The development of this story leading to the formation of

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:14:11
03.qxd 15/11/2001 15:45 Page 59

Concrete and the structural use of cements in England before 1890 59

Figure 3.18 F.G. Edwards


reinforced concrete joists
with nailing strips as illus-
trated in Patent No. 2941,
1891.

the Concrete Institute which became the Institution of Structural Engineers is told
by Anita Witten in Chapter 14.
Other patents were taken out for what was effectively reinforced concrete, by
Gen. Scott,61 of Royal Albert Hall fame, Matthew Allen,17 who in 1862 was one of the
first to advocate Portland cement and breeze concrete, F.G. Edwards,62 for I section
joists with a breeze concrete nailing strip in the top (Figure 3.18), and others; but
none of them came into general use or can be expected to have survived in any
quantity.
However, it would not be unexpected to find examples of the two floors
mentioned earlier — Fox & Barrett and Dennett (or later Dennett & Ingle) in use
in buildings constructed before 1890.

In-situ walls

The other type of patent in the lists for the 1860s, 1870s and 1880s for concrete
work is for apparatus for forming concrete walls, taken out by Tall (7), Drake (3),
Payne (2) and others. These were for formwork systems and sometimes included
loose descriptions of reinforcement and recipes for concrete — there is even one
system of climbing formwork! Most were more ingenious than useful and do not
appear to have benefited their inventors or the industry. Their products do not
appear to have survived in any quantity. They were generally advocated as a less
costly way of providing ‘workers’ dwellings’, a philanthropic tendency of that time,
and hence it would not be totally unexpected to find mid and late Victorian small
houses with in-situ concrete walls. Anerley New Church, now converted to hous-
ing, by W.J.E. Henley, manager of the Concrete Building Company (1883), a
remarkable neo-Gothic in-situ concrete building shows that it is always wise to be
alert for the unexpected.63
Discussions at the RIBA were a dialogue between the disciples of concrete and
the sceptics, who thought that even if it remained standing when the formwork
was struck it would crumble away within a year or two. There was probably right
on both sides, because they were generally talking of different materials — the
sceptics had in mind examples of concrete made with lime or defective cement
and the disciples carried out tests and supervised the work closely to ensure a sat-
isfactory product. It is probable that some examples of the latter survive, such as
63 Lincoln’s Inn Fields designed by William Simmons, with in-situ mass concrete
walls and Hyatt’s floors64 (Figures 3.19 and 3.20).
Official acceptance of in-situ concrete as a material for walls came in clause 2a
of the Bye-Laws made in 1886 under the provisions of the Metropolis Management
and Building Acts Amendment Act 1878, where it could be substituted for brick-
work of the same thickness. The concrete had to be of Portland cement, clean sand,
and clean ballast, gravel, broken bricks, or furnace clinker, passing a 2 in ring, in
proportions 1 : 2 : 3, carefully mixed with clean water, and carried up regularly, in
parallel frames of equal height.65

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:14:11
03.qxd 15/11/2001 15:45 Page 60

60 Historic concrete

Figure 3.19 63 Lincoln’s Inn


Fields with in-situ mass
concrete walls and Hyatt’s
floors.

Figure 3.20 Hyatt’s


‘gridiron’ floor (p. 62 in
Ref. 28).

Precast walls and precast dressings to brickwork

It is not generally realized that a substantial proportion of the bay columns with ornate
capitals and other painted decorative features on the elevations of late Victorian
and Edwardian speculative housing are of precast concrete — an economical and
more durable substitute for the soft easily carved stone alternative — nor is it
realised that the use of precast components in buildings pre-dates most other uses
of concrete.
From the early years of the 19th century, stucco decoration was frequently
precast, using Roman cement,66 rather than being formed in-situ, and, in 1832
and 1834 William Ranger patented67–69 and used precast concrete blocks, made
with lime slaked with boiling water, for building dock and sea walls and a num-
ber of buildings in Brighton and London and possibly elsewhere. It is probable

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:14:11
03.qxd 15/11/2001 15:45 Page 61

Concrete and the structural use of cements in England before 1890 61

Figure 3.21 Lascelles’


system at Central Buffet.

that they might still be found behind stucco on terraces in Brighton of that date.
Ranger’s concrete also had military applications as described by John Weiler in
Chapter 19. Precast concrete, or as it was generally termed ‘artificial stone’, con-
tinued to be used, as whole blocks, or probably more commonly facings to mass
concrete walls, usually with dovetail strips or ribs to bond them to the in-situ work.
Castle House, Bridgewater, dating from the 1851, incorporates precast concrete
facing panels, an interesting and possibly unique form of iron and tile floor, and
a pitched roof with stretcher bond crucks probably laid in neat Roman cement.70
In 1995 there was an opportunity to inspect an ingenious precast system, patented
by William Lascelles in 1875, used in the construction of the Central Buffet and
Dock Managers Office at the Royal Albert Dock in London’s Docklands whose
elevations have been conserved and restored.71,72
Lascelles’ system73 employed 3 ft  2 ft breeze concrete slabs, screwed to timber
studs. They were 1–11⁄2 in thick, reinforced with two diagonal wrought iron rods.
Similar panels could be used on the inside in lieu of plaster, as they were in the
ceilings of houses at Sydenham,74 and on the floors in lieu of boards, but were
not at Royal Albert Dock. Some of the wall slabs at Royal Albert Dock had a rough-
cast finish but they could be cast to resemble brickwork or tile hanging and be
coloured accordingly. The Central Buffet also incorporates a further layer of pre-
cast components outside the slabs, fixed through them to the timber framing, of
columns, with bases and ornamental capitals, friezes and decorative panels, and
a precast concrete cornice incorporating the gutter (Figure 3.21).
Lascelles’ concrete could be self-coloured — red, buff or grey. Red, obtained by
lining the mould with a grout of ‘Spanish brown’ was used at the Royal Albert Dock

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:14:11
03.qxd 15/11/2001 15:45 Page 62

62 Historic concrete

buildings and the finish on surviving unweathered areas is good enough to be


mistaken for terracotta. Even the fire surrounds and overmantels in the Central
Buffet are Lascelles’ precast concrete.
Lascelles’ brochure lists the range or architectural dressings of all kinds sup-
plied by the company, many no doubt from stock, and it would have been from
workshops such as his that the speculative builder would purchase the bay columns
and other decorative work for speculative housing.
Lascelles’ system is likely to have been chosen for the buildings at Royal Albert
Dock on the grounds of speed and economy. Moulds for precast work could be
made quickly and cheaply in Lascelles’ joinery works so that the precast concrete
components could be delivered to site within weeks of being ordered. In his
contribution to the discussion at the RIBA in 1876,75 Lascelles said the slabs were
removed from the mould in three or four days and were then ready for fixing to
the walls.
To produce such decoration the only alternatives would be terracotta or stone;
the former would take months for pattern and mould making, drying and firing,
while stone would have to be sourced as well as carved, so precast concrete was
undoubtedly both quicker and cheaper.
It is interesting to note that one of the latest concrete repair techniques —
re-alkalization — has been used to conserve a substantial proportion of Lascelles’
units at the Royal Albert Dock buildings.
One of the leading Victorian architects prepared to use this new material was
Norman Shaw, who seems to have been retained by Lascelles to produce typical
house designs, and it was he who designed the elaborate facade of Lascelles’ award-
winning pavilion at the Paris Exhibition of 1878.

Summary of The primary purpose of this chapter has been to introduce engineers of today to
the development of the structural uses in buildings of cements during the 19th
structural uses of century and hence to describe forms of construction they are likely to encounter
cement and when working on Victorian buildings.
concrete in the Until 1890 concrete was not used in Britain for building frames, only for slabs,
19th century lintels and occasionally walls. Most forms of fireproof flooring which had been
developed since 1850 incorporated concrete or screed and by 1890 this would
have been made with Portland cement, which had by then replaced the lime and
occasionally gypsum used as the binder in the 1850s and 1860s.
Fireproof flooring, frequently finished with timber boards on battens, was used
in commercial and institutional buildings but not exclusively except for the stairs,
landings and corridors forming the escape routes. It was also used in the higher
class of domestic buildings but only in the lower floors, or sometimes at ground
and second floor levels — to separate the kitchens from the family living rooms
and to separate the bedrooms from the rooms below.
Reinforced concrete beams and columns would be unusual, except for lintels
although concrete casing to rolled wrought iron beams (steel would be unlikely
before 1890) and occasionally to columns is not uncommon. The concrete would
more than likely be made with coke breeze aggregate. Concrete in floors had either
breeze or occasionally broken brick aggregate. Gravel was uncommon but some
patent systems used crushed limestone.
Two other forms of construction have been mentioned. Hoop iron which can
be expected to be encountered in the brickwork of substantial buildings, some-
times in quantity, sometimes only in a couple of courses in each storey height, and
tile creasing which was used for fireproof flooring in houses and also in external
construction where timber would be liable to rot.
During the second half of the 19th century, a number of people had set down
and published and patented the principles of reinforced concrete, but the forms

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:14:11
03.qxd 15/11/2001 15:45 Page 63

Concrete and the structural use of cements in England before 1890 63

of construction in general use indicated that no real understanding of the behav-


iour of reinforcement in conjunction with concrete existed in general practice. The
ingredients were all there waiting for Hennebique and others to compile the
recipes.

References 1. LCC (General Powers) Act 1909.


2. Godwin, G., Prize essay upon the nature and properties of concrete, and its applica-
tion to construction up to the present period. Trans. IBA, 1836, 1, 1–37.
3. Abraham, R., Concrete used at Westminster New Bridewell. Trans. IBA, 1836, 1, 38–39.
4. Taylor, G.L., An account of the methods used in underpinning the long
storehouse at His Majesty’s Dock Yard, Chatham, in the year 1834. Trans. IBA, 1836,
1, 40–43.
5. Brunel, M.I., Particulars of some experiments on the mode of binding brick con-
struction. Trans. IBA, 1836, 1, 61– 64.
6. Smeaton, J., Narrative of the Building of the Eddystone Lighthouse, London, 1791.
7. Parker, J., A certain cement or terras to be used in aquatic and other buildings and
stucco work. Patent No. 2120, 1796.
8. Thurston, A.P., Parker’s ‘Roman’ Cement. Trans. Newcomen Soc., 19, 1938–39,
193–206.
9. Bristow, I.C., Exterior renders designed to imitate stone. Trans. ASCHB, 22, 1997,
13–30.
10. Pasley, C.W., Outline of a course of practical architecture, Chatham, 1862, but first
issued as lithographed notes in 1826. It is clearly from that earlier date and was not
revised before being printed in 1862, p. 91. (Reprinted Donhead, 2001.)
11. Pasley., C.W., Observations on Limes, Calcareous Cements, Mortars, Stuccos and
Concrete, etc., 1st edn., London, Weale, 1838, 38.
12. Pasley, C.W., 1838, 84.
13. Pasley, C.W., 1826, 184.
14. Skempton, A.W., Portland cements 1843–1887. Trans. Newcomen Soc., 1967, 35, 117–52.
15. Dobson, E., A Rudimentary Treatise on Foundations and Concrete Works, London,
Weale, 1850. (Reprinted Bath, 1970.)
16. Smith, T.R., New materials and recent inventions connected with building. Trans. RIBA,
1874–75, 199–216, 221–30.
17. Allen, M., Stairs, etc. Patent No. 244, 1862.
18. Crook, J.M., Sir Robert Smirke: A pioneer of concrete construction. Trans. Newcomen
Soc., 389, 1965, 5–22.
19. Pasley, C.W., 1838, 16, 265–70.
20. Pasley, C.W., 1826, 16, 25–26, 27.
21. Barley, M.W., The English Cottage and Farmhouse, London, 1961, 83–95, 103, 259,
263.
22. Stevens, H.J., Plaster floors in Derby district. In: discussion following Burnell’s paper
(Ref. 38). Trans. RIBA, 1854, 63.
23. Allen, C. Bruce, Cottage Building and Hints for Improved Dwellings for the Labour-
ing Classes, 6th edn., London, 1867: 40.
24. Foulston, J., The Public Buildings Erected in the West of England, London, 1838.
25. Fox, H.H., Fire-proof roofs, floors and ceilings. Patent No. 10047, 1844.
26. Barrett, J., On the construction of fire-proof buildings. Proc. ICE 12, 1852–53, 251.
27. Dennett, C.C., Floors and ceilings. Patent No. 685, 1857.
28. Hyatt, T., Portland-Cement-Concrete Combined with Iron, as a Building
Material. London, 1877 (reprinted by ACI, Detroit, 1976).
29. Sutcliffe, G.L., Concrete: Its Nature and Uses, London, 1893, Chapter 20.
30. Potter, T., Concrete: Its Use in Building and the Construction of Concrete Walls, Floors
etc., 2nd edn., Winchester, 1891; 3rd edn., London, 1908.
31. Webster, J.J., Fire-proof construction. Proc. ICE, 105, 1891, 249–86.
32. Farrow, Frederick, Fire-resisting Floors Used in London. BFPC Red Book No. 7, 1898.
33. Lawford, G.M., Fireproof floors. Trans. Soc. Engrs, 1889, 43–70.
34. Pasley, C.W., 1838, 243.
35. Pasley, C.W., 1838, 169, 239.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:14:11
03.qxd 15/11/2001 15:45 Page 64

64 Historic concrete

36. Wonnacott, T.H., On the use of Portland cement concrete as a building material. Trans.
RIBA, 1870–71, 175–80.
37 Pasley, C.W., 1838, 233–40.
38. Brunel, M.I., Trans. IBA 1836, 61–64.
39. Verity, T. & F. T., Specification of work required to be done in the erection and com-
pletion of club premises at Nos. 96 and 97 Piccadilly, 1890.
40. Sutcliffe, G.L., London, 1893, Chapter 20.
41. Tall, J., In: discussion to Payne’s paper (Ref. 55). Trans. RIBA, 1875–76, 232.
42. Seddon, Major H.C., Experiments with Concrete Slabs. School of Military
Engineering: Chatham, 1880.
43. Sutcliffe, G.L., 1893, 263.
44. Fowler, C., On terrace roofs. Trans. IBA, 1836, 1, 47–51.
45. Pasley, C.W., Observations etc., 2nd edn, Part 1, London 1847, 174–75.
46. Wilkinson, W.B., Construction of fireproof buildings etc. Patent No. 2293, 1854.
47. Cassie, W.F., Early reinforced concrete in Newcastle-upon-Tyne, Magazine of Concrete
Research, 1955, 25–30.
48. Brown, J.M., W.B. Wilkinson (1819–1902) and his place in the History of Reinforced
Concrete. Trans. Newcomen Soc., 39, 1967, 129–42.
49. Burnell, H.H., Description of the French method of constructing iron floors, and dis-
cussion. Trans. RIBA, 1853–54, 36–74.
50. Lewis, T.H., Fire-proof materials and construction. Trans. RIBA, 1864–65, 109–26.
51. Wonnacott, T.H., On the use of Portland cement concrete as a building material. Trans.
RIBA, 1870–71, 175–80.
52. Blomfield, A.W., Remarks on concrete building. Trans. RIBA, 1870–71, 181–87.
53. Seddon, Captain, Our present knowledge of building materials and how to improve
it. Trans. RIBA, 1871–72, 143–57, 177–84.
54. Smith, T.R., On new materials and recent inventions connected with building. Trans.
RIBA, 1874–75, 199–216, 221–30.
55. Payne, A, Concrete as a building material. Trans. RIBA, 1875–76, 179–92, 225–54.
56. Cates, A., Concrete and fire-resisting constructions. Trans. RIBA, 1877–78, 296–312.
57. Hyatt, T., 1877, 4.
58. Pasley, C.W., 1838, 168.
59. Barrett, J., On the construction of fire-proof buildings. Proc. ICE, 1852–53, 12, 244–72.
60. Hurst, Lawrance, Edwin O. Sachs — Engineer & Fireman. In: David Wilmore (ed.)
Edwin O. Sachs Architect, Stagehand, Engineer & Fireman, Braisty Wood, 1998: 126.
61. Scott, H.Y.D., Floors and roofs. Patent No. 452, 1867.
62. Edwards, F.G., Concrete building etc. Patent No. 2941, 1891.
63. Concrete Quarterly 88/90, April–September, 1971, 50–51.
64. The Builder, 8th December, 1888, 421, 422.
65. 41 and 42 Vic. cap. 32, 22 July 1878. Cited on p. 69 of Harper, R.H., Victorian build-
ing regulations, London, 1985.
66. Bristow, I.C., 1997, 19.
67. Pasley, C.W., 1838, 18, 141, 252.
68. Ranger, William, Artificial stone. Patent No. 6341, 1832.
69. Ranger, William, Artificial stone. Patent No. 6729, 1834.
70. New Civil Engineer, 30th October 1986, 18.
71. New Builder, 28th April, 1995, 32, 34.
72. McFarland, B., Woodhouse, J., Jeanes, C., Royal Albert Dock, ‘the gentle touch’, Con-
struction Repair, 1995, 5, 20–23.
73. Lascelles, W.H., Constructing buildings. Patent No. 2151, 1875.
74. Stanley, C.C., Highlights in the History of Concrete. Cement and Concrete Association,
1979, 26.
75. Lascelles, W.H., In: Discussion on Payne’s paper (Ref. 44). Trans. RIBA, 1875–76, 185.

Further Reading Batty Langley, The London Prices of Bricklayers Materials and Works etc., London, 1749.
Braidwood, J., On fire-proof buildings. Proc. ICE, 1849, 8, 141–61.
Burn, R. Scott, The New Guide to Masonry, Bricklaying and Plastering. John G. Murdoch,
London, c. 1872.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:14:11
03.qxd 15/11/2001 15:45 Page 65

Concrete and the structural use of cements in England before 1890 65

de Courcy, J.W., The emergence of reinforced concrete, 1750–1910. Struct. Engr., 1987,
65A, 315–22, 1988, 66, 128–30.
Francis, A.J., The Cement Industry 1796–1914: A History. Newton Abbot, 1977.
Halstead, P.E., The Early History of Portland Cement. Trans. Newcomen Soc., 1961–62,
34, 37–54.
Hamilton, S.B., A Note on the History of Reinforced Concrete in Buildings, National Build-
ing Studies Special Report No. 24, HMSO: London, 1956.
Hamilton, S.B., A Short History of the Structural Fire Protection of Buildings, National
Building Studies Special Report No. 27, HMSO: London, 1958.
Newlon, H. (ed.), A Selection of Historic American Papers on Concrete 1876– 1926,
American Concrete Institute, Detroit, 1976 (includes papers by Thaddeus Hyatt,
W.E. Ward, A.N. Talbot, A.R. Lord, C.A.P. Turner, E.L. Ransome and D.A. Adams).
Newman, J., Notes on Concrete and Works in Concrete, London, 1887.
Reid, H., A Practical Treatise on Concrete and How to Make it, London, 1869.
Reid, H., The Science and Art of the Manufacture of Portland Cement, London, 1877.
Semple, G., A Treatise on Building in Water, Dublin, 1776.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:14:11
04.qxd 15/11/2001 15:46 Page 67

4 The era of the proprietary systems

Michael Bussell
Synopsis This chapter reviews developments in reinforced concrete construction, mainly
in Britain, from the 1890s until the First World War. It focuses on the proprietary
systems that offered structural design in this new composite material as an adjunct
to the supply of reinforcement, or to the construction of the complete reinforced
concrete structure. Of these the Hennebique system, through its British agent, L.G.
Mouchel, was the most prolific, with a remarkable record of expansion following
its arrival in 1897. Many other systems were patented, although only a few found
wide use, notably those of Truscon (using the Kahn system), and Considère.
The Appendix contains further information to assist in identifying proprietary
reinforcement, floors, and systems of this period in existing construction.

The early years The 1854 patent by William Boutland Wilkinson of Newcastle upon Tyne was the
first to propose the use of iron as reinforcement to concrete, recognizing the rela-
of reinforced tive weakness of concrete in tension.1 Like those of the Frenchmen Lambot and
concrete in Britain François Coignet in 1855, Wilkinson’s patent attracted little interest in the British
building industry. It was not until 1892 that François Hennebique, a French con-
tractor, obtained a British patent for his system, which, with others later, would
find wide use in Britain.2
The principal concern was to provide an economical and fire-resistant form
of construction — something which was widely sought as a solution to the fatal
and expensive fires that frequently consumed mills, warehouses and public build-
ings in particular. Attempts to deal with this problem had begun in the 19th
Century; some of the most notable later 19th-century solutions have been reviewed
in Chapter 3. Such solutions were often based on beams and columns of iron or
steel. Floors used with such frames included brick vaulting, and flat and vaulted
concrete slabs. Although themselves ‘fireproof ’ (or at least incombustible), they
did not always protect the vulnerable bottom flanges of the beams, nor the columns,
from distortion or collapse when exposed to fire.
Hennebique’s patent was based on plain round bars with fish-tailed ends, and
stirrups of flat strips, all at that time being of mild steel. The concrete cover to the
bars and stirrups afforded protection against fire, for a period depending on the
thickness of the cover. The bars provided tensile resistance in beams and slabs,
and supplemented the compressive capacity of the concrete in columns and walls.
The stirrups provided shear resistance, although they were not mechanically
anchored in the compression zone, as is the norm nowadays. Column bars were
linked by strips of wire. A later development of the Hennebique system in 1897
was to provide bent-up bars in beams to provide hogging resistance and supple-
ment shear capacity.
At this time, such construction was generally known as ‘ferro-concrete’ rather
than as reinforced concrete. Significantly, the English version of the Hennebique
house journal, first published in English in 1909 by Mouchel, was titled Ferro-
Concrete (see below).
The question arises how such structures were designed. Hennebique carried
out tests to establish the strength of beams, and had the technical assistance of a

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:14:40
04.qxd 15/11/2001 15:46 Page 68

68 Historic concrete

Belgian engineer.2 His approach was experience-based. Edmond Coignet, son of


François, and Napoléon de Tédesco presented a paper to the French Society of
Civil Engineers in 1894 which laid down the basis for calculations on the modular
ratio method.3 This assumed that plane sections remained plane, that concrete car-
ried no tensile stress, and that capacity was dictated by limiting, or ‘permissible
stresses’ under service load. This approach remained valid and in use for most of
the 20th century. (Design methods of the time are discussed further in Chapter 5.)
Outside Britain, others were developing their ideas on the reinforcement of con-
crete. Indeed, it has to be said that, after Wilkinson, the initiative for the devel-
opment of reinforced concrete came from the Continent (notably France and
Germany) and from the USA. Britain lagged behind, and the early application of
reinforced concrete construction in Britain was essentially due to the ‘import’ of
Continental and American systems.
Joseph Monier obtained a French patent in 1867 for reinforcing concrete plant
tubs with wire and rods. His ideas were taken up by the German G.A. Wayss, and
the ‘Monier system’ — regarded on the Continent as a synonym for reinforced
concrete — was widely used there, but found little application in Britain.4 In
the USA, Thaddeus Hyatt developed reinforcing systems using nuts,4 or cross-
bars placed through holes in square bars,5 to anchor the steel to the concrete.
E.L. Ransome obtained an American patent in 1884 for a square twisted bar which
was stronger (being cold-worked), and had superior bond performance.4
Numerous other patents were obtained, and systems developed, using various
reinforcing section profiles. Expanded metal was introduced as reinforcement to
concrete around 1890, and was widely used.4 The Kahn bar of 1902–1903 is of
unusual profile, being a square section with two projecting strips on diagonally
opposite corners. These are slit to be bent up diagonally in short lengths, form-
ing shear reinforcement. Kahn saw these as forming a ‘trussed’ beam, and indeed
the business which used his system in the USA, and later in Britain, was called the
Trussed Concrete Steel Company, affectionately remembered by many as Truscon.
A detailed recital of the various patent reinforcement profiles and arrangements,
and their related construction systems, would be unduly lengthy. It will, however,
be helpful for today’s engineer to be able to identify the commonest profiles and
systems likely to be found when investigating existing construction of this period.
These are illustrated in the Appendix. Further details of these (and many other)
systems can be found in contemporary texts such as Cassell4 or Marsh.5 Table 4.1
lists the proprietary systems in common use in Great Britain in 1907.
Hennebique’s 1892 patent was followed by the opening of an office in Brussels,
and the construction of two early reinforced concrete structures in France — a refin-
ery in Paris in 1894, and a framed mill in Tourcoing in 1895. In 1897 he obtained
a further British patent, and appointed L.G. Mouchel as his agent in Britain, with
offices in Victoria Street, Westminster (the traditional home of consulting engineers
until recent years). Hennebique’s first building in Britain was Weaver’s Mill of 1897
in Swansea (Figure 4.1, demolished 1984). The working drawings for this were
done in Nantes in France, from where too were supplied the cement, aggregate
and reinforcement; the licensed contractor was also French!2 Apparently the only
indigenous element of the construction was the water used in mixing the concrete.
The expansion of the Hennebique enterprise was remarkable. By 1899, 3061
projects had been undertaken.2 A major reason for this success was Hennebique’s
insistence on the use of suitable materials and experienced labour, to ensure that
the resulting construction was of a good standard. This explains the use of French
materials and a French contractor for Weaver’s Mill, although it was more com-
mon for the company to license contractors whose work would be, in part at least,
undertaken and supervised by men who had gained experience with Hennebique.
In modern jargon, one would say that the process was one of ‘technology trans-
fer’, and certainly such was essential, bearing in mind the phenomenal workload

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:14:40
04.qxd 15/11/2001 15:46 Page 69

The era of the proprietary systems 69

Table 4.1 The common proprietary systems in use in Great Britain in 1907

No. Name of Form of Form of Form of shear Method of fixing Direction of shear
system tension bars compression reinforcement shear reinforcement reinforcement
bars

1 Coignet Round straight Round straight Round rods bent to Looped under tension Vertical
bars bars U-shape bars and twisted above
compression bars
1a Coignet Round straight Round straight Bent up ends of extra Continuous with extra Diagonal
bars and round bars tension bars tension bars
bars bent up near
supports
2 Considère Round straight Round straight (a) Bent up ends of (a) Continuous with (a) Diagonal
bars and round bars extra tension bars and extra tension bars and (b) Vertical
bars bent up near (b) round rods lapped (b) bent round tension
supports around the main tension and compression bars
and compression bars
3 Hennebique Round straight Round straight Steel strip bent to Sprung on to tension Vertical
bars and round bars U-shape and made bars and bent over for
bars bent up near with spring clip anchorage in concrete
supports
4 Improved Round straight Round straight Round rods wound Bent round tension Spiral
Construction bars and round bars around the main tension and compression bars
bars bent up and compression bars
near supports
5 Indented Corrugated square Corrugated Bent-up ends of Continuous with Diagonal
bars, bent up near square straight tension bars tension bars
supports bars
6 Johnson Round straight – [sic] Trough of wire lattice Woven with tension Vertical or diagonal
bars woven with with rectangular or bars strands, according
wire lattice with diamond mesh to mesh used
7 Kahn Square bars Square straight Wings attached to Continuous with Diagonal
generally straight, bars main part of tension tension and
sometimes bent and compression bars compression bars
up towards
supports
8 Ridley- Angle bars Angle or other Trough of corrugated Riveted or bolted Continuous plate
Cammell straight, and bars, straight sheeting
corrugated
sheeting
9 Wells Twin round bars Round straight Steel strip hangers and Bent round tension Vertical
connected by bars bonders [sic] bars
short web
10 Williams Rolled steel Rolled steel (a) Round bars (b) rolled (a) Ends split for (a) Vertical
sections, sections, steel sections (c) spiral anchorage in concrete (b) Diagonal
straight straight coils of steel wire (b) riveted or bolted to (c) Spiral
sometimes used in tension and
addition compression bars

Source: Concrete and Constructional Engineering, 1907–1908, 2, 433.

in those early years and later. By 1909 the Hennebique system had been used for
nearly 20,000 structures, and the company had 62 offices: 43 in Europe (includ-
ing Britain), 12 in the USA, four in Asia and three in Africa. A list of British
Hennebique projects issued in 1911 by what had by then become L.G. Mouchel
& Partners totalled 1073.2 (A second notable reason for Hennebique’s success was
his company’s assiduous attention to promoting its achievements.)
A standard specification of 1917 by L.G. Mouchel & Partners6 is of interest.
It covers ‘ferro-concrete’, and cost 2s. 6d. (121⁄2p). Steel for reinforcement was to

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:14:40
04.qxd 15/11/2001 15:46 Page 70

70 Historic concrete

Figure 4.1 Weaver’s Mill,


Swansea (1897, demolished)
(by courtesy of J.W. Figg).

comply with BS 15 of 1912,7 to be Trisec bars (a patent high tensile steel), or


to be shell discard of prescribed quality (the First World War still had a year to
run). Concrete proportions by volume were approximately 1.1 : 2 : 4 of Portland
cement : sand : coarse aggregate. Provision was made for load testing if required.
(Such testing was common practice at that time.)
The deflection criterion for passing a load test carried out on the completed
structure was onerous by today’s standards. Under 11⁄2 times the imposed load, the
structure was not to deflect more than 1/600 times the span! It should, however,
be borne in mind that such structures were designed using elastic modular ratio
theory, with permissible stresses typically of 16,000 lb/in2 (110 N/mm2) in the
reinforcement and of 600 lb/in2 (4.1 N/mm2) in the compressed concrete. Con-
sequently the section sizes and the reinforcement were more generous than a
present-day design would require. A further factor — certainly for the floor slabs —
is that the framing plans usually provided beams in two directions (on the prece-
dent of iron and steel frames). The typical floor slab was accordingly supported
on all four sides and, when loaded, would tend to behave more as a shallow dome
in compression than as a slab in flexure, with the beams acting both as supports
and perimeter ties. Such behaviour would generate smaller deflections in the slabs.
A typical if early surviving Hennebique building structure is the Co-operative
Wholesale Society (CWS) warehouse on Quayside, Newcastle-upon-Tyne5,8
(Figure 4.2). The building generally dates from 1897 to 1900, although the ninth
floor was added in 1908. The foundation is a raft in view of poor ground condi-
tions. Actually, this ‘raft’ is a series of haunched slabs spanning between substan-
tial ground beams. These slabs are 7 in (178 mm) thick at their centres, with a
square column grid of 14 ft 6 in (4.42 m). Floors were designed for an imposed
load of no less than 6 cwt/ft2 (32.3 kN/m2). Typical suspended floor slabs — also
7 in (178 mm) thick — were satisfactorily test-loaded to 50% above this value.

Procurement of How were such buildings procured? Clearly Hennebique and his competitors were
in effect providing a ‘design and build’ service, although the basic structural
concrete layouts and performance specification would usually be defined by the client and
construction his architects.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:14:40
04.qxd 15/11/2001 15:46 Page 71

The era of the proprietary systems 71

Figure 4.2 Co-operative


Wholesale Society
Warehouse at Quayside,
Newcastle-upon-Tyne before
recent repairs (1897–1900).

There were very few independent consulting engineers practising at the turn
of the century, so far as building design was concerned. Sven Bylander, for
example, who had worked on steel framed structures in the USA, was employed
by Waring White, its builder, for the structural design of the 1904 steel-framed
Ritz Hotel in London. This building, incidentally, had concrete floor slabs
reinforced using the ‘Columbian’ system with Bonna bars of cruciform or ‘cross
of Lorraine’ section (see Appendix).
Architects would generally be competent to design most elements of the build-
ing, including foundations and structural masonry. These elements were largely
sized by empirical rules relating foundation width to wall thickness, and wall thick-
ness to height of building. Similarly, tables and standard textbooks were available
to assist the architect to design and size timber floors and roofs.
When it came to steel or concrete structures, however, it was normal for their
design — that is the calculations and drawing-up of details — to be carried out
by ‘specialists’. For steelwork this would be the steel fabricators, and for reinforced
concrete it would be Hennebique, Truscon, Considère and others.
Nor was the appointment of a specialist just a matter of selecting one such
company to implement the architect’s scheme.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:14:40
04.qxd 15/11/2001 15:46 Page 72

72 Historic concrete

The example of the recently refurbished YMCA building in Manchester is illu-


minating.9 Its structure was originally conceived in 1908 by the architects Wood-
house, Corbett & Dean as a steel skeleton with brick walls. Two particular features
of the building led to a decision to consider reinforced concrete. One was the pro-
vision of a swimming pool on the top floor, the box structure for this logically being
best done using reinforced concrete. The other was the need to have a column-
free hall at lower level. The original plan was to span this by storey-height steel
trusses built into masonry walls on the floor above, but clearly it would be a sens-
ible alternative to design the walls in concrete as deep beams, with considerable
savings in steel and altogether simpler construction.
The drawings were sent out to five specialists, of which the Trussed Concrete
Steel Company was successful in its bid. Each was required to prepare a scheme
in concrete and to submit it together with a tender prepared by a contractor, to
be nominated by the specialist. So the ‘main’ contractor was, effectively, the choice
of the concrete specialist. This apparently back-to-front approach makes sense
when the importance of good materials and workmanship, already stressed, is taken
into account. In their own interests, the concrete specialists might be expected
to propose only contractors licensed by or approved by themselves to use their
system.
Four of the five specialists submitted schemes with the tenders. It then appears
that the architects spot-checked selected typical designed members from each
scheme and calculated their strengths using the 1907 RIBA Joint Committee
Report.10 These calculations were compared with the required strengths, to give
an indication of the soundness of the tenderers’ designs.
Having accepted the Trussed Concrete tender using Kahn bars, the client
required the final design by the specialists to follow the RIBA Report methods,
with C.F. Marsh, author of the first British book on reinforced concrete,5 acting
as checking engineer. The specialists were required to pay his fee for this service!
There were at this time no regulations for reinforced concrete construction in
Manchester, and Mr Marsh’s ‘seal of approval’ was noted as being most helpful in
obtaining building control consent.
The picture one has, therefore, is of the design and construction of reinforced
concrete at this time clearly being a specialist activity. Established concrete spe-
cialists such as Hennebique were anxious, of course, to increase turnover and
profits, but were also no doubt concerned that use of unsuitable materials and inex-
perienced labour, without established standards or codes of practice, could lead
to failures and consequent loss of confidence in reinforced concrete as a structural
material. They were willing to engage in competitive bidding against each other
for work — indeed, they had no choice — and that necessarily meant preparing
designs and pricing schemes competitively, with no guarantee of payment. Against
the previously quoted figure of 3061 projects realized by Hennebique in the
period 1892–99 must be set 8078 not realized.2
The ethical position for a professional engineer employed by a specialist con-
crete firm was in those days ambiguous; at least one ICE member was expelled
for over-zealous efforts to ‘procure’ work for his employer. The fiercely compet-
itive commercial environment may explain this eagerness, although of course it
could not (officially) be condoned by Edwardian professionals.

The systems Hennebique was, with Mouchel, for several years the only major concrete specialist
in Britain. The company began working in Britain in 1897. The Trussed Concrete
develop Steel Company took out a British patent in 1903 to cover use of its American Kahn
bar.11 In 1904 Edmond Coignet built a group of tobacco warehouses in Bristol
(recently demolished) while the British Reinforced Concrete Co. (BRC) was estab-
lished in 1905. Armand Considère took out a British patent in 1902 for the use

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:14:40
04.qxd 15/11/2001 15:46 Page 73

The era of the proprietary systems 73

of helical binding in columns, based on tests he had conducted. These showed —


as is now well appreciated — that adequate restraint to the compression bars is
essential, both to restrain them from premature buckling and to enhance the
strength of the concrete core. But this form of reinforcing was not used until 1907.
Considère later developed the U-hook to anchor main bars. Other forms of
reinforcement included expanded metal, as illustrated in the Appendix.
Anchorage and bond were recognized as important design matters, and in this
respect the relatively inferior performance of the plain round or square bar was
one factor in the proliferation of patents for the use of deformed or profiled bars.
Of these, the Ransome and Kahn bars have already been noted. Others included
Johnson’s indented bar, patented unsurprisingly by the Patent Indented Steel Bar
Company, and much used by the British Concrete Steel Company. An advertise-
ment in a contemporary textbook12 by the former company notes that this bar
profile has a continuous mechanical bond with the concrete, without slip. Signifi-
cantly, it also points out that the bar is offered without fee or royalty, and may
be used for any concrete system. In other words, the company was offering a
product — the bar — to anyone who would buy it for use in their own reinforced
concrete construction. This is in strong contrast to the ‘concrete specialists’, mainly
from the Continent, who were offering a complete system.
The clients for early reinforced concrete structures included, perhaps para-
doxically, some of the more ‘conservative’ bodies such as the Government’s Office
of Works, dock and harbour companies, railways, water boards, and agricultural
and co-operative societies. Their interest in the new construction material was,
surely, commercially motivated. Concrete was clearly economically competitive with
steel and masonry from the outset, for if this had not been the case then its use
would never have been considered. At the same time, it was seen to offer other
advantages. The reinforcement appeared to be inherently protected against cor-
rosion by its embedment in the concrete. Only later did it become embarrassingly
clear that the actual thickness of concrete cover, the concrete mix design, its
compaction, and (particularly) water content were all vitally important in
determining durability. The concrete cover was also acknowledged to protect the
reinforcement against fire. Concrete was pourable, so that structures could be built
with non-orthogonal forms at only slightly greater cost. The Manchester YMCA
swimming pool has already been cited as a case where the mouldable nature of
concrete, its structural strength, and its potentially water-resistant nature combined
to offer the ‘ideal’ material. Water tanks, silos, bunkers, and other essentially
functional structures were obviously similarly suited to concrete construction.
The Government’s initial attitude towards reinforced concrete was ambivalent.
On the one hand, the Local Government Board was unwilling to provide loan
finance for local authorities to build in reinforced concrete on the same terms as
were applied to traditional construction and even to the relatively recent newcomer,
steel framing. But many public architects (then as now) were keenly interested in
new structural developments. Sir Henry Tanner became Chief Architect in the
Office of Works in 1898 and encouraged the use of reinforced concrete in its pro-
jects.13 Crown buildings were exempt from building regulations, and so there was
no procedural objection to using reinforced concrete for their construction at a
time when building bye-laws made no provision for this new material. Conse-
quently, the Public Office and the larger Sorting Office of the General Post Office
in King Edward Street in the City of London12 were built on the Hennebique sys-
tem in 1907–10. The demolition of the Sorting Office in 1998 regrettably deprived
London of one of its earliest and boldest reinforced concrete structures. However,
the demolition exposed briefly the ambitious scale of the construction (Figure 4.3).
It also showed the simple butt-connection of column bars in the monolithic frame,
a detail that might surprise today’s engineer accustomed to providing lapped
connections (Figure 4.4).

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:14:40
04.qxd 15/11/2001 15:46 Page 74

74 Historic concrete

Figure 4.3 A cross section


through the Sorting Office
of the General Post Office,
King Edward Street, City of
London (1907–1910), exposed
during demolition in 1998
(by courtesy of Lawrance
Hurst).

It is no coincidence that Sir Henry took the chair of the RIBA Concrete
Committee when it was established in 1905. He was also an active member of the
Concrete Institute. (His role in promoting the wider use of concrete is discussed
further in Chapters 5 and 14.)
The ready embrace of reinforced concrete by public utilities (including railway
companies and port authorities), and by commercial bodies, was undoubtedly
spurred by practical considerations of economy, including what was perceived as
minimal maintenance cost. Today we may question this strictly utilitarian approach
on aesthetic grounds. Unfortunately, too, durability was inadequately understood,
as will be considered in more detail in Chapter 5. Weaver’s Mill in Swansea
(Figure 4.1) was not a visual delight as it stood abandoned in the early 1980s, and
it was demolished despite being the oldest surviving reinforced concrete structure
in Britain. It is not certain which structure now holds that title, although the
1897–1900 CWS Quayside warehouse in Newcastle (Figure 4.2) is a contender.
(It is shown here before recent repairs.)
Many concrete structures of this period were bold in concept. One of the most
spectacular was the 1909 Royal Liver Building at Pier Head, Liverpool, architect
W. Aubrey Thomas4 and once more engineered by Mouchel using the Hennebique
system (Figure 4.5). One hundred and sixty-seven feet (51 m) high at roof level,

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:14:40
04.qxd 15/11/2001 15:46 Page 75

The era of the proprietary systems 75

Figure 4.4 Column in the


Sorting Office of the General
Post Office during demolition,
showing butt-jointed main
bars with small tubes used
as location aids (1907–1910)
(by courtesy of Lawrance
Hurst).

and crowned by the Liver bird with an overall height of 310 ft (94 m), it offered
arriving transatlantic voyagers the sight of what was then the nearest resemblance
in Europe to a New York ‘skyscraper’.
The tallest concrete building in the world before this structure was erected had
been the 18-storey Ingalls building in Cincinnati, Ohio, of 1902, at 210 ft (64 m).
This was reinforced on the Ransome system of twisted bars, mixed with plain round
bars. Column main bars were spliced using early ‘couplers’ — sleeves of wrought
iron tubing filled with grout (similar to the details seen in the GPO Sorting Office,
shown in Figure 4.4). Writing of the Ingalls building, Twelvetrees notes wryly that
‘the architectural design is not remarkable for novelty’,12 and indeed it can be said
that many early concrete buildings, as opposed to structures, did not advertise their
concrete construction. From the outside, and often even from the inside, the struc-
ture could be of steel or of concrete — it was hard to tell. The expressionist use
of concrete in habitable buildings had to wait for the birth of the modern move-
ment (see Chapter 2). On the other hand, functional structures were usually clear
in displaying their concrete form and finish.
This review suggests that, until the First World War, the concrete specialists ‘had
it all their own way’ so far as design was concerned. While this is true in terms of
the preparation of designs, drawings and schedules, it was often required of the

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:14:40
04.qxd 15/11/2001 15:46 Page 76

76 Historic concrete

Figure 4.5 Royal Liver


Building, Pier Head, Liverpool
(1909) (by courtesy of British
Cement Association).

specialists that they also satisfy an independent consultant. The example of Marsh
and the Manchester YMCA building has already been described above. As for con-
struction, the specialists either did the work themselves, or they licensed contractors
who were guided by the specialists in sound construction practice, usually in return
for paying a royalty for using the system.

Construction Actual construction practice was inevitably more basic than today, although the
principles of sound practice were little different. Materials were to be clean and
practice properly stored; formwork was to be clean, rigid and secure. It was in the mix-
ing, transporting, placing and treatment of concrete that the main differences were
to be found.
All concrete was site-batched — there were no ready-mix trucks, no skips man-
oeuvred by tower cranes, no pumps. Hand-mixing was common, although
mixing machines were already being used. Concrete was mixed by volume pro-
portions (a practice that is still to be found today on smaller sites), or in ‘recipe’
form. A good example is the 1917 Mouchel specification.6 This typically called for
6 cwt (305 kg) of cement to be batched with 131⁄2 ft3 (0.38 m3) of sand and 27 ft3
(0.76 m3) of coarse aggregate to give a 1.1 : 2 : 4 mix, with a probable cube strength
of around 15–20 N/mm2.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:14:40
04.qxd 15/11/2001 15:46 Page 77

The era of the proprietary systems 77

Figure 4.6 Typical concrete


punning tools from the early
20th century.

Transporting was likely to be by tipper trucks running on a narrow-gauge rail-


way, by chute, or by wheelbarrow once the fresh concrete had been raised to the
required level by a hoist or a derrick. Placing would be by spade or shovel, work-
ing the mix as far as possible into its final location. It would then be compacted,
up to a point, by ramming, tamping, or ‘punning’ by hand. The tools for this work
came in a variety of shapes (Figure 4.6), including a rod with an enlarged box-
end and another shaped like a hockey stick.4 Not until 1917 did Eugène Freyssinet
establish the essential importance of mechanically compacting the concrete to expel
air and ensure thorough filling of the profiles, completely enclosing the steel
reinforcement with concrete. Even then, it was 1924 before he himself used
vibrating tools in practice. Before then, the importance of filling the shutters and
fully surrounding the reinforcement with concrete was understood; but, without
mechanical compaction, and with (often) congested reinforcement, the all-too-
obvious answer was to make the concrete wetter, so that it flowed more readily
and was ‘self-compacting’. Some specialists, including Ransome, argued in the
1900s that a wet mix would indeed be superior to a dry mix.11 It is perhaps just
a coincidence that Duff Abrams completed his research on concrete mixes in 1918,
only a year after Freyssinet’s work on the need for mechanical compaction.
Abrams’s law declares that the water–cement ratio is a fundamental determinant
of concrete strength. In simple terms, a ‘wetter’ concrete will be weaker when
hardened than a ‘drier’ concrete.

Variations of Numerous construction techniques and structural forms were introduced during
the period under review which, at first glance, might be thought to be later in
technique and origin. Some more notable examples are briefly described below.
form
Precasting

Examples of early precast cladding have been given in Chapter 3. Precasting of


structural elements was the subject of Hennebique patents as early as 1897, both
as self-contained units and as permanent shuttering with projecting reinforcement
to provide composite action with in-situ concrete. The British Precast Concrete
Federation was formed as early as 1918.14

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:14:40
04.qxd 15/11/2001 15:46 Page 78

78 Historic concrete

A simple precast floor system was revealed in the 1904–1905 Joshua Hoyle
building in Manchester prior to refurbishment. This building has a steel frame
and terracotta cladding, and — clearly an innovation for this time — an unattended
automatic lift.15 However, it is the floor units that are of most interest. They are
precast unreinforced half arches that are laid in pairs to sit on the lower flange of
the I-beams, butting at the crown. The units thus form simple three-pinned mass
concrete arches. The coarse aggregate is clinker (burnt coal), and timber shavings
as are nowadays found in woodwool slabs. Re-use proposals had to take account
of the potential combustibility of such concrete.

Filler joist floors and clinker concrete

The introduction of ‘filler joist’ floors in the second half of the 19th century was
described in Chapter 3. They continued to be used in the early part of the 20th
century. Unreinforced concrete slabs, spanning effectively as flat arches, were sup-
ported by I-beams or joists at typically 0.6–1.0 m spacing (Figure 4.7). Originally,
the joists were of wrought iron, while the concrete would contain either lime or
Portland cement as binder, with coarse aggregate of broken brick, slag, clinker or
‘breeze’ from coal fires, or stone. Incomplete combustion of the coal could leave
a combustible fraction in the floor. Steel, as it became more widely available from
the 1890s, replaced wrought iron in the joists, while the concrete was increasingly
made with Portland cement, sand and stone aggregate, typically as a 1 : 2 : 4 mix.
This filler joist floor construction was widely used in offices and blocks of
flats, notably the ‘mansion blocks’ of Edwardian times and later, where its high
mass gave good resistance to air-borne sound. Timber flooring laid on battens
(particularly when carpeted) also gave good structure-borne sound resistance,
resulting in excellent acoustic insulation between dwellings.
Tests to assess the enhancement of strength by what is nowadays termed com-
posite action between the filler joists and the concrete were carried out at the
National Physical Laboratory as late as 1922–23.16 The results were incorporated
into BS 44917 in 1932, and indeed remain in the 1969 metric edition still, with
later amendments, in use today. However, the wide adoption of orthodox reinforced
concrete floors, both in situ and precast, resulted in the demise of filler joist con-
struction, with its requirement of formwork for the invariably in-situ infill concrete.
A problem often found with clinker concrete in such construction arises when
the floor or roof becomes wet, as in neglected bathrooms or kitchens, or on flat
roofs with degraded waterproofing. The clinker contains compounds of sulphur,
and also of nitrogen and chlorine, which in wet conditions can severely aggravate
corrosion of the iron or steel joists, leading also to spalling of adjacent concrete.
Repair can be expensive and disruptive.

Figure 4.7 Typical section


through filler joist floor
(by courtesy of Construction
Industry Research and
Information Association,
Report 111, Structural
renovation of traditional
buildings).

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:14:40
04.qxd 15/11/2001 15:46 Page 79

The era of the proprietary systems 79

Concrete blocks

Concrete was quickly recognized as a potentially cheaper but adequate substitute


for fired-clay materials including bricks and tiles. A patent for hollow concrete
blocks had been taken out as early as 1850,18 and block-making machines were
in use by c. 1860.14 Steam curing was in use by the early 1900s18 to accelerate
curing and increase productive use of the block moulds.

Asbestos–cement and woodwool

The first asbestos–cement sheets are believed to have been made in Austria in 1900.
Within a decade the material was being widely used in its basic flat sheet form as
a durable, non-corroding building board and roof sheeting. Corrugated sheeting
followed soon after in the early 1910s, achieving longer spans from its deeper pro-
file.14 This is a relatively early example of fibre-reinforced cement, a predecessor
of grc or glass-reinforced cement, used for architectural cladding in recent decades.
Another cement-based composite with a lengthy history is woodwool which, like
clinker concrete, makes use of what would otherwise be a waste product. Wood
shavings from the planing of timber and cement paste are pressed together to form
slabs, which are lightweight and provide good thermal insulation.

Unreinforced concrete

The emphasis in this chapter is mainly on reinforced concrete, but the use of ‘plain’
or unreinforced concrete was — and remains — still widespread. This followed
the precedent of using lime-based concrete (described in Chapter 3) for founda-
tions pads, strips, and rafts, and other structural elements subject to essentially
compressive loads. Plain concrete was an obvious choice where bulk or mass was
needed, as in gravity dams for reservoirs, retaining walls to uphold the sides of
railway cuttings, and marine works. It was also used in the walls of housing and
other buildings, where its presence may be deduced from the thicker wall section,
the presence of visible horizontal daywork joints, and often the erosion of the
cement paste to leave a stony surface.
The term ‘mass concrete’ should perhaps have been confined to the use of
the material in bulk, but in practice it has always been synonymous with plain or
unreinforced concrete.

Shotcrete

Shotcrete, also known as gunite and nowadays as sprayed concrete, was developed
almost a century ago in 1907 by one Carl Ethan Akeley, an American naturalist
and taxidermist.19 Seeking a means of creating animal models and mounts, he
devised the idea of ‘shooting’ a pressurized blend of cement and sand onto a metal
armature, adding water as the mixture left the ‘cement gun’. Built up in thin dense
layers, this produced realistic free-form profiles with a hard durable surface.
The Cement-Gun Company was formed in Allentown, Pennsylvania, in 1910.
It developed Akeley’s invention and applied it on a larger scale to building and
civil engineering uses. The company registered the trade name of ‘Gunite’ for the
product, which was quickly adopted for hydraulic work and particularly for tun-
nel linings (see Chapter 15). It also found wide use as sprayed fire protection to
steel frames, and for repair and strengthening works, when the ability to add a
reinforcement cage and then form the concrete profile without formwork proved
very effective.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:14:40
04.qxd 15/11/2001 15:46 Page 80

80 Historic concrete

The process was introduced into the UK in the 1920s. The original mixing
method came to be known as ‘dry-mix’. Later, the ‘wet-mix’ method was intro-
duced, in which the water was mixed with the cement and sand before it left the
gun. This in effect uses the same principle as present-day concrete pumping,
although the discharge speed was much higher. In both methods ‘rebound’ of some
of the concrete was inevitable, although an experienced operator could limit this,
and would also control the amount of water added.

Publications of the For those faced with appraising or refurbishing such structures today, it may be
helpful to review the technical information and design guidance available to clients,
time architects and structural practitioners during the years 1897–1915. The libraries
of the Institutions of Civil and Structural Engineers, and also of the Royal Insti-
tute of British Architects, have good collections of useful material and can between
them provide all the references cited here.
This pioneering period began with construction of the first British Hennebique
Mouchel designs, and closed with publication of the London County Council
Reinforced Concrete Regulations20 which offered the first ‘code of practice’ for
reinforced concrete design. Regulatory practice is discussed further in the next
chapter; here, attention is briefly drawn to contemporary texts and specialist
literature.
As already noted, Reinforced Concrete by C.F. Marsh appeared in 1904, being the
first British textbook on reinforced concrete design and construction.5 It is a sub-
stantial text, well illustrated, and can be fairly called a ‘state-of-the-art’ review. Later
editions followed in collaboration with W. Dunn. These gave increasing space to
calculation methods as the use of reinforced concrete widened.18
A more enduring record of concrete practice began for the general reader with
the publication in 1906 of the first issue of Concrete and Constructional Engineering,
which continued until 1966. This journal is a rich source of information on both
individual buildings and practice of the time.
Essentially a house journal, but none the less useful for that, was Hennebique’s
Le Béton Armé, produced from 1898. An English sister journal, Ferro-Concrete, began
publication in 1909. Today this is a valuable if scarce source of detailed informa-
tion for British Hennebique buildings such as those mentioned here. The early
editor was W. Noble Twelvetrees, also the author of early guides to theory and
practice in reinforced concrete.12,21
The Concrete Institute was set up in 1908 for architects, engineers and con-
tractors, publishing its first Transactions in 1909. Its development and metamor-
phosis into the Institution of Structural Engineers is traced in Chapter 14.
Other concrete specialists published literature — both technically supportive and
openly promotional — during this period, while various ‘independent’ textbooks
started to appear. Their growth into a wide range of design guides is outlined in
Chapter 5.
The Appendix gives more comprehensive guidance on sources of contempo-
rary information.

The era comes to The outbreak of the First World War in 1914 brought many things to an end. From
the viewpoint of structural history and that of today’s structural engineer, it can
an end: The First be said to mark the end of the ‘golden age’ of the proprietary systems. The pub-
World War and lication of the London County Council Reinforced Concrete Regulations in 1915
later put an acceptable method of designing reinforced concrete structures into the
public domain.20 Using plain mild steel bars, the engineer or numerate architect or
builder could now do the calculations and design concrete structures. That is not
to say that the specialists disappeared; but an element of ‘free market’ theory had

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:14:40
04.qxd 15/11/2001 15:46 Page 81

The era of the proprietary systems 81

arrived. Henceforward, it would not need a specialist to design concrete structures;


and the designs could be assessed against published codes, and approved under
regulatory bye-laws. Nevertheless, companies such as Hennebique continued
to design and build structures. L.G. Mouchel, for example, formed a consulting
engineering practice that continues to this day. Companies making reinforcement
continued to offer design services — indeed, a few still do.
Another aspect of the First World War was, of course, the application of
reinforced concrete to military structures, most commonly as protection against
shelling. The defensive ‘pill-box’ has become for some an archaeological novelty,
but it should be recalled that its design, however empirical, involved the first seri-
ous attempts to assess the effects of blast loading on reinforced concrete structures.22

Additional general Additional general sources for the period covered in this chapter include studies
by de Courcy,23 Stanley,24 and Mainstone.25 Some sources of guidance for those
sources of called upon to assess or alter such structures is cited at the end of Chapter 5, which
information on traces the continuing development of concrete in the 20th century, and is
this period supplemented by more comprehensive guidance in the Appendix.

Acknowledgements The author wishes to thank his fellow co-authors for their encouragement and
for helpful criticism of the text. The source of Table 4.1 was drawn to the author’s
attention by Mr B.N. Sharp. Sir Alan Muir Wood kindly provided advice on the
development of shotcrete.

References 1. Brown, J.M., W.B. Wilkinson (1819–1902) and his place in the history of reinforced
concrete. Trans. Newcomen Soc., 1966–67, 39, 29–142.
2. Cusack, P., François Hennebique: the specialist organisation and the success of ferro-
concrete: 1892–1909. Trans. Newcomen Soc., 1984–85, 56, 71– 86.
3. Coignet, E., Tédesco, N. de., Du calcul des ouvrages en ciment avec ossature metallique.
La Société des Ingénieurs Civils de France: Paris, 1894.
4. Jones, B.E. (ed.), Cassell’s Reinforced Concrete, 2nd edn. Waverley Book Company:
London, 1920.
5. Marsh, C.F., Reinforced Concrete. Constable: London, 1904.
6. L.G. Mouchel & Partners, Standard Specification for Ferro-Concrete. L.G. Mouchel
& Partners: London, 1917.
7. Engineering Standards Committee. Standard Specification for Structural Steel for
Bridges and General Building Construction. Engineering Standards Committee:
London, 1912, BS 15.
8. Anon., Ferro-concrete warehouse at Newcastle-on-Tyne. Engineering, 1903, April 17,
514–15.
9. Lakeman, A., The YMCA building, Manchester. Concrete & Constructional Engi-
neering, 1911, 6, 368–77 (and discussion, 501–15).
10. Joint Committee on Reinforced Concrete. Report of the Joint Committee on Reinforced
Concrete. J. Roy. Inst. Br. Archit. (3rd ser.), 1907, 14(15), 513– 41.
11. Hamilton, S.B., A Note on the History of Reinforced Concrete in Buildings. National
Building Studies Special Report No. 24. HMSO: London, 1956.
12. Twelvetrees, W.N., Concrete-Steel Buildings. Whittaker & Co.: London, 1907.
13. Gray, A.S., Edwardian Architecture: A Biographical Dictionary, 2nd edn. Wordsworth
Editions: Ware, 1988.
14. Hudson, K., Building Materials. Longman: London, 1972.
15. Anon., A Piccadilly improvement. Manchester City News, 1905, January 21.
16. National Physical Laboratory Reports, 1922, p. 8; 1923, p. 169. Etchells, E.F. (ed.),
Modern Steelwork. Nash & Alexander: London, 1927.
17. British Standards Institution. Specification for the Use of Structural Steel in Building.
BSI: London, 1932, BS 449.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:14:40
04.qxd 15/11/2001 15:46 Page 82

82 Historic concrete

18. Marsh, C.F., Dunn, W., Manual of Reinforced Concrete and Concrete Block Con-
struction. Constable: London, 1908 (and various other editions).
19. King, E.H., Shotcrete. In: Bickel, J.O. et al. (eds), Tunnel Engineering Handbook,
2nd edn. Chapman & Hall: New York and London, 1996, 220–30.
20. London County Council. Reinforced Concrete Regulations. LCC: London, 1915.
21. Twelvetrees, W.N., Concrete-Steel. Whittaker & Co.: London, 1906.
22. Mallory, K., Ottar, A., Architecture of Aggression: A History of Military Architecture
in North West Europe 1900–1945. Architectural Press: London, 1973.
23. De Courcy, J.W., The emergence of reinforced concrete 1750–1910. Struct. Engr, 1987,
65A(9), 315–22 (and discussion of paper in Struct. Engr, 1988, 66(8), 128–30).
24. Stanley, C.C., Highlights in the History of Concrete. Cement & Concrete Association:
Slough, 1979.
25. Mainstone, R.J., Developments in Structural Form, 2nd edn. Architectural Press:
Oxford, 1998.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:14:40
05.qxd 15/11/2001 15:47 Page 83

5 The development of reinforced


concrete design and practice

Michael Bussell
Synopsis The previous chapter described the introduction of reinforced concrete into the
UK via proprietary systems, which flourished from the end of the 19th century
until the First World War. This was in the absence of an agreed — or rather a
codified — design method, which did not appear until 1915. This chapter reviews
developments in the understanding of reinforced concrete behaviour, and charts
the gradual standardization of structural design, materials, codes, standards and
textbooks up to 1948 when CP 114 first appeared. It continues this review briefly
from 1948 to the present day. Some significant structures built between the First
World War and 1948 are noted. Landmarks in architectural concrete use and other
notable developments in shell roofs, prestressed concrete, bridges, and maritime
structures are covered in more detail in other chapters.
Appendix B contains information to assist in identifying proprietary concrete
floors of this period in existing construction.

Early Early studies of the behaviour of reinforced concrete were led by American, French
and German workers. The first book on reinforced concrete (as opposed to articles)
understanding of was published in 1877 by Thaddeus Hyatt, an American living in London.1 He
reinforced concrete used data from tests carried out at David Kirkaldy’s testing laboratory in South-
behaviour wark. His work recognized that composite behaviour between iron or steel and
concrete is dependent on bond between the two materials, and is aided by the near
equivalence of their coefficients of thermal expansion. Hyatt’s work acknowledged
that, at least under working loads, strain compatibility renders stresses in materials
proportional to their relative elastic moduli.2
G.A. Wayss published in 1887 a book on the Monier system,3 which his com-
pany was using in Germany and elsewhere. Working with K. Koenen, he concluded
that the steel should be designed to take all the tensile stresses, and also that bond
between steel and concrete was essential to transfer the internal forces to maintain
equilibrium. A similar study was published in 1894 in France by Edmond Coignet
and Napoléon de Tédesco,4 and this — together with a paper by P. Christophe of
1899, published as a book in 19025 — laid the foundations for elastic modular ratio
theory. This was practised until well after the Second World War, and indeed is
still used for serviceability calculations in today’s limit-state codes.

Early design In Britain the first textbook, by C.F. Marsh,6 appeared in 1904; but it was 1906
before the Royal Institute of British Architects appointed a Reinforced Concrete
guidance Committee to review the use of structural reinforced concrete. Its first report was
published in 1907.7
Events in Britain now gained momentum. BS 12, covering Portland cement,
appeared in 1904, reflecting concern about the quality of cement. Similar concern
led to the establishment of a Special Commission on Concrete Aggregates by
the British Fire Prevention Committee in 1906. The previous chapter drew atten-
tion to the use of clinker aggregates. These often contained unburnt coal; in the

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:16:44
05.qxd 15/11/2001 15:47 Page 84

84 Historic concrete

event of a fire the nominally ‘fireproof ’ floor could then actually support com-
bustion! The Committee’s founder and chairman was E.O. Sachs, who was also
editor of Concrete and Constructional Engineering, which began publication in the
same year.
The year 1908 saw the formation of the Concrete Institute, an open forum for
all interested in the use of reinforced concrete, not just ‘the specialists’. This, in
due course, became the Institution of Structural Engineers. Chapter 14 gives a
fuller account of its birth and development.
The year 1909 was notable for the passing of the London County Council (Gen-
eral Powers) Act.8 This is well known for its schedule covering — for the first time
in British building regulations — the design of steel, wrought iron and cast iron
structures, although by the time of its issue steel was almost exclusively the only
ferrous structural metal in use. The Act also made provision for the introduction
of regulations controlling the use of reinforced concrete in inner London, although
these would not appear until six years later. (It also, for the first time, specified
floor loadings to be taken into account in design.)
The Institution of Civil Engineers set up its own Committee on Reinforced
Concrete, which reported in 1910.9 The committee reviewed the concrete speci-
fications of the specialist firms (Hennebique, etc.), overseas rules on concrete, and
available test data. It also heard statements from practising consulting engineers,
who were generally reluctant to take responsibility for designing larger concrete
projects. Of particular relevance for today’s appraising engineer is the evidence
for wide variation in cover requirements (or none at all) and a preference for using
wetter mixes to achieve ‘good’ compaction, when mechanical compacting methods
were neither yet available nor recognized to be beneficial. Low cover and a weaker
‘wet’ concrete have obvious implications for durability. This is now understood,
but was not recognized at the time.
In 1911 the RIBA Committee was restructured as a Joint Committee, with
added representation from the Concrete Institute and the London County
Council (LCC). It produced a revised report in the same year.10 This again did
not make specific recommendations for cover or care in use of wetter concrete
mixes, so that durability was not well addressed. Fire resistance needs were
taken into account, with cover of 1⁄2 in (13 mm) being specified for slabs and 1 in
(25 mm) for beams. A 1 : 2 : 4 concrete mix was assumed, with a cube strength of
1800 lb/in2 (12.4 N/mm2) at 28 days. The permissible concrete compressive
stress was to be one-third of this value. The standard notation for symbols, intro-
duced in 1909 by the Concrete Institute, was adopted. Shear reinforcement
was to be provided when concrete shear stress exceeded 60 lb/in2 (0.41 N/mm2) —
one-tenth of the permissible compressive stress — and was to comprise either
bent-up bars or stirrups spaced no further apart than the beam depth. Column
axial compressive capacity was recognized to be enhanced when helical bind-
ing was present, acknowledging the containing effect of this steel. Design for
bending was based on the by then generally accepted elastic modular ratio
theory, using a ratio of 15.2 Concern about reinforcement anchorage to the con-
crete was reflected in guidance that bar ends should be split, bent, or otherwise
mechanically secured.

The 1915 These recommendations were largely embodied in the LCC’s Reinforced Concrete
Regulations of 1915.11 The symbol Q (with units of stress) was defined as the ratio
Reinforced of applied moment to (member width times the square of effective depth); it was
Concrete a useful and long-lived aid to engineers making calculations (with a slide rule, of
Regulations course), as it was a measure of how hard the section was working. Tensile and shear
steel was to be effectively anchored, by hooking or otherwise, at its ends. Hook
internal diameter was to be four bar diameters.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:16:44
05.qxd 15/11/2001 15:47 Page 85

The development of reinforced concrete design and practice 85

Cover was again dictated by fire considerations. Columns (known then as ‘pil-
lars’ if vertical, otherwise as ‘struts’) were to have cover of 11⁄2 in (38 mm), or the
bar diameter if this were greater.

From 1918 to 1934 The LCC regulations offered the first codified British design and construction guid-
ance for reinforced concrete, but they were little used until the end of the First
World War. The next decade, the 1920s, was a period of often (but not always)
poor quality construction, carried out by building contractors who were frequently
unaware of the need for adequate cover to steel and the need for proper treatment
of concrete. This contrasted sharply with the pre-war period, when the specialist
concrete firms might admittedly charge higher prices, but would take care to build
a sound structure. They had both the understanding of what was needed, and
the motive of maintaining sound reputations.
A dramatic example of a good 1920s reinforced concrete structure is the 1926
New Royal Horticultural Hall in Westminster, London, by architects Easton and
Robertson (Figure 5.1). Design of the structure was by British Reinforced Con-
crete Ltd., one of the longer-lasting specialist firms providing a design service. Such
a structure shows a clear use of concrete arches that could never be mistaken in
form or appearance for either steel or masonry.
Another notable structure of this period, facing imminent demolition at the
time of writing, is Wembley Stadium (1921–24, Figure 5.2). It was designed by
the architect Maxwell Ayrton and the engineer Owen Williams for the British
Empire Exhibition, to accommodate 125,000 people. The contractor was Sir Robert
McAlpine. This was one of the several large structures on the exhibition site, includ-
ing the Empire Pool. The stadium was presented more in the style of a masonry
structure both in form and in the false jointing applied to concrete surfaces to give
the appearance of ashlar masonry. This was on the ‘public’ faces of the building.
The engineer Oscar Faber was disappointed with the more utilitarian concrete
work elsewhere.12

Figure 5.1 The New Royal


Horticultural Hall (1926)
(by courtesy of the British
Architectural Library).

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:16:44
05.qxd 15/11/2001 15:47 Page 86

86 Historic concrete

Figure 5.2 Wembley


Stadium (1921–24)

Williams himself went on to engineer some rugged Highland bridges (see


Chapter 11), before becoming architect–engineer for the 1930 Dorchester Hotel,
London (in collaboration with architect W. Curtis Green and consulting engineers
Considère & Partners) and many other structures. Of these perhaps the most notable
are the 1931 Daily Express building in London, its 1939 successor in Manchester,
and the 1932 Boots ‘wets’ factory at Beeston, Nottingham (Figure 5.3).12,13
The two last-mentioned buildings are particularly significant in their use of flat
slabs. These had been developed independently in the USA and in Switzerland,
as an alternative to the traditional drop beam solution. Beams were essential in a
frame of iron or steel because these materials are used as one-dimensional ‘sticks’,
whereas concrete can be readily formed to any required shape. In addition, the
omission of downstanding beams simplifies the formwork and eases construction.
Nevertheless, early concrete floors and roofs were usually structured with one-
way or, more commonly, two-way grids of beams supporting slabs. Indeed, in some
buildings the concrete beams are haunched near the column supports to look more
as if they are steel beams cased in concrete and supported on stiffened gussets
(Figure 5.4)! Flat slab column heads, in contrast, are clearly not derived from
steelwork practice (Figure 5.5). O.W. Norcross took out a patent for flat slabs in
the USA in 1902. His fellow-countryman, C.A.P. Turner, was building flat slabs by
1906, and the first American load tests on flat slabs were performed by A.R. Lord
in 1910.
American practice favoured four layers of slab reinforcement, rather than two
as is now customary. The additional reinforcement layers at 45° to the column grid
were provided in pragmatic recognition that the principal moments were not

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:16:44
05.qxd 15/11/2001 15:47 Page 87

The development of reinforced concrete design and practice 87

Figure 5.3 Boots ‘wets’


factory (1932). Under grey
skies before recent restora-
tion and cleaning.

Figure 5.4 Early concrete


structure resembling
steel-framed construction
(Co-operative Wholesale
Society warehouse on
Quayside, Newcastle-upon-
Tyne, a Hennebique structure
of 1897–1900).

always parallel to the column lines. Nevertheless, the simpler arrangement of two
layers at right angles can resist the bending moment components adequately if
designed for the appropriate values.
Robert Maillart in Switzerland advocated this simpler approach, and based his
designs on tests carried out in 1908 and 1913–14.14 ‘Mushroom’ column heads of
more or less elegance were developed independently in the USA and mainland
Europe in this period, with the Americans leading in the adoption also of ‘drop’
panels as another way to cope with high shear stresses around columns.
Some use were made of flat slabs in the UK before the early 1930s by innovative
engineers such as Sven Bylander, but wider application was probably inhibited
because design guidance was not readily available.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:16:44
05.qxd 15/11/2001 15:47 Page 88

88 Historic concrete

Figure 5.5 Flat slab


reinforced concrete
construction with ‘mushroom’
heads (Boots ‘wets’ factory,
1932, by Sir Owen Williams)
(by courtesy of the Builder
Group).

The first British Design of flat slabs was one of the significant additions to guidance provided in
the first British code of practice (as opposed to regulations) in 1934.15
concrete code Another was improved guidance on column design. Oscar Faber, a practising
engineer with a strong interest in structural behaviour, questioned the validity of
the modular ratio design method in columns. He carried out tests in 1927 to assess
the effects of shrinkage and creep on the current assumptions for reinforced con-
crete design, and concluded that the modular ratio method was illogical for
columns. It obliged the designer to assume that the column steel was elastically
stressed at, typically, 15 times the compressive stress in the adjacent concrete. The
reality was quite different. Creep of the compressed concrete would increase com-
pression in the steel, and shrinkage would shed further load from the concrete
into the steel. Work at the Building Research Station confirmed this.16,17 So it was
logical on that evidence to abandon the modular ratio method for columns, and
to use instead the arithmetic sum of allowable loads on concrete and steel to give

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:16:44
05.qxd 15/11/2001 15:47 Page 89

The development of reinforced concrete design and practice 89

the column’s capacity. (This was, of course, still some time away from the more
recent ‘plastic’ approach of working out the column’s ultimate strength, based on
the combined contribution of both concrete and steel, and taking its load capacity
as a suitably factored proportion of that.)
The 1934 code was the result of a review of current practice by the Reinforced
Concrete Structures Committee appointed by the Building Board of the Department
of Scientific and Industrial Research in 1931. The 1915 regulations, which formed
their starting point, had been based on the Joint Committee’s report of 1911 —
so there were two decades of experience and development to be considered.
Attention was given to the strength requirements for concrete mixes. Volume
batching was still the norm (as it was to be until the 1950s). There were four classes,
I–IV, these being designated respectively 1 : 1 : 2, 1 : 1.2 : 2.4, 1 : 1.5 : 3, and 1 : 2 : 4;
these were the proportions of cement : fine aggregate : coarse aggregate by volume.
But each class could be of three grades, use of which depended on the level of
control and supervision. Ordinary Grade concrete was to have cube tests taken
only when directed by the designer. Higher Grade concrete was to have prescribed
preliminary and works tests (including daily works consistence or ‘slump’ tests),
and the work was to be carried out under the direction of a foreman and a super-
visor or clerk of works, both to be experienced in such work. For Special Grade
concrete, further controls on materials consistency were called for, while the
designer was to design the structure taking account of continuity in all members.
In return for this, greater stresses were permitted on the concrete. For a 1 : 2 : 4
Ordinary Grade concrete, the permissible compressive stress of 600 lb/in2
(4.1 N/mm2) in the 1915 regulations was increased now to 750 lb/in2 (5.2 N/mm2).
The corresponding Higher Grade figure was 950 lb/in2 (6.5 N/mm2). Special Grade
compressive stresses were to be calculated from preliminary test cube results
divided by 5, but not to be more than 25% above the Higher Grade figures.
The modular ratio for design was now linked to the concrete strength (recog-
nizing that Young’s modulus for concrete is, of course, related to its strength). The
modular ratio in Imperial units was 40,000/(3 times the permissible bending stress,
i.e. the specified 28-day works cube strength); in SI units the numerator would
be 276. For an Ordinary Grade 1 : 2 : 4 mix, with a 28-day works cube strength of
2250 lb/in2 (15.5 N/mm2), the modular ratio would now be about 18.
For reinforcement, permissible stresses in bending and shear were given as 0.45
times the steel yield point stress. For ordinary mild steel this was 18,000 lb/in2
(124 N/mm2). In compression the permissible stresses were three-quarters of those
for bending. Reference was also made to BS 15 for a higher-grade steel and to
BS 165 for hard-drawn steel wire,18,19 for which higher permissible stresses were
given, acknowledging the developing use of high-tensile steels for rod and fabric.
The handbook to the code20 comments on the benefits of mechanical vibration
as an effective means of compacting concrete using a lower water–cement ratio,
with benefits to strength and hardening rate. Again durability was not explicitly
identified as a consideration.
The factor of safety for reinforcement was declared in the handbook as 1/0.45,
i.e. 2.2, on yield. That for concrete was 3 on 28-day works cube strength. The hand-
book warned that use of higher-strength steel could increase the risk of cracking
in tension zones. Also, the modular ratio design method meant that the compressive
steel stress in flexure was limited to, typically, 11–18 times the adjacent concrete
stress; so there was no great benefit, in general, in using high-tensile steel for
strength unless cracking was either acceptable or investigated. Neither did the code
recognize any higher permissible bond stress for deformed (cold-worked) bars,
although it had long been argued, rightly, that deformed or ribbed bars had
improved anchorage and bond properties.
Guidance was given on the calculation of moments in flat slabs and continuous
members, and redistribution of hogging moments by up to 15% was permitted in

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:16:44
05.qxd 15/11/2001 15:47 Page 90

90 Historic concrete

beams and slabs. This acknowledged the relaxation of hogging moments due to
creep in the compressed concrete at monolithic supports. It was also of practical
benefit in reducing steel congestion and the associated risk of poor concrete
compaction over supports.
Appendices to the code included loadings, as a British standard on loadings had
yet to appear.
The code was based on working-load design, a philosophy that was unchallenged
in the UK until the issue of CP 110 in 1972, as discussed below.

Structural The code arrived in time to be of use to the new generation of structural engi-
neers and contractors, mostly from outside Britain, who were keen to collaborate
concrete in the with architects of the Modern Movement. Ove Arup, Oscar Faber, and Felix
1930s Samuely relished the chance to design unusual structures with architects such as
Berthold Lubetkin and the Tecton group. Chapter 2 has considered this in more
detail; here it is of interest to note two buildings which were structured with only
limited assistance from the code.
The 1933 Penguin Pool at London Zoo (Arup and Samuely with Tecton,
Figures 5.6 and 5.7) was an exercise in structural gymnastics in which the spiral
ramps are clearly subject to torsion. British concrete codes before 1972 were coy
about torsion, although spiral staircases and other structures reliant on torsional
strength had been built as early as 1900. So the calculations were made from first
(if rather substantial) engineering mathematical principles.
Similarly, the pioneering 1935 Highpoint I flats at Highgate, London (Arup and
Tecton) had a ‘box-frame’ structure above ground-floor level, with slabs carried
on walls with minimal downstanding beams (Figures 5.8 and 5.9). The code of
1934 required reinforced concrete walls to be designed and built to an equivalent
standard of strength to that for other members, but gave no further guidance.
This hinted that the orthodox steel-oriented view of structures being made up of
columns and beams was not easily shaken off.
In 1938 the Ministry of Health promulgated new model building bye-laws which
barely mentioned concrete.21 However, it was by now easier to obtain building

Figure 5.6 Spiral ramp


reinforcement for The
Penguin Pool, London Zoo
(1933) (by courtesy of Ove
Arup & Partners).

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:16:44
05.qxd 15/11/2001 15:47 Page 91

The development of reinforced concrete design and practice 91

regulations approval with the aid of the established 1915 London regulations, and
subsequently the 1934 code, than it had been in the early 1930s. Further encour-
agement came from the LCC’s 1938 guide to construction of buildings in inner
London22 and from a code issued by the Building Industries National Council in
1939,23 although both were somewhat pushed aside by the events of the Second
World War.

Figure 5.7 The Penguin


Pool, London Zoo completed
(by courtesy of Ove Arup &
Partners).

Figure 5.8 Highpoint I flats


at Highgate, London under
construction (1935) (by
courtesy of Ove Arup &
Partners).

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:16:44
05.qxd 15/11/2001 15:47 Page 92

92 Historic concrete

Figure 5.9 Highpoint I flats


completed (by courtesy of
Ove Arup & Partners).

Materials developments proceeded. The 1934 code countenanced use of rapid-


hardening Portland cement, Portland-blastfurnace cement, and — significantly —
high-alumina cement, even though not all were covered by British standards.
Chapter 6 gives fuller attention to changes in cement and concrete.
The long-awaited BS 882 for coarse and fine aggregates appeared in 1940.24
Reinforcement standards were developing. BS 785 appeared in 1938 for hot-rolled
bars and hard-drawn wire,25 BS 1144 in 1943 for cold-worked bars,26 and BS 1221
in 1945 for steel fabric, i.e. mesh.27 Developments in reinforcement strengths and
permissible tension stresses in bending are reviewed in Table 5.1.

The Second World But this is to run ahead of the Second World War, which at once both halted most
civilian projects and generated huge demand for military works, both offensive
War and defensive. With awareness of the surely inevitable war, and the equal certainty
of bombing raids on town and cities, many local authorities had been already con-
sidering air raid precautions. Arup and Lubetkin argued for massive concrete shel-
ters, in view of their greater safety. Circular shelters were conceived with spiral
floor slabs that could be used as car parks after the war. Construction of these as
proposed would be on the ‘top–down’ principle, as was later widely used on com-
mercial sites in the 1980s for speed of building and minimising damage to adja-
cent buildings from ground movements.28 For various reasons (mainly political
and economic) their proposals were not pursued.
Equally original in concept, but this time realized, were the concrete structures
of the Mulberry Harbours, designed and built for the Allied invasion of occupied
France.29 One can imagine the design of these being carried out with an eye to
rapid construction and ‘buildability’ (see Chapter 13).
One effect of the war was inevitably that many structures were subject to blast
and fire.2 Study of blast-damaged concrete structures led to conclusions that will
surprise few engineers who have, of sad necessity, been involved in more recent
investigations of structures damaged by terrorism or gas explosions. Structural
redundancy was desirable, so that a damaged structure might still be able to stand
even though suffering serious local damage. Generous reinforcement lap lengths
should be provided. Reinforcement should be provided where reversals of stress,

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:16:44
05.qxd 15/11/2001 15:47 Page 93

The development of reinforced concrete design and practice 93

Table 5.1 Developments in specified reinforcement strength and permissible tensile stresses in bending from 1915
(Part 1 and Part 2)

Code and year British Standard Steel type Ultimate Yield stress Permissible tensile
tensile strength (fy) lb/in2 (N/mm2) stress due to bending
lb/in2 (N/mm2) lb/in2 (N/mm2)

Part 1
London Reinforced – Mild – – 16,000 (110)
Concrete Regulations,
1915
Building Research BS 15 Mild 62,720–71,680 – 18,000 (124)
Board (DSIR) (432–494)
Code, 1933 BS 15 Defined yield – 44,000 (minimum) 0.45 fy, i.e. 20,000
point (303) (138)
London Byelaws, 1938 – All – – 18,000 (124)
CP 114: 1948 BS 785 Mild 62,720–71,680 – 18,000 (124)
(432–494)
Medium 73,920–85,120 36,960–43,680 0.5 fy, maximum
tensile* (510–587) (255–301) 27,000 (186)
High tensile* 82,880–96,320 42,560–51,520 0.5 fy, maximum
(571–664) (293–355) 27,000 (186)
BS 1144 Cold worked 0.5 fy, maximum
single twisted: 27000 (186)
ø  3/8 in 80,000 (551) 70,000 (482)
ø  3/8 in 70,000 (482) 60,000 (414)
Cold worked 63,000 (434) 54,000 (372)
twin twisted
Fabric to To BS 785 As above for BS 785
BS 1221 To BS 1144 As above for BS 1144
Expanded metal 75,000 (517) 50,000 (345) 20,000 (138) ø  11⁄2 in
CP 114: 1957 BS 785 Mild, no defined 62,720–71,680 – 20,000 (138); ø  11⁄2 in
yield point (432–494) 18,000 (124)
BS 785, Guaranteed yield – fy 0.5 fy, maximum
BS 1144, stress, high-bond 30,000 (207)
BS 1221 or mesh
CP 114: 1965 BS 785, Guaranteed yield N/A fy 0.55 fy maximum
BS 1144, stress, high-bond (33,000) 227 for
BS 1221 or mesh ø  7/8 in, (30,000)
207 for ø  7/8 in
CP 114: Part 2: BS 785 Mild, no defined 432–494 – 140 for ø  40 mm,
1969 (Metric) yield point 125 for ø  40 mm
BS 785, Guaranteed yield – fy 0.55 fy maximum 230
BS 1144, stress, high-bond for ø  20 mm, 210
BS 1221 or mesh for ø  20 mm
CP 110: 1972 BS 4449 Hot rolled mild – 250† –
BS 4449 Hot rolled high – 410† –
yield
BS 4461 Cold worked high – Maximum 460 for –
yield ø  16 mm, 425
for ø  16 mm†
BS 4482 Hard drawn wire – 485† –
(ø  12 mm)
CP 110: 1972 BS 4449, Grade 460/425 – Maximum 460 for –
(amended 1980) BS 4461 ø  20 mm, 425
for ø  20 mm†
BS 4482, Hard drawn wire – 485† –
BS 4483 (ø  12 mm)

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:16:44
05.qxd 15/11/2001 15:47 Page 94

94 Historic concrete

Table 5.1 Continued

Code and year British Standard Steel type Ultimate Yield stress Permissible tensile
tensile strength (fy) lb/in2 (N/mm2) stress due to bending
lb/in2 (N/mm2) lb/in2 (N/mm2)

CP 110: 1972 BS 4449 Grade 460 – 460† –


(amended 1983)
BS 8110: 1985 BS 4449 Hot rolled mild – 250† –
BS 4449, High yield – 460† –
BS 4482, (hot rolled or
BS 4483 cold worked)

*Ultimate tensile strength and yield stress varied depending on bar diameter, smaller bars having higher strengths.

Redefined as characteristic strength in and after CP 110: 1972.
Note: during the mid-1960s, the Lancashire Steel Manufacturing Co. produced a hot rolled ribbed bar (Lancs 80) with fy of 82,000 lb/in2 (565 N/mm2)
which was used with a permissible tension stress due to bending of 0.55 fy, i.e. 44,000 lb/in2 (303 N/mm2).

as from uplift pressure, could otherwise cause supports and slabs to fail in tension
or shear. Similarly, fire is most likely to severely damage structures with highly
stressed steel having shallow cover, particularly cold-worked steel which loses its
strength enhancement and reverts to mild steel at around 400°C.
A committee convened by the Institution of Structural Engineers and chaired
by Oscar Faber sat in 1942–43 with a brief to propose post-war building methods
which would produce rapid, economic construction.30 The particular circumstances
were, of course, related to the severe destruction of large areas of towns and cities
by bombing, allied to the generally accepted need to improve housing standards.
The committee’s technical conclusions were generally sound, prudent and
undramatic, such as that the reinforcement standard BS 78525 should include yield
stress. Previously it gave only ultimate tensile strength figures. Reinforcement could
then be stressed at one-half of this yield stress, effectively reducing the steel
factor of safety from 2.2 to 2. Permissible concrete compressive stresses should be
increased by 10%; increased stresses due to wind load only should be permitted;
prestressed concrete should be encouraged, with a code to be provided when
experience allowed (it came out as CP 115 in 1959).31 The mechanical vibration of
concrete should be encouraged, although — as before — the argument for this was
stated to be on grounds of improved strength, without reference to durability.
Of some interest were the report’s recommendations relating to improved design
and construction methods. Design and construction in reinforced concrete should
be carried out only by those with suitable experience. More effort should be made
to train concretors and to give them the status of craftsmen. Formworking and
steel fixing should also be recognized as specific trade activities, and supervisors
should be suitably trained. Standard specifications should be adopted. These calls
echo across more than half a century with a somewhat depressing familiarity.
One particular recommendation was for the engineering designer to be involved
at the outset of a project. This is, or should be, now taken for granted; but many
architects were still keen to take their design to an advanced stage, so that beam
and column layouts and dimensions were basically fixed, before a consulting engin-
eer or design-and-build contractor was appointed (who would then complete the
detailed design). One reason for this could be that an architect was understand-
ably loath to relinquish control of the structural planning. Another consideration
was that the architect’s fee might be reduced by the client if a consulting engineer
were appointed, or indeed the architect might be expected to pay the consultant’s
fee himself. This problem did not arise if a concrete specialist were employed.
In contrast, there were some who favoured the closely involved approach. Owen
Williams as architect-engineer straddled both disciplines. Ove Arup — having
left the contracting firm of Kier for which he designed the Penguin Pool and

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:16:44
05.qxd 15/11/2001 15:47 Page 95

The development of reinforced concrete design and practice 95

Highpoint — collaborated as an independent consulting engineer with like-minded


architects, founding Ove Arup & Partners in 1946. Oscar Faber was an early prac-
titioner of multi-disciplinary engineering consultancy.

CP 114: 1948 Soon after the Faber report appeared, the war ended and work began on updat-
ing the 1934 code. Conditions were difficult. Structural steelwork and reinforcement
were in short supply, which gave an incentive to prestressed concrete, as prestress-
ing steel was not rationed (see Chapter 10). In the absence of orthodox reinforce-
ment, many concrete elements of this period were reinforced with old tram rails,
scrap electrical conduit, and the like. (This approach, of necessity rather than choice,
is an interesting precursor of today’s environmentally-favoured ‘recycling’.)
It was decided that the new code should fall within the structural series being
prepared by the British Standards Institution (originally numbers 111–118 cov-
ering the common structural materials), and accordingly CP 114 first appeared
in 1948.32 Water–cement ratio was prescribed, with a 20% reduction in water for
vibrated concrete. Such vibrated concrete could be stressed 10% higher in compres-
sion. Cover on external faces was to be increased, as also for internally corrosive
conditions, by 1⁄2 in (13 mm) above the nominal internal value of 1⁄2 in (13 mm), or
the bar diameter if this were greater.
A maintenance clause was introduced, which recommended inspections at
suggested intervals of 3–5 years to identify cracking or corrosion of reinforcement.
It pointed out that a little work done when the problem was small would be prefer-
able to remedying a major problem if it were neglected. Unfortunately, the cost
of access for external inspections was often seen as too high (especially for high-
rise buildings) and, as now, maintenance inspections were all too often deferred.
A supplementary code, CP 114.100–114.105,33 was introduced in 1950. This
covered suspended concrete floors and roofs (including stairs) and comprised six
‘sub-codes’. Sub-code 114.100 gave general recommendations for these elements,
while the remaining five sub-codes added specific recommendations for solid slabs
supported by beams; flat slabs; ribbed slabs; pre-cast elements; and filler joist
construction.

Post-war The design of concrete structures in the late 1940s and early 1950s was restricted
by shortages of materials (particularly reinforcement) and planning controls. Never-
reconstruction theless, some notable structures were built. These included the Brynmawr rub-
ber factory (described in Chapter 8) and some early multi-storey public housing
projects such as the 1949 Rosebery Avenue project in Finsbury, London, by
Tecton and Ove Arup & Partners34 (Figures 5.10 and 5.11). This developed the
Highpoint principles, and used a novel jacking system to fix and strip the reusable
formwork.
In Bristol, Felix Samuely designed an unusual factory structure (Colodense,
1951, Figures 5.12 and 5.13). This combined precast concrete arch frames with
prestressed ties, and used composite floors with prestressed planks placed, as if
they were reinforcement, in tension zones.35 The high strength of prestressing steel
allowed more efficient use of the limited quantities of steel available in periods of
shortage.
Much rebuilding was needed after the war, and various solutions were employed
using concrete. Prefabricated housing and bungalows were developed to a myr-
iad of designs. Soon would come the large panel precast concrete systems for tower
blocks of flats (and, often overlooked, many lower-rise buildings, too).36,37 Some
systems were purpose-designed for building types, such as CLASP (Consortium
of Local Authority Schools Projects). Although conceived around a steel frame,
this system included precast concrete cladding units, the use of architectural

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:16:44
05.qxd 15/11/2001 15:47 Page 96

96 Historic concrete

Figure 5.10 Box-frame flats


under construction at
Rosebery Avenue, Finsbury,
London (1949) (by courtesy
of Ove Arup & Partners).

Figure 5.11 Detail of


shuttering system used at
Rosebery Avenue (by
courtesy of Ove Arup &
Partners).

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:16:44
05.qxd 15/11/2001 15:47 Page 97

The development of reinforced concrete design and practice 97

Figure 5.12 The Colodense


factory under construction at
Malago, Bristol (1951) (by
courtesy of F.J. Samuely &
Partners).

Figure 5.13 Detail of


prestressed planks used as
reinforcement in the
Colodense factory (by
courtesy of F.J. Samuely &
Partners).

cladding in concrete38 having grown vigorously since the war. (A subsequent period
when exposed concrete fell out of favour, largely due to poor appearance after
weathering when not thoughtfully detailed, has recently been succeeded by another
upsurge in the use of ‘architectural concrete’.)
Another form of concrete construction widely used for housing in the early post-
war period was ‘no-fines’, which as its name indicates was based on a concrete made
with cement and coarse aggregate only, omitting the sand. The resulting ‘honey-
comb’ consistency was thermally efficient and resistant to water penetration if built
of reasonable thickness. Originally developed in Holland,36 the technique was
employed almost exclusively in the UK by the contracting firm Wimpey, who

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:16:44
05.qxd 15/11/2001 15:47 Page 98

98 Historic concrete

combined it with reusable shuttering to produce economical housing using rela-


tively unskilled labour. Reportedly some 300,000 low-rise dwellings were built;39
the relatively modest strength of the material led to construction that resem-
bled traditional masonry wall construction, without large windows or narrow piers
between them. No-fines was also used in high-rise concrete frames as infill walling.
Proprietary concrete floor systems proliferated, mostly incorporating precast
components. A selection in use in the middle part of the 20th century is illustrated
in Appendix B.
Some floors, such as those by Siegwart and Armocrete, dated their origins back
to early this century. Others, of more recent origin, used prestressed wires in long-
line moulds to produce standard concrete joists that could literally be cut to the
required length after the concrete had hardened. For these, high-alumina cement
(HAC — developed in France and originally known as ciment fondu) was widely
if not always used in the concrete, as its very rapid hardening and gain of strength
allowed swift turnaround and efficient use of the moulds. Several localized
failures of purpose-designed HAC units in the mid-1970s drew attention to some
loss of strength due to chemical changes in the HAC.40 However, since then, no
further failures have been reported in what remain as several million square metres
of such construction, most of which is different in the details from the structures
that suffered failure, and contains mass-produced quality-controlled units.
Other precast floor systems rely on composite action between precast units, or
ribs, and an in-situ infill. The precast sections can often span in the temporary
condition between supports and carry the wet concrete weight without propping.
With joists and ribs a variety of infill blocks, often hollow, can be laid to reduce
floor weight and cost. Still other floor systems include reinforced or prestressed
hollow-core and plank units. All such flooring systems generally, of course, elim-
inate the need for formwork and minimize or eliminate use of propping.

Codes and It may be useful to link briefly the code of 1948, the first CP 114, forward through
its successors to today’s BS 8110.41
standards since The 1957 edition of CP 114,42 although still based on working-load design
1948 philosophy, introduced the load-factor method for slab and beam design as an alterna-
tive to modular ratio design. It also allowed mixes to be specified by strength —
typically 3000, 3750, or 4500 lb/in2 (21, 26, or 31 N/mm2) — and to be designed
by the contractor to achieve this specified figure. This gave scope for economy in
the mix design. A 10% permissible stress increase was allowed for a designed-mix
concrete. The load-factor approach was based on ultimate load theory, but loads
considered in design remained service loads, and the concrete cube strength and
steel yield stress were factored down to give permissible stresses. The factor of safety
for steel was now 1.8, a further 10% reduction, while that for concrete was 3 for
nominal (by volume) mixes and 2.73 for designed mixes, a 10% reduction over
nominal mixes.
The sub-codes CP 114.100–114.104 issued in 1950 were incorporated into the
1957 edition of CP114, while CP114.105 for filler joist construction was withdrawn
from the concrete code as its recommendations were, rather more logically, to be
found in the structural code for steel, BS 449.
The amended 1965 CP 114,43 a reprinting, recommended that all beams (except
for lintels and other such minor items) should be provided with nominal shear
reinforcement.
Separate codes for prestressed concrete design, CP 115,31 and for precast
concrete design, CP 116,44 appeared in 1959 and 1965 respectively.
Metric editions of all three codes were introduced in 1969, with CP 116 being
amended in 197045 to give enhanced recommendations for tying and robustness,
following the Ronan Point collapse in 1968.46

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:16:44
05.qxd 15/11/2001 15:47 Page 99

The development of reinforced concrete design and practice 99

In 1972 the three codes were unified into one document, CP 110.47 This adopted
limit-state design exclusively, with design generally being based on strength and
stability, and with detailing rules being used to satisfy stiffness and serviceability
once basic member sizes were known.
This in turn was replaced by the present BS 8110, first issued in 1985.41 The
European pre-standard DD ENV 1992-1-148 was issued in 1992.
Some structural engineers were not persuaded that longer, more complex codes
based on limit-state principles were necessary, or indeed desirable, for the design
of routine building structures. In 1987 a referendum of its members was held by
the Institution of Structural Engineers, in which a large majority of those voting
were in favour of retaining the ‘permissible stress’ design methods embodied in
CP 114 and its predecessors. In response, the Institution commissioned prepar-
ation of a design manual incorporating permissible stress recommendations
updated from CP 114, together with supplementary design guidance.49 This
appeared in 1991, six years after a similar design manual based on BS 8110.50
Standards for reinforcement and concrete have also evolved, reflecting
changes in practice and materials.
High yield reinforcement has gradually displaced mild steel. Initially the higher
strength was achieved by cold-working, typically by stretching and/or twisting the
bars (either singly or in pairs) to produce a distinctive ‘barley sugar’ spiral pro-
file. This was reflected in the brand names of bars such as Twisteel and Twin Twisted
(also known as Isteg). Tentor bars were probably named in recognition of their
being subject to both tension and torsion during the cold-working process. Fabric
or mesh has usually been made from hard-drawn wire, spot-welded at the rod
intersections with care taken to avoid annealing the steel back to a weaker mild
state. More recently, hot-rolled high yield reinforcing bars have been widely used.
For concrete, durability considerations have at last received their necessary recog-
nition, with minimum concrete grades, cement contents, and maximum
water–cement ratio being now defined in BS 8110 for various conditions of expo-
sure. Typical grades of concrete are consequently now likely to give 35–40 N/mm2
cube crushing strength — well above the 21–31.5 range typically quoted in CP 114.
Changes in reinforcement strengths and permissible stresses are given in
Table 5.1. Typical concrete grades and permissible flexural compressive stresses
have changed too, as shown in Table 5.2.

Changes in design This chapter has attempted to chart the evolution of design theory and practice
from before the turn of the century to 1948, and briefly beyond there to the pre-
practice sent day. This chapter’s title could equally be ‘the development of standards for
the use of reinforced concrete’. Certainly, in more recent years, the tendency has
been for more and more design issues and details to be codified and prescribed.
There is thus less incentive for the designer — particularly of ‘routine’ structures —
to take other routes, even though current building regulations for structure (e.g.
for England and Wales51) make it clear that the designer is not obliged to use an
available code if another sound method can be justified.
Increasingly, too, the whole process of concrete design is being automated. Today,
engineers and technicians can model, view, analyse, design and detail structures
sitting in front of a PC. This is somewhat different from the days — still within
the experience of engineers working today — when the tools of the trade were a
drawing board and T-square, pencil, pad, slide-rule, and one or more design
aids — books, and ‘ready reckoner’ charts and tables.
Older practitioners will remember, too, the use of moment distribution as devised
by Professor Hardy Cross,52,53 itself an advance on the laborious hand solutions
of slope-deflection equations or the slightly less tedious ‘theorem of three moments’.
Concrete codes from the outset recognized the monolithic nature of reinforced

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:16:44
05.qxd 15/11/2001 15:47 Page 100

100 Historic concrete

Table 5.2 Developments in specified concrete compressive strength and permissible flexural compressive stresses
from 1915

Code and year Concrete grade or mix Batch proportions Works cube strength Permissible flexural
by volume at 28 days compressive stress
lb/in2 (N/mm2) lb/in2 (N/mm2)

London Reinforced – 1:1:2 2200 (15.2) 750 (5.2)


Concrete Regulations, 1.5:2: 4 2000 (13.8) 700 (4.8)
1915 1.2:2: 4 1800 (12.4) 650 (4.5)
1:2:4 1600 (11.0) 600 (4.1)
Building Research Ordinary grade 1:1:2 2925 (20.2) 975 (6.7)
Board (DSIR) Code, 1 : 1.2 : 2.4 2775 (19.1) 925 (6.4)
1933 1 : 1.5 : 3 2550 (17.6) 850 (5.9)
1:2:4 2250 (15.5) 750 (5.2)
Higher grade 1:1:2 3750 (25.8) 1250 (8.6)
1 : 1.2 : 2.4 3600 (24.8) 1200 (8.3)
1 : 1.5 : 3 3300 (22.7) 1100 (7.6)
1:2:4 2850 (19.6) 950 (6.5)
Special grade 1:1:2 Not more than 25% above values for higher
1 : 1.2 : 2.4 grade concrete
1 : 1.5 : 3
1:2:4
London Byelaws, Ordinary and higher grades 1:1:2 As 1933 Code (1 : 1–2 : 2.4 mix not included)
1938 1 : 1.5 : 3
1:2:4
Special grade – Not included from 1933 Code
CP 114: 1948 Aggregate to BS 882 1:1:2 4500 (31.0) 1500 (10.3)
1 : 1.5 : 3 3750 (25.8) 1250 (8.6)
1:2:4 3000 (20.7) 1000 (6.9)
CP 114: 1957 Ordinary 1:1:2 4500 (31.0) 1500 (10.3)
1 : 1.5 : 3 3750 (25.8) 1250 (8.6)
1:2:4 3000 (20.7) 1000 (6.9)
Special designed mix N/A uw uw/3
CP 114: 1965 Ordinary All As CP 114: 1957
Designed mix N/A uw uw/2.73
CP 110: 1972 Prescribed mix Prescribed in Table 50 (15) N/A
(by weight) (20)
(25)
(30)
Designed mix N/A fcu N/A
BS 8110: 1985 Ordinary prescribed mix See BS 5328 (by weight) Concrete grade must be chosen with regard
Designed mix N/A to exposure conditions and durability
Special prescribed mix See BS 5328 (by weight) (see 33.3 and Tables 3.2–3.4)

concrete, and formulae were provided to assist in the calculation of fixity moments.
Nevertheless, continuous beams and slabs of varying spans and frames of asym-
metrical dimensions and member sizes could oblige the designer to undertake time-
consuming and tedious calculations.

Technical The following briefly highlights aspects of the much more comprehensive advice
on information sources to be found in the Appendix.
information and Of the standard textbooks, that by Reynolds54 has been, over six decades, one
design guidance of the most used guides, together with the explanatory handbooks to the various
design codes from 1934.20,55,56
In addition to textbooks already noted, special mention should be made of the
magazine Concrete and Constructional Engineering, published from 1906 until 1966.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:16:44
05.qxd 15/11/2001 15:47 Page 101

The development of reinforced concrete design and practice 101

Much useful guidance for designers and constructors was contained in the numer-
ous books from Concrete Publications Limited, which are listed in the Appendix.
The Cement and Concrete Association (founded in 1935, and now the British
Cement Association), and the Concrete Society (founded in 1966), have both
published many useful design and practice guides and other reference works, as
have other more specialist sources. Such guidance, whether in books, on CD-ROM,
or in some yet-to-be-imagined future medium, emphasises that design and
construction in reinforced concrete is not entirely circumscribed by codes, but is
indeed a practice.
For today’s engineer called upon to assess or alter such structures, these contem-
porary publications are invaluable. The library of the Institution of Civil Engineers
and that of the Institution of Structural Engineers both hold excellent collections
of such material. More recent works include a review of UK reinforcement
standards57 and a paper on British concrete codes and regulations prior to the
issue of CP110 in 1972.58 Numerous reports and pamphlets on the identification,
investigation, and assessment of proprietary concrete systems for housing and
other building types are available from the Building Research Establishment. The
Institution of Structural Engineers has published a general guide to the appraisal
of existing structures.59
Guidance on the repair and maintenance of concrete structures is also widely
available. A BRE report60 describes general principles, while the assessment and
repair of corrosion-damaged concrete is discussed by Pullar-Strecker.61 BS 6089
gives guidance on the assessment of concrete strength in existing structures.62 A
growing number of BRE and Concrete Society publications deal with specific
aspects of investigation, assessment and repair.

Acknowledgements The author wishes to thank his fellow co-authors for their encouragement and
for helpful criticism of the text. He also thanks Mr D.K. Doran for information
on the use of ‘no-fines’ concrete, and Dr L.G. Booth for greatly enlarging on the
author’s passing reference to the works produced by Concrete Publications
Limited. Dr Booth’s bibliography of their works is included in the Appendix with
his kind permission.

References 1. Hyatt, T., An Account of Some Experiments with Portland-Cement Concrete, Com-
bined with Iron as a Building Material, with Reference to Economy of Metal in Con-
struction, and for Security Against Fire in the Making of Roofs, Floors and Walking
Surfaces. Chiswick Press: London, 1877.
2. Hamilton, S.B., A Note on the History of Reinforced Concrete in Buildings. HMSO:
London, National Building Studies Special Report No. 24, 1956.
3. Wayss, G.A., Das System Monier. G.A. Wayss: Berlin, 1887.
4. Coignet, E., Tédesco, N. de., Du calcul des ouvrages en ciment avec ossature metallique.
La Société des Ingénieurs Civils de France: Paris, 1894.
5. Christophe, P., Le beton armé et ses applications, 2nd edn. Béranger: Paris, 1902.
6. Marsh, C.F., Reinforced Concrete. Constable: London, 1904.
7. Joint Committee on Reinforced Concrete, Report of the Joint Committee on Reinforced
Concrete. J. Roy. Inst. Br. Archit. (3rd ser.), 1907, 14(15), 513–41 (discussion 497–505).
8. London County Council. London County Council (General Powers) Act. HMSO:
London, 1909.
9. Committee on Reinforced Concrete, Interim Report on Reinforced Concrete. Insti-
tution of Civil Engineers: London, 1910.
10. Joint Committee on Reinforced Concrete, Second Report of the Joint Committee on
Reinforced Concrete. Royal Institute of British Architects: London, 1911.
11. London County Council, Reinforced Concrete Regulations. LCC: London, 1915.
12. Cottam, D., Sir Owen Williams 1890–1969. Architectural Association: London,
1986.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:16:44
05.qxd 15/11/2001 15:47 Page 102

102 Historic concrete

13. Architects’ Journal, Messrs Boots Factory, Beeston, Nottingham. Arch. J., 1932, 76, 3
August, 125–36.
14. Billington, D.P., Robert Maillart’s Bridges: The Art of Engineering. Princeton University
Press: Princeton, NJ, 1979.
15. Reinforced Concrete Structures Committee, Report of the Reinforced Concrete Struc-
tures Committee of the Building Research Board. HMSO: London, 1933.
16. Glanville, W.H., Studies in Reinforced Concrete II: Shrinkage Stresses. HMSO:
London, Building Research Technical Paper No. 11, 1930.
17. Glanville, W.H., Studies in Reinforced Concrete III: The Creep or Flow of Concrete
Under Load. HMSO: London, Building Research Technical Paper No. 12, 1930.
18. Engineering Standards Committee, Standard Specification for Structural Steel Bridges
and General Building Construction. Engineering Standards Committee: London, 1912,
BS 15.
19. British Standards Institution, Hard Drawn Steel Wire for Concrete Reinforcement. BSI:
London, 1929, BS 165.
20. Scott, W.L., Glanville, W.H., Explanatory Handbook on the Code of Practice for Rein-
forced Concrete. Concrete Publications Limited: London, 1934 (with later editions
accompanying and explaining the editions of CP 114 of 1948, 1957 and 1965).
21. Ministry of Health, Model Byelaws Series IV Buildings. HMSO: London, 1938.
22. London County Council, Construction of Buildings in London. LCC: London, 1938.
23. Building Industries National Council, Code of Practice for the Use of Reinforced
Concrete in the Construction of Buildings. BINC: London, 1939.
24. British Standards Institution, Coarse and Fine Aggregates from Natural Sources for
Concrete. BSI: London, 1940, BS 882.
25. British Standards Institution, Rolled Steel Bars and Hard-Drawn Steel Wire for Con-
crete Reinforcement. BSI: London, 1938, BS 785.
26. British Standards Institution, Cold Twisted Steel Bars for Concrete Reinforcement. BSI:
London, 1943, BS 1144.
27. British Standards Institution, Steel Fabric for Concrete Reinforcement. BSI: London,
1945, BS 1221 (covered expanded metal, as well as hard-drawn steel wire and twisted
steel fabric).
28. Arup, O.N., Design, Cost, Construction and Relative Safety of Trench, Surface, Bomb-
proof and Other Air-raid Shelters. Concrete Publications Limited: London, 1939.
29. Institution of Civil Engineers, The Civil Engineer in War, Vol. 2. ICE: London, 1948.
30. Reinforced Concrete Structures Committee, Reinforced Concrete Structures. HMSO:
London, Post-war Building Studies No. 8, 1944.
31. British Standards Institution, The Structural Use of Prestressed Concrete in Buildings.
BSI: London, 1959, CP 115.
32. Codes of Practice Committee for Civil Engineering, Public Works and Building, The
Structural Use of Normal Reinforced Concrete in Buildings. British Standards Insti-
tution: London, 1948, CP 114.
33. The Council for Codes of Practice for Buildings, Suspended Concrete Floors and Roofs
(Including Stairs). British Standards Institution: London, 1950, CP114.100–114.005.
34. Collins, A.R. (ed.), Structural Engineering — Two Centuries of British Achievement.
Tarot Print: Chislehurst, 1983.
35. Higgs, M. (ed.), Felix James Samuely. Archit. Assoc. J., 1960, 76, June, 2–31.
36. Finnimore, B., Houses from the Factory: System Building and the Welfare State,
1942–74. Rivers Oram Press: London, 1989.
37. White, R.B., Prefabrication: A History of its Development in Great Britain. HMSO:
London, National Building Studies Special Report No. 36, 1965.
38. Morris, A.E.J., Precast Concrete Cladding. Fountain Press: London, 1966.
39. Reeves, B.R., Martin G.R., The Structural Condition of Wimpey No-fines Low-rise
Dwellings. BRE: Garston, 1989, Report BR153.
40. Bate, S.C.C., High Alumina Cement Concrete in Existing Building Structures. HMSO:
London, 1984, Report BR235.
41. British Standards Institution, Structural Use of Concrete. BSI: London, 1985, BS 8110.
42. British Standards Institution, The Structural Use of Reinforced Concrete in Buildings.
BSI: London, 1957, CP 114.
43. British Standards Institution, The Structural Use of Reinforced Concrete in Buildings.
BSI: London, 1965, CP 114 (reset and reprinted).

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:16:44
05.qxd 15/11/2001 15:47 Page 103

The development of reinforced concrete design and practice 103

44. British Standards Institution, The Structural Use of Precast Concrete. BSI: London,
1965, CP 116.
45. British Standards Institution, Large Panel Structures and Structural Connections in
Precast Concrete. BSI: London, 1970, Addendum No. 1 (1970) to CP 116: 1965 and
CP 116: Part 2: 1969.
46. Ministry of Housing and Local Government, Report of the Inquiry into the Collapse
of Flats at Ronan Point, Canning Town. HMSO: London, 1968.
47. British Standards Institution, The Structural Use of Concrete. BSI, London, 1972,
CP 110.
48. British Standards Institution, Eurocode 2: Design of Concrete Structures. General Rules
for Buildings (Together with United Kingdom National Application Document). BSI:
London, 1992, DD ENV 1992-1-1.
49. Institution of Structural Engineers, Recommendations for the Permissible Stress Design
of Reinforced Concrete Building Structures. ISE: London, 1991.
50. Institution of Structural Engineers and Institution of Civil Engineers, Manual for the
Design of Reinforced Concrete Building Structures. ISE: London, 1985.
51. Department of the Environment and the Welsh Office, The Building Regulations 1991:
Approved Document A — Structure. HMSO: London, 1991.
52. Cross, H., Continuity as a factor in reinforced concrete design. J. Am. Concrete Inst.,
1929, 25, December, 669–711.
53. Cross, H., Analysis of continuous frames by distributing fixed-end moments. Trans.
Am. Soc. Civil Eng., 1932, 96, 1–10 (discussion 11–156).
54. Reynolds, C.E., Reinforced Concrete Designer’s Handbook. Concrete Publications Lim-
ited: London, 1932 (later editions in 1939, 1946, 1948, 1957, 1961, 1971, 1974, 1981
and 1988).
55. Bate, S.C.C. et al., Handbook on the Unified Code for Structural Concrete (CP 110:
1972). Cement and Concrete Association, London, 1972.
56. Rowe, R. et al., Handbook to British Standard BS8110: 1985, Structural Use of Con-
crete. Palladian Publications: London, 1987.
57. Steel Reinforcement Commission, UK reinforcement standards 1938 to 1990. Concrete,
1990, 24(3), 40–41.
58. Matthews, D.D., The background to CP110: 1972: The structural use of concrete —
notes on earlier British concrete codes and regulations. Unified Code Symposium, Insti-
tution of Structural Engineers: London, 1973, 5–18.
59. Institution of Structural Engineers, Appraisal of Existing Structures, 2nd edn. ISE:
London, 1996.
60. Currie, R.J., Robery, P.C., Repair and Maintenance of Reinforced Concrete. BRE:
Garston, 1994, Report BR254.
61. Pullar-Strecker, P., Corrosion Damaged Concrete: Assessment and Repair. CIRIA/
Butterworth: London, 1987.
62. British Standards Institution, Guide to Assessment of Concrete Strength in Existing
Structures. BSI: London, 1981, BS 6089.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:16:44
06.qxd 15/11/2001 15:48 Page 105

6 Cement and concrete as materials:


changes in properties, production and
performance

George Somerville
Synopsis This chapter records the changes in cement and concrete as materials since 1900,
in terms of their properties, production and performance. While largely material
orientated, the influence of changes in design standards and construction practices
is also covered. A major factor is increase in strength; as this has occurred, the
emphasis in design has switched to durability, although an integrated approach —
design, materials and construction quality — has not yet fully evolved. It is shown
that change has been triggered by a range of factors, including production meth-
ods, customer demand for improved performance, construction methods and
changes in fashion.

Introduction The starting point for this chapter is the early years of the 20th century, when the
first attempts were made in the UK to produce authoritative specifications for both
cement and concrete. Other contributions in this series cover the earlier devel-
opments for both materials — mainly the inventive and entrepreneurial periods
in the 19th century. A whole range of cements were then available and reinforced
concrete was very much promoted via proprietary systems, each having its own
design method. Without wishing to duplicate that coverage, the author did find
it necessary to delve into previous practice in order to establish clearly his start-
ing point; useful references in this respect are the works by Gooding and
Halstead,1 Halstead,2 Francis,3 Davis4 and Hamilton.5 The first four of these relate
to cement; the last is almost a standard work on reinforced concrete.
There were three significant events in the first decade of the century, which were
to shape the future in the UK. The year 1900 saw the formation of The Associ-
ated Portland Cement Manufacturers (1900) Ltd, an amalgamation of 24 com-
panies. Then in 1904 the first edition of BS 12,6 a specification for Portland cement,
was published by the Engineering Standards Committee of the ICE. Three years
later in 1907, what was effectively the first Code of Practice for reinforced con-
crete was published by the RIBA.7 Before 1904, there were very many different
cements on the market, more than matched by the number of client specifications.
The agreement of a common BS 12 was therefore a major step, subsequently
underpinning the design code,7 which had a strong emphasis on materials as well
as on design calculations.
Using these events as a base, one objective of this chapter is to record the changes
that have since occurred in the characteristics of cement and concrete, while trying
to understand the underlying reasons for these. The second and final objective is
to outline the significance of these changes, in terms of the required and actual
performance of concrete structures in service. The emphasis is therefore on the
significance of change, and not on the basic characteristics of the materials — for
which reference should be made to standard works, such as the books by Lea8
and Neville.9

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:17:14
06.qxd 15/11/2001 15:48 Page 106

106 Historic concrete

Cement standards Since BS 126 was first published, there have been 14 editions in total, the most recent
one being in 1991.10 In 1904, the characteristics that were specified included fine-
ness, specific gravity, chemical composition, tensile strength (from briquettes), set-
ting time and soundness. The basic constituents then, as now, were Portland cement
clinker and calcium sulphate (to regulate setting time). Requirements for fineness
and specific gravity have now gone, and the limits for other characteristics have
changed, as indeed have methods of sampling and testing, of defining chemical
composition, and of testing for strength. It is therefore virtually impossible to make
comparisons on pure specification terms.
Cement standards have changed and developed in other ways. Still operating from
a base of Portland cement clinker and calcium sulphate, there is a range of manu-
factured cements available, created by introducing additional main or minor con-
stituents and additives. In producing specifications for these, there is now a major
European influence. A recent publication11 indicates what cements are available,
while giving guidance on selection. Table 6.1, taken from that publication, provides
a summary; it is interesting to note that the dominant factor in the selection process
is durability, representing a substantial change in performance requirements.
Table 6.1 indicates how the different manufactured cements have evolved, in their
own right. The British Standard for Portland slag cement (BS 146) first appeared
in 1923, while that for Portland fly ash cement (BS 6588) first appeared in 1991.
Much of the additional development recorded in Table 6.1 has been triggered

Table 6.1 Comparison of British and European cements

Cement British Cement Notation Clinker Content of


designation Standard type to in DD ENV content (%) other main
to DD ENV cement DD ENV 197-1 CEM constituents (%)
197-1 197-1

Portland cement BS 12: 1991 I I 95–100 –


BS 4027*
Portland slag BS 146: 1991 II II/A-S 80–94 6–20
cement II/B-S 65–79 21–35
Portland silica None II/A-D 90–94 6–10
fume cement
Portland pozzolana None II/A-P 80–94 6–20
cement II/B-P 65–79 21–35
II/A-Q 80–94 6–20
II/B-Q 65–79 21–35
Portland fly ash BS 6588: 1991 II/A-V 80–94 6–20
cement BS 6588: 1991 II/B-V 65–79 21–35
None II/A-W 80–94 6–20
II/B-W 65–79 21–35
Portland burnt None II/A/T 80–94 6–20
shale cement II/B-T 65–79 21–35
Portland limestone BS 7583: 1992 II/A-L 80–94 6–20
cement None II/B-L 65–79 21–35
Portland composite None II/A-M 80–94 6–20
cement II/B-M 65–79 21–35
Blast furnace BS 146: 1991† III III/A 35–64 36–65
cement None† III/B 20–34 66–80
None† III/C 5–19 81–95
Pozzolanic cement None IV IV/A 65–89 11–35
BS 6610: 1991 IV/B 45–64 36–55
Composite cement None V V/A 40–64 36–60
V/B 20–39 61–80

*Cement to BS 4027, sulphate-resisting Portland cement, is included here since it complies with DD ENV
197-1, CEM I, although it will eventually be covered specifically in a future part of EN 197.

BS 4246: 1991 covers a blast furnace slag content of 50–85%.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:17:14
06.qxd 15/11/2001 15:48 Page 107

Cement and concrete as materials 107

by the move towards European standards. Work on European cement standards


started perhaps 25 years ago and was initially very slow; however, it all acceler-
ated in the 1980s, and what we are looking at in Table 6.1 is essentially the first
editions. However, both slag and pulverized fuel ash (PFA) also have their own
British standard,12,13 and a significant part of UK practice is to combine these with
Portland cement as mixer blends. This practice is comprehensively reviewed in a
Concrete Society report;14 again, durability features strongly in the recommen-
dations made.

Changes in the General


properties of
In the early 1980s, a growing awareness of increases in the strength of ordinary
Portland cement: Portland cement led to several published papers giving details of changes in a
the effect on number of characteristics, as well as reviewing the effects of these on the proper-
concrete ties of concrete. This process was begun by Corish and Jackson,15 followed by
Nixon,16 the Concrete Society17 and, most recently, by Corish.18 The data presented
are comprehensive and will not be repeated here, except to illustrate some of the
key points.

Changes in strength

Table 6.2 is an abridged version of Corish’s18 data on cement strength from 1960 to
1992. Also in this table, an attempt has been made to extrapolate back to the post-
war period, using BRE data from the Concrete Society;17 this is for the weighted
mean values only, and has to be treated with some caution, although the general
trend is approximately correct. Extrapolation back from that period is difficult,
because of the lack of reliable data; there is a strong testing influence here —
prewar testing showed considerable variability, and the methods used may well have
underestimated the real strength of the cement. It is interesting to note that in intro-
ducing the then new CP 114, Faber,19 in 1949, commented that the strength of cement
and rebars had not increased in the previous 10 years; it maybe deduced therefore
that the top figures in Table 6.2 also applied in the late 1930s. Going back beyond
that period, the author found no reliable test data; the only clue was perhaps the
cube strength required on the common 1 : 2 : 4 concrete, being 16.55 N/mm2 at
28 days in 1907,7 and 20.69 N/mm2 in 1933.20 Having made these qualifications,
Figure 6.1 gives an estimate of estimated average cement strength up to 1980,

Table 6.2 Changes in cement strength, according to Corish18

Year Concrete strength (N/mm2) Ratio

3-day 7-day 28-day 3 -day : 28-day 7 -day : 28-day

Weighted mean Range Weighted mean Range Weighted mean Range

Postwar (13) – (20) – (32) – 0.406 0.625


1952 (14) – (22) – (33) – 0.424 0.667
1957 (15) – (23) – (34) – 0.441 0.676
1960 16 14–20 24 20–27 35 27–42 0.457 0.686
1965 17 14–24 25 21–31 37 32–42 0.459 0.676
1970 20 13–26 28 22–35 40 35–46 0.500 0.700
1975 21 15–26 30 22–35 42 36–45 0.500 0.714
1980 24 18–28 33 24–38 44 36–46 0.545 0.750
1985 25 22–29 35 30–38 46 42–48 0.554 0.761
1990 26 22–29 36 31–38 47 42–49 0.553 0.766
1992 25 21–28 34 30–38 46 42–49 0.554 0.739

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:17:14
06.qxd 15/11/2001 15:48 Page 108

108 Historic concrete

Figure 6.1 Estimates of


actual and BS minimum
28-day cement strengths, in
terms of concrete testing at
0.6 water/cement ratio.

Table 6.3 Range of C3S and C2S in Portland cement, 1914–1990 8,17,18

1914–1922 1928–1930 1944 1960 1980 1990

C3S (%) 15–48 19–58 30–50 36–55 45–64 46–60


C2S (%) 15–26 53–14 45–20 37–12 30–11 28–13

relative to the minimum strength in BS 12. Comparisons with Table 6.2 would indi-
cate that the average curve levels off, in the 1980s and 1990s at about 46–47 N/mm2.
Table 6.2 shows an increase in 28-day strength of just over 30% in the period
1960–1992, but with a tendency to level off from 1985 onwards. It has been noted
that fineness has not changed over this period,17 as determined by the air perme-
ability test for surface area. However, this test gives no real measure of particle dis-
tribution, and there is evidence that modern cements have become more uniform,
with less coarse and less very fine material. Conceivably, this is a contributor to
increased strength. A further factor is the increased efficiency in kiln operation,
particularly in the cooling phase, affecting the formation of alite (C3S) crystals.
Equally significant in Table 6.2 is the ratio of early strength to 28-day strength.
This is attributed to changes in the proportions of tricalcium silicate (C3S) to dical-
cium silicate (C2S).15,17,18 Lea8 has indicated that these changes have been going
on for most of this century (see Table 6.3), and therefore it is inferred that the
ratio of early strength to 28-day strength has been increasing steadily over that
time, but with a tendency to level off from 1980 onwards.
A review of the literature suggests that these changes have been brought about
by a combination of factors. Improvements in the manufacture of cement are one
such factor, both in terms of the process itself and in quality control and testing.
Changes in customer demand are a second factor, which shows itself in two ways:
in increases in general strength requirements in sequential design codes (as know-
ledge of behaviour increased); and in changes in the way in which concrete is pro-
duced, e.g. via ready-mixed concrete or precast concrete (where a high early
strength is important). A third factor is the introduction of new technologies
(e.g. prestressed concrete) and construction methods.
So what are the effects of these strength changes for concrete and for perform-
ance of structures in service? This has been discussed by Nixon and Spooner.21
The increase in the early strength of concrete is clearly an advantage in precast
concrete and in modern construction methods, with the emphasis on speed of
construction and buildability. However, the higher early strength implies less

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:17:14
06.qxd 15/11/2001 15:48 Page 109

Cement and concrete as materials 109

strength gain up to 28 days. Nixon and Spooner21 demonstrate that there is still
significant strength gain beyond 28 days, although this is no longer taken into
account in UK structural codes.
Possibly the major concern is with durability and with the resistance to corro-
sion provided by the quality of the cover concrete. With the increase in cement
strength, concrete specifications based only on strength could be satisfied with lower
cement contents and higher water/cement (w/c) ratios — both of these are trends
which increase permeability and decrease durability. In fact, this problem was rec-
ognized in the early 1980s22 between the issuing of CP 110 in 1972 and the appear-
ance of its successor BS 811023 in 1985. Without going into detail, the importance
of w/c ratio and cement content was recognized for durability, as was the practi-
cality of testing for strength in compliance terms: the net effect is durability grades
of concrete in BS 8110, and, since cement characteristics have now substantially
stabilized (Tables 6.2 and 6.3) these can be used with confidence. It is interesting
to note that action on these matters has been taken only relatively recently —
although the importance of w/c ratio was first established in 1918.24

Alkalis

The alkali content of cement became an issue in the 1970s in relation to alkali–
silica reaction (ASR), an expansive reaction in concrete brought about by critical
combinations of reactive aggregates, a high reactive alkali concrete, and moisture.
ASR first came to light in the USA in the 1930s, with the earliest-built structure
dating back to 1914. In the UK, the American experience was known about in
the late 1940s, but conclusions from investigative work in the 1950s and 1960s
indicated that there were no known deposits of reactive aggregates in the UK; this
was still the situation in 1971, when BRE Digest 126 was published.
In 1976, diagnosis of the reaction in foundation blocks for three electricity sub-
stations in south-west England and in a dam in Jersey changed the situation, and
around 200 cases have since been identified. Much development work has been done
in the past 20 years, first to produce specifications which minimize the risk of the
reaction, and second to assess the effects on those structures already diagnosed.
Guidance to minimize the risk of ASR is now incorporated into British Stan-
dards,25 based on work by BRE (Digest 330) and the Concrete Society (Technical
Report 30). Control is generally achieved by limiting the reactive alkali content of
the concrete mix to 3.0 kg/m3 of Na2O equivalent; this can be done by limiting the
cement content or by selecting Portland cement, ground-granulated blast-furnace
slag (GGBS) or PFA with low reactive alkali contents. For cement contents less than
500 kg/m3, a Portland cement with a certified average alkali content of less than
0.60% is deemed to satisfy.
Changes in the levels of alkalis in cement are therefore of interest. Table 6.4,
an abridged version of data by Corish,18 shows the trend since 1960. After peak-
ing in the early 1970s, levels have come down following positive efforts by the
cement industry; however, the most significant factor has been the elimination
of UK-produced high-alkali cement. Current research is aimed at reviewing the
present level of 3 kg/m3 in concrete.

Other factors

The review of cement properties15–18 covered many other factors, where change
is perhaps less significant. These include:
• Setting times: Initial setting time has fallen since 1960 by about 20%.
• Soundness: Of major concern 100 years ago, this is no longer an issue.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:17:14
06.qxd 15/11/2001 15:48 Page 110

110 Historic concrete

Table 6.4 Alkalis in Portland cement since 1960

Year Alkalis as Na2O equivalent (%)

Weighted mean Range

1960 (0.64) (0.4)–(1.2)


1965 (0.66) (0.5)–(1.2)
1970 0.69 0.5–1.2
1975 0.69 0.4–1.1
1980 0.61 0.4–1.1
1985 0.65 0.5–1.0
1990 0.62 0.41–0.83
1992 0.62 0.50–0.83

• Reactivity: Improvements in manufacturing, in the preparation of the raw mater-


ials and in the burning and cooling of the clinker have changed the mineral-
ogy of cements. Improvements in finished grinding have changed the particle
size distributions of cements. As a result, the activity of the cement has increased.
• Heat of hydration: The early rate of heat development has increased. Heat of
hydration is generally only an issue in the adiabatic conditions prevailing in
large sections, with the possibility of thermal cracking. Special low-heat cements
are available, if then considered necessary.
A relatively recent concern is long-term expansion due to delayed ettringite
formation (DEF) in concretes subjected to early heat curing.26 While research is
continuing, the procedure for avoiding DEF is well understood, and recommen-
dations are contained in the Department of Transport Specification for Highway
Works Part 5, and in the draft European concrete Standard pr-ENV 206. In
essence, limits are set on the temperature of the concrete, on the curing temper-
ature and on the rate of temperature increase.

The use of PFA and Table 6.1 gives details of relevant standards for manufactured cements. Both PFA
and slag have their own standards12,13 and the use of these materials, when com-
GGBS in concrete bined with Portland cement at the concrete mixer, has increased substantially since
the 1970s. In mix design terms, a prime concern for all concrete is durability, and
equivalence to a factory-made cement of the same composition is deemed satis-
factory if the mixer blends comply with the same grades achieved by concrete
made with the manufactured cement.23,25 Effective curing is more significant for
composite cements and mixer blends.
There is little doubt that the use of composite cements or mixer blends is advan-
tageous in many durability situations, and guidance is available.11,14 However, dur-
ability considerations have dominated developments to such an extent that they
are considered almost the only issue in deciding what concrete to use; there has
to be a balance with other performance requirements, such as speed of construc-
tion, and a perspective on this will emerge in due course.
In the context of this chapter — properties, production and performance of
concrete — the use of PFA and slag is significant. In relation to documented
changes, time is still short in a historical context. There is a great deal of labora-
tory evidence of the benefits, and it is known that the constituent materials can
be reliably produced to a common standard. There are also well-documented pro-
cedures for demonstrating the equivalence of mixer blends, in terms of sampling
(testing for strength and calculation of chemical composition). As yet, there is
little documented evidence in the public domain of a change in performance in
time in the manner recorded for Portland cement.15–18

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:17:14
06.qxd 15/11/2001 15:48 Page 111

Cement and concrete as materials 111

Table 6.5 The main general codes for structural concrete since 1907

Group Year Brief listing

1 1907 Report of the RIBA Joint Committee7


1911 Second report of the same RIBA Committee
1916 LCC Regulations
2 1933 The DSIR Code20
1938 LCC Regulations
1939 Building Industries National Council Code
3 1948 CP 114 Reinforced concrete
1957 CP 114 (with amendments in 1965, 1967 and 1973)
1959 CP 115 Prestressed concrete (with an amendment in 1973)
1965 CP 116 Precast concrete (with Addendum 1 in 1970)
4 1972 CP 110 Structural concrete
1985 BS 8110 replacing and updating CP 110

Concrete: impact Design standards


of changes in There have been four more or less distinct sets of rules for the general design
design standards and construction of structural concrete in the UK this century. These are listed in
and construction Table 6.5; this ignores other important codes for particular application (e.g. BS 5400
practices for bridges), and the recent developments with Eurocodes, but is sufficient for
illustrating the historical interaction between design rules and the development of
concrete as a material. A useful summary of this interaction is provided by Matthews.27
In the early part of this development — certainly up to 1948 and, arguably, as
far as 1972 — there was a close relationship between material requirements, design
rules and quality of construction. This is not surprising in 1907, since all concrete
was designed by specialist contractors, having their own patented systems, and one
of the main tasks of the RIBA Committee was to reconcile the different approaches
while taking due account of the laws of mechanics! However, this integration
was maintained and indeed emphasized in subsequent codes. Both the DSIR Code
and CP 114 (1948) rewarded extra care in supervision and testing by the use of
higher permissible stresses as a fraction of the required works cube strength. The
subsequent segregation of material, design and construction matters really began
in the late 1960s, and first showed itself in 1973, when a leaflet on ‘specification of
concrete for durability’ was included in a paper by Matthews;27 this has now devel-
oped to the point where design requirements23 are separate from concrete speci-
fications — effectively mirroring changes in the operating mode of the industry.
The next trend to note is the increases in strength over the years. The century
began with 1 : 2 : 4 concrete and 28-day cube strengths of 16–20 N/mm2. With CP
114 (1948) the specified works cube strength was just over 20 N/mm2 for 1 : 2 : 4 con-
crete, but the existence of richer mixes was recognized (up to 1 : 1 : 2) and the con-
cept of designed mixes foresaw specified works cube strengths to 50 N/mm2 or more.
Both CP 115 (1959) for prestressed concrete, and CP 116 (1965) for precast, raised
the level higher. By the time of the unified code — CP 110 in 1972 (Table 6.5) —
there had also been a change in the method of specification, with designed mixes
taking over firmly from what was now called prescribed mixes. This aspect of change
is summarized by Beeby and Hawes.28 In the last decade, there has been growing
interest in high-strength or high-performance concrete, with strengths in excess of
100 N/mm2 being achieved; in part, this is in response to durability concerns.
A close analysis of these changes shows a chicken-and-egg effect. The early speci-
fied concrete strengths were achievable with the cements available at that time;
permissible stresses were between 20% and 25% of the cube strength. Permissible
stresses increased as knowledge increased (up 20% by 1933) and new applications

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:17:14
06.qxd 15/11/2001 15:48 Page 112

112 Historic concrete

demanded higher-strength concrete and richer mixes. Broadly, cement strengths


responded to these new needs; the one exception to this was in 1948, when a
special clause was inserted into CP 114, indicating what to do if the specified
strengths could not be achieved!
Concrete technology played a significant part in these developments. In 1948, a
10% increase in working stresses was permitted when the concrete was vibrated. The
1950s saw a sharp increase in research on mix design and quality control. It is fas-
cinating to compare the proceedings of two conferences on mix design and quali-
ty control in 195429 and 1964.30 Here, it is easy to detect the death knell of the works
cube strength and its associated test regime; statistical control was the future.
The influence of research — or sometimes the lack of influence — on design stand-
ards and material development is also of interest. Two examples will suffice to make
the point. The discovery of the importance of water/cement ratio on all concrete
properties is generally attributed to Abrams24 as long ago as 1918. This factor was
slow to gain recognition, but it is now usually quoted in most codes in relation to
durability requirements; in terms of control of concrete quality, however, it has led
to little more than the development of the slump test for workability. On a more
positive note, Faber’s research on time-dependent effects in the 1920s followed by
Glanville’s work at BRE (BRE Technical Papers 11 and12, 1930) led to a new treat-
ment for column design in the DSIR Code of 1933, which negated the previous
modular ratio approach, and accepted that the failure load was the sum of the yield
of the steel and the capacity of the concrete; this in fact is ultimate load theory, not
recognized formally in code terms until 1957. Time-dependent effects were, of
course, integrated into the calculation of losses for prestressed concrete, but, even
now, in general design, their consideration is relegated effectively to special struc-
tures where movement is considered to be significant. Regrettably, design is largely
about strength, stiffness and stability, with insufficient attention to deformation and
strain; concrete structures in service have often suffered as a result.
The net effect of these interactive changes in concrete properties and design
standards is concrete structures which have become progressively lighter over the
years,28 with the ability to deform more, and yet constructed with materials which
in themselves are stronger and stiffer. There is therefore less margin for error,
in terms of effect on in-service performance.

Construction practice

Until the 1960s, responsibility for concrete making and placing rested with the
contractor. All codes and manuals emphasized the need for good workmanship
and skilled operatives. To quote Faber in 1949:19
‘It is, I think, difficult to exaggerate the importance of this Clause [on
supervision and workmanship], since in determining what is a reasonable
factor of safety the excellence of the detailing and of the actual execution,
plays an extremely important part’.
In the 1990s, procurement procedures are different, with the emphasis on speed
of construction, construction management and specialist subcontracting; most con-
crete arrives on site in a ready-mix truck or in the form of precast concrete com-
ponents. There are, therefore, more links in the chain which transposes basic
materials into concrete in the final structure. While each link has undoubtedly
become more efficient in itself, responsibility for the chain has become more
diffuse, and care is essential to ensure that nothing is lost in the process. In partic-
ular, the actual placing, compacting and curing of concrete (having the right cover)
is an installation sensitive operation, and an understanding by site operatives as
to the significance of these activities is just as important as expertise on the site

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:17:14
06.qxd 15/11/2001 15:48 Page 113

Cement and concrete as materials 113

itself. Feedback from performance in service has demonstrated variable end results,
and the good advice readily available31–33 is clearly not always followed. The human
element, especially in communications, is critical.
Changes in structural form (and fashions) and demands for new types of struc-
ture have had a major influence on construction practice, and a knock-on effect
on the required properties of fresh and hardened concrete. The era of shell roofs
and concrete trusses has been and gone. Dams were replaced by nuclear plants
and offshore platforms. Novel construction methods, bigger spans, thinner sec-
tions, etc., have all contributed to the introduction of pumping and large area
pours.
In the precasting arena, the century has seen the passing of several significant
phases, not without trauma. In response to demand for housing after the Second
World War, prefabricated reinforced concrete housing was introduced. A decade
or so later, there was the industrialized building period of the 1960s. Neither was
a technical or social success; both contributed to a growing awareness of the risks
of corrosion. Another contributed to the corrosion issue — again mainly in the
precast field — was the widespread use of calcium chloride as an accelerator which
began in the postwar period; its use was effectively banned in 1977, in concrete
containing embedded metal (it had been banned earlier for prestressed concrete).
Calcium chloride, as an admixture, is perhaps responsible for the general wari-
ness among specifiers for the use of admixtures for other purposes. There has been
a general increase in the use of admixtures over this century, but perhaps not as
much as there should have been. Admixtures are available both as accelerators
and retarders, for air entrainment and waterproofing and, especially, as plasticizers
and super-plasticizers.34

Durability and Lack of durability is perceived as a relatively new phenomenon, conceivably with
changes in the basic materials — and the introduction of new ones — being a con-
whole-life costing tributing factor. Concern over durability is not new; many technical papers dating
back to the 1920s and 1930s drew attention to it, and this concern is reflected in the
codes of that period. However, there are now very many more structures which have
been in service for significant periods, thus emphasizing both the scale and nature
of the problem.
With hindsight, what has been missing is the positive consideration of durabil-
ity as a major performance criterion, to be compared with the provision of strength,
stiffness and stability.35 The history of the industry has seen change by evolution,
punctuated by an occasional leap forward triggered either by a major technical
innovation (e.g. prestressed concrete) or changes in society’s needs (e.g. more hous-
ing in the middle part of the century, or the need to keep the motorway system
clear by using deicing salts). Great leaps forward all too often mean that practice
gets ahead of technology, but even the significance of slow evolutionary changes
is not always picked up — there is no machinery to do that properly.
Durability is certainly a materials issue, but feedback from performance in
service shows that design and construction are equally important. Changes in
materials cannot be considered in isolation; the issue of design standards
(including environmental loads) and of construction methods and quality are also
significant — the reason why they have been addressed in this chapter.
The plethora of papers in the technical literature over the last 20 years35 means
that we now have a better understanding of durability, often quantified in phys-
ical or performance terms. Application of that understanding in practice still lags
behind, however, particularly in terms of an integrated design, construction and
materials approach. This is slowly changing. The emphasis is still too much on
materials, but individual aggressive actions are being identified and addressed;
the basis for doing that is illustrated in Table 6.6.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:17:14
06.qxd 15/11/2001 15:48 Page 114

114 Historic concrete

Table 6.6 Types of aggressive action for which material specifications have been
developed

Aggressive action General approach Comments

Sulphate attack Quantify the action Specific material and mix proportions are
recommended in most codes for defined
ranges of sulphate concentration.
Alkali–silica Define ranges of The basic reaction and its possible effects are
reaction intensity for it now well understood. Recommendations to
minimize the risk of damage are published.
Freezing and Produce a specification Dealt with by choice of materials, mix
thawing for each range proportions and concrete grade. Air entrainment
for lower grades. Detail to exposure to moisture.
Abrasion Specifications to cover aggregate properties,
concrete grade and mix proportions, compaction
and curing, methods of finishing, etc.

The recipe approach in Table 6.6 is perhaps most appropriate for aggressive
actions which directly affect the concrete. Possibly the most serious durability issue
is corrosion of reinforcement, due either to carbonation of the concrete or to the
ingress of chloride ions from sources such as deicing salts or seawater. In the
laboratory, carbonation of concrete has been known about for over 60 years; in
practice, it became an issue only in the 1970s when the first wave of postwar con-
struction had been in service for 25–30 years. Real awareness of corrosion due to
chlorides was heightened by the use of deicing salts on motorways which began
around 1960. This is a major subject, with the historical aspect being summarized
by Somerville.35 It is in this area that the interplay between design, material and
construction aspects is most significant in influencing performance in service.
On a broader front, the technical performance of structures with time was first
addressed formally with a code of practice in 1950;36 this gave guidance on design
lives for different types of building and building component. It was largely ignored.
Its successor37 has received more attention, since the passage of time has high-
lighted the importance of durability in relation to function. Until the last decade,
the time factor was not recognized as part of the design process — levels of strength,
stiffness, stability and serviceability were provided by calculation and tacitly
assumed to obtain for the entire useful life (itself not normally given in the design
brief). Whole-life costing38 is beginning to change that. While still relatively new,
this has the merit of recognizing that adequate performance is required over a
significant period of time, and, in achieving that, should once again lead to the
integration of design, construction and material factors.

High-alumina HAC was first developed in France at the beginning of the century as a chem-
ically resistant cement — from a fusion of limestone and bauxite in a reverber-
cement (HAC) atory furnace. Early uses were in foundations and piling. However, its high early
concrete strength subsequently led to widespread use in precast concrete elements for build-
ings, particularly after the Second World War. At its peak, some 25 manufactur-
ers produced a wide range of X- and I-section beam and slab elements; it has been
estimated that nearly 90% of HAC use in the UK was for this application,
with roughly 17  106 m2 of floors and roofs in some 30,000 buildings.
Major collapses of school roofs in Camden in 1973 and Stepney in 197440 —
together with other local failures — led to major investigations by the Building
Research Station and to detailed guidance on the assessment of buildings
containing HAC concrete.41 A recent BRE Digest42 confirmed the validity of that

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:17:14
06.qxd 15/11/2001 15:48 Page 115

Cement and concrete as materials 115

earlier guidance, but expressed concern over the danger of chemical attack and
drew attention to the risk of reinforcement corrosion due to carbonation.
The technical concern with HAC concrete was an inevitable change in miner-
alogical composition which led to a reduction in strength (especially in warm, damp
conditions) of up to 50%. Even with design based on the reduced (converted)
strength, there was still a risk. In effect, HAC concrete is no longer used for struc-
tural purposes in buildings and foundations, although there has been recent inter-
est in using its high early strength for repair purposes, especially in roads and
pavements — with concrete having a w/c ratio less than 0.40 and a cement content
greater than 400 kg/m3.

Concluding As its title implies, the main objective of this chapter has been to record change —
remarks with the emphasis on the properties of cement and concrete in this century.
Hopefully, that has been done.
Recording the significance of that change on in-service performance was a
second objective. This has proven more difficult. What has emerged in this regard
is the interactive influence of changes in design and detailing, materials and tech-
nology, construction and procurement methods. Coupled with that is the influ-
ence of fashion on structural form and, perhaps more important, the impact of
customer demand. This has shown itself in three ways: volume of demand, which
comes in waves, leading, for example, to the industrialized building phase of the
1960s; increases in design standards, e.g. in terms of loads or quality; and changes
in use or function of ‘normal’ structures such as bridges or building, or for new
types of structure, e.g. offshore platforms or superstores. Overriding all these
changes is the shift in attitude to structural performance. Durability is now prob-
ably as significant as strength and stiffness, as we grapple with environmental issues
and whole-life costing. We now know much more about that, but we still require
the wisdom to apply that knowledge in practice.

References 1. Gooding, P., Halstead, P.E., The early history of cement in England. Proc. Third Int.
Symp. Chem. Cement, London, 1952.
2. Halstead, P.E., The early history of Portland cement. Paper read to the Newcomen
Society, Science Museum, London, 6 December 1961.
3. Francis, A.J., The Cement Industry 1796–1914: A History. David and Charles:
London, 1977.
4. Davis, A.C., A Hundred Years of Portland Cement. Concrete Publications Ltd.:
London, 1927.
5. Hamilton, S.B., A Note on the History of Reinforced Concrete in Buildings. National
Building Studies Special Report No. 24. HMSO: London, 1956.
6. The Engineering Standards Committee, British Standard Specification for Portland
Cement. Crosby Lockwood & Son: London, 1904.
7. Royal Institute of British Architects, Report of the Joint Committee on Reinforced
Concrete. J. R. Inst. Br. Archit., 1907, 3rd series, 14, No. 15.
8. Lea, F.M., The Chemistry of Cement and Concrete, 3rd edn. Edward Arnold:
London, 1983.
9. Neville, A.M., Properties of Concrete, 3rd edn. Pitman: London, 1982.
10. British Standards Institution, Specification for Portland Cement. BSI: Milton Keynes,
1991, BS 12.
11. Spooner, D.C., A Guide to the Properties and Selection of Cements Conforming to
British and European Standards. Interim Technical Note 13. British Cement Associ-
ation: Crowthorne, 1995.
12. British Standards Institution, Pulverised-Fuel Ash. Part 1. Specification for Pulverised-
Fuel Ash for Use with Portland Cement. BSI: Milton Keynes, 1993, BS 3892:
Part 1.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:17:14
06.qxd 15/11/2001 15:48 Page 116

116 Historic concrete

13. British Standards Institution, Specification for Ground Granulated Blast Furnace Slag
for Use with Portland Cement. BSI: Milton Keynes, 1992, BS 6699.
14. Concrete Society, The Use of ggbs and pfs in Concrete. Technical Report No. 40.
Concrete Society: Slough, 1991.
15. Corish, A.T., Jackson, P.J., Portland cement properties — past and present. Concrete,
1982, 16(7), 16–18.
16. Nixon, P.J., Changes in Portland Cement Properties and Their Effects on Concrete.
Information Paper IP3/86. Building Research Establishment: Garston, 1986.
17. Concrete Society, Changes in the Properties of Ordinary Portland Cement and Their
Effects on Concrete. Technical Report No. 29. Concrete Society: Slough, 1987.
18. Corish, P.T., Portland cement properties — updated. Concrete, 1994, 28(1), 25–28.
19. Faber, O., The structural use of normal reinforced concrete in buildings. Struct. Engr,
1949, April, 193–208.
20. Report of the Reinforced Concrete Structures Committee of the Building Research
Board. HMSO: London, 1933.
21. Nixon, P.J., Spooner, D.C., Concrete proof for British cement. Concrete, 1993, 27(5),
41–44.
22. Deacon, R.C., Dewar, J.D., Concrete durability — specifying more simply and surely
by strength. Concrete, 1982, 16(2), 19–21.
23. British Standards Institution, Structural Use of Concrete. Part 1. Code of Practice for
Design and Construction. BSI: Milton Keynes, 1985, BS 8110.
24. Abrams, D., Bulletin No. 1. Lewis Institute: Chicago, 1918.
25. British Standards Institution, Concrete. Part 1. Guide to Specifying Concrete. BSI: Mil-
ton Keynes, 1991, BS 5328.
26. Lawrence, C.D., Dalziel, J.A., Hobbs, D.W., Sulphate Attack Arising from Delayed
Ettringite Formation. Interim Technical Note 12. British Cement Association:
Crowthorne, 1990.
27. Matthews, D.D., The background to CP110: 1972 ‘The structural use of concrete’. Notes
on earlier British codes and regulations. Proc. Unified Code Symp., 27 September 1973.
The Institution of Structural Engineers and the Concrete Society: London, 1973.
28. Beeby, A.W., Hawes, F.L., Action and Reaction in Concrete Design, 1935–1985. C&CA
Reprint 3/86. British Cement Association: Crowthorne, 1986.
29. Cement and Concrete Association, Mix Design and Quality Control of Concrete. Proc.
Symp., 11–13 May 1954. British Cement Association: Crowthorne, 1954.
30. Cement and Concrete Association, Proc. Symp. Concrete Quality, November 1964.
British Cement Association: Crowthorne, 1964.
31. British Cement Association, Concrete on Site (formerly the ‘Man on the job’ series).
Publication 45.201. British Cement Association: Crowthorne, 1993.
32. Dewar, J.D., Anderson, R., Manual of Ready-mixed Concrete. Blackie & Son: Glasgow,
1988.
33. Murdock, L.J., Brook, K.M., Dewar, J.D., Concrete Materials and Practice, 6th edn.
Edward Arnold: London, 1991.
34. Hewlett, P.C. (ed.), Cement Admixtures — Uses and Application, 2nd edn. Longman
Scientific and Technical: London, 1988.
35. Somerville, G., The design life of concrete structures. Struct. Engr, 1986, 64A(2).
36. British Standards Institution, Code of Functional Requirements of Buildings. CP3:
Chapter IX: Durability. BSI: London, 1950.
37. British Standards Institution, Guide to: Durability of Buildings and Building Elements,
Products and Components. BSI: London, 1992, BS 7543.
38. Concrete Bridge Development Group, Whole life Costing of Concrete Bridges. Proc.
Sem., 25 April 1995. CBDG: Crowthorne, 1995.
39. Department for Education and Science. Report on the collapse of the roof of the assem-
bly hall of the Camden School for Girls. HMSO: London, 1973.
40. S.C.C. Bate. Report on the failure of roof beams at St John Cass’s School, stepney. BRE
CP 68/74.
41. Building Regulations Advisory Committee, Report by Sub-committee. (High Alumina
Cement Concrete) Document BRAC (75). HMSO: London, 1975.
42. Building Research Establishment, Assessment of Existing High Alumina Cement
Concrete Construction in the UK. BRE Digest 392. BRE: Garston, 1994.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:17:14
07.qxd 15/11/2001 15:49 Page 117

7 Concrete foundations and


substructures: a historical review

Mike Chrimes
Synopsis Concrete has been used extensively in foundations since the revival of interest in
its use two centuries ago. This paper outlines changes in its applications and use
for foundations and substructures over this period, paying particular attention to
early 20th century developments with reinforced concrete, and some post-Second
World War innovations such as diaphragm walls and the use of large diameter
bored piles in London Clay.

Introduction Mankind early recognised the value of sound foundations: the biblical references
are well known.1 Despite archaeological and written evidence of past foundation
techniques the history of foundation engineering, as with much of the more prac-
tical side of civil engineering, has yet to be written. Professor Skempton has
provided useful reviews of the development of soil mechanics,2,3 Heyman has
described the evolution of earth pressure theory and retaining wall design,4 and
Glossop described the development of specialist techniques,5,6 but little has been
written on the foundations themselves.7–9
For centuries timber was the chief means of providing adequate foundations for
structures where ground conditions were unlikely to sustain the superstructure,
being used for piles and also timber platforms and grillages. By the early 19th
century many other expedients had been employed to improve the ground and
spread the load of a structure — using fascine work (bundles of branches, etc.),
crushed chalk, and sand. Empirical rules were developed for masonry footings,
increasing the bearing surface with depth by using successively wider courses of
masonry. No doubt many mistakes were made, but there was a considerable body
of practical knowledge about the problems that could be encountered in ground
engineering, the need for adequate site investigation, and the expedients that
would be employed to secure good foundations.10 –13
From the end of the 18th century, the requirement for good foundation prac-
tice was of increasing importance. The large factory buildings of the Industrial
Revolution required foundations designed to sustain not only the static load of a
multi-storey building, but also the static and dynamic effects of heavy machinery.
Foundations were also needed for new warehouses in docks and alongside railways
and canals, and for larger and more heavily loaded bridges.
It is believed that the Albion Mill, designed by Samuel Wyatt and erected 1783–86
close to the south end of Blackfriars Bridge, was the earliest example of a raft foun-
dation, of masonry, covering the entire area of ground under a large building; it
can be taken as an example of best foundation practice of the time.14

Early use of The construction of Albion Mill took place at a time when British engineers were
taking increasing interest in the properties of mortars, most notably John Smeaton
concrete in whose researches into pozzolannas were published in 1791.15 Shortly afterwards
foundations Telford investigated the properties of Parker’s ‘Roman cement’ on behalf of the

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:49 Page 118

118 Historic concrete

British Fisheries Society (Paxton 15a). These investigations were particularly con-
cerned with identifying mortars that would set and be durable under water. In this
context English engineers had been using foreign natural pozzolanas, particularly
so called ‘Dutch’ trass for over a century. Italian pozzolana had been used by Charles
II’s engineers in their attempts to build a breakwater at Tangiers. Eighteenth cen-
tury lock and sluice work by engineers such as John Grundy specified the use of
trass for the mortar, and Colonel Henry Watson is known to have transported large
quantitites to Calcutta for use in constructing a dockyard there around 1780.
However, although concrete was used in the medieval period in Europe, in
Britain there is no evidence of its use being specified as a foundation material in
the modern sense before the early 19th century.16 Engineers’ knowledge of
its potential was probably limited to the description of the use of ‘beton’ for the
foundations at Toulon (1748) in Belidor’s Architecture hydraulique,17 which all the
leading engineers of the period: Smeaton, Rennie and Telford, are known to have
possessed (Figure 7.1a). French interest in the use of concrete in the early 19th
century, culminating in Vicat’s work at Souillac, had no immediate impact on
British practice.18–21
George Semple was the first British engineer who suggested the use of concrete
for bridge foundations.22 Lime ‘concrete’, described initially as ‘grouted gravel’
was first used by Sir Robert Smirke in 1817 for remedial work on the foundations
of the Penitentiary at Millbank, then under construction. Claims for its earlier use
at East India Docks appear unsubstantiated.23,24 Smirke used it again at Lancaster
Place (1820–23), Sir Robert Peel’s House, 4 Whitehall Gardens (1822–23), and,
most famously, to underpin the walls of the new Customs House, using a mix of
one measure of Dorking (quick) lime to seven or eight of Thames ballast.25,26
His success at Custom House was widely reported. The contractors for many
of his works were Samuel Baker and Sons, one of the leading contractors of the
early 19th century, and this must have helped spread knowledge of the technique.
Smirke used concrete again for the foundation of the British Museum (1833),
and the Oxford and Cambridge Club (1835). The contractors for the latter were
Grissell and Peto, who completed the foundations contracts for the Palace of
Westminster, where lime concrete was again used.27,28

Iron hoops in
brickwork

Figure 7.1 (a) Machine for


placing concrete underwater
at Toulon Harbour. Concrete
(b) Chimney built at Fulham
Gasworks on quicksand. (a) (b)

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:49 Page 119

Concrete foundations and substructures 119

The earliest published specification for the use of concrete was the work of
another architect, Robert Abraham, responsible for the new Bridewell prison (1830)
in Tothill Fields Westminster (23). This talks about 1 ft layers of concrete being
pitched from a height of 9 ft and ‘immediately puddled and trodden down by men
constantly employed in the works.’ Less information is available about the early
use of concrete foundations for industrial structures, but, for Fulham gasworks,
on ground described as quicksand, Samuel Clegg used concrete for the founda-
tions of a gasworks chimney in 1829 (Figure 7.1b). The concrete, of a 1 : 5 lime :
ballast mix was placed to form a block 8 ft deep and 20 ft wide, surmounted by
York stone landings and a brick footing 8 ft deep, offset. The chimney settled at
161⁄2 in, but there were no other problems. In the second half of the century con-
crete was used extensively in gasworks, particularly for gasholder tanks (see Gould),
and beneath retort houses as an alternative to rafts of brickwork and inverted
arches, although brickwork arches were used in combination with concrete rafts
at Beckton (1868) to reduce the weight.
Concrete was used by the contractor Hugh McIntosh for the foundations of the
London–Greenwich Railway viaduct (1835).23,24,29 In one area he had to excavate
to a depth of 18 ft through a peat layer. Concrete was laid by filling the founda-
tions with a layer of water perhaps 1 ft deep (305 mm) and tipping in it a mixture
of 1 : 3 : 6 lime : sand : gravel which was then turned by shovels in the water. This
was apparently developed from the methods used at East India Docks where Ralph
Walker had suggested gravel tipped into water for the foundations for the Wall.
It proved difficult in practice to mix the concrete adequately.
George Ledwell Taylor, architect for the government dockyards, attempted
to underpin the foundations of a storehouse in Chatham Dockyard in 183430,31
(Figure 7.2). The contractor, Ranger, had developed his own patent for concrete
using a 1 : 6 lime : gravel combination mixed with hot water, and compressed between
timber forms. Whereas at the Custom House the whole of the excavated area was
filled with concrete, at Chatham the concrete was contained within formwork
affixed at the width of the original brickwork (7 ft) (2.13 m).
For the stores at the Clarence Victualling Yard, Gosport, where the ground was
‘ooze or mud’ Taylor sank boxes 4 ft square (1.22 m) into the mud, excavated the
enclosed space, which was pumped dry to a sound base, and filled with concrete to
act as a foundation for piers carrying iron columns at 15 ft (4.57 m) centres. He also
used Ranger’s patents in Chatham and Woolwich dockyard for dock and wharf walls.

Figure 7.2 Underpinning the


Store House, Chatham
Dockyard, 1834.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:49 Page 120

120 Historic concrete

These proved vulnerable to water and frost action, and Pasley’s observations to
this effect may have delayed the use of concrete for retaining walls without any
masonry facing.24
As the 19th century progressed lime concrete gained increased acceptance as a
foundation material. Under 1879 bye-laws passed by the provisions of the Metro-
polis Management and Buildings Act, 1878, Section 16, the site of all buildings in
the London area, unless founded in gravel, sand, or ‘natural virgin soil’ had to
be covered in a layer of concrete at least 6 in thick (152 mm), with the foundations
of the walls at least 9 in thick (229 mm) projecting at least 4 in (101.6 mm) off each
side of the footings of the walls. The mix for the concrete was either 1 : 6 lime aggre-
gate or 1 : 8 of Portland cement aggregate.32,33
Experiments with Roman cement, with its flash set, were unsatisfactory.34 –35 Even
when it was successfully applied, as with the Great North Road at Archway, it was
a very tedious process, but as Portland cement gained acceptance the possibility
of stronger concrete for foundations grew.36 In London’s docks one can observe
engineers’ increasing confidence in concrete — from its use as a foundation, to a
backing material for retaining walls, and finally mass concrete walls.37–39 James
Walker described the use of a very weak concrete at West India Docks — essen-
tially a bed of gravel covered in a solution of lime, with a mix proportion of 1 : 20.
At St Katharine’s in 1826, Telford used a 12 in bed of concrete for the founda-
tions of the dock walls.26,40 West India Junction Dock (1850–53) had brick walls
on mass concrete foundations, Commercial Docks’ South Dock (1851–55) and
Millwall Docks (1865–68) had combinations of brick facing and concrete backing.
Royal Albert Dock (1876–80) had mass concrete walls (Figure 7.3).
Concrete was also successfully introduced as a cut-off for dam foundations.41–44
At Woodhead the initial dam design was a failure due to excessive leakage through
the fissured rock, and for his second design (1870–71) Bateman used concrete for
the cut-off, which, unlike the original clay puddle, was not susceptible to erosion.
This was followed by a concrete cut-off (1876) for the Upper Barden reservoir.
By then possibly the first modern concrete dam had been designed by British
engineers — the dam for Geelong waterworks, erected 1873–74.45–46 It was origi-
nally intended to build the dam from local stone, but in the absence of suitable
masonry a concrete mix of 1 part Portland cement to 61⁄2 parts of aggregate, placed
in rammed layers of 7 in (177.8 mm) in thickness, was used.

Raft foundations A plain concrete slab became a typical foundation in the last quarter of the 19th cen-
tury. In one spectacular example in 1895 the 3 ft thick concrete raft or pad for a
and concrete 185 ft diameter steel gasholder at Middlesborough began to subside shortly after con-
footings struction. The site was underlain with soft clay, and it had been hoped the overall
load spread by the raft of 1.2 t/ft2 would be sustainable. The problem of differential
settlement may have been aggravated by the ground on one side being compacted
by nearby railway lines. The solution was to drive boreholes on the higher side; the
weight of the tank forced out soft clay through the boreholes (James IGSE).46a
Concrete in combination with iron, and later steel, was regularly used for the
foundations of load bearing columns and other heavy loads by civil engineers by
this time (Figure 7.4).47 In the 1890s such practices would have been encouraged
by knowledge of the work of engineers and architects in Chicago, where steel beams
were used in combination with concrete for the first generation of skyscrapers.48
The Blackpool Tower was founded on four concrete blocks 34 ft square (10.36 m)
and 12 ft thick (3.66 m), in which 12 in  6 in (304 mm  152 mm) steel girders
were embedded. The High Court at Calcutta, built in the late 1860s, was founded
on a raft of concrete which contained two layers of hoop-iron interlaced to form
an 18 in square (457 mm) mesh, an expedient which is likely to have been used
elsewhere in the following decades.49

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:49 Page 121

Concrete foundations and substructures 121

(a) (b)

Figure 7.3 Dock wall profiles


at (a) St Katharine’s Dock,
1826; (b) West India Docks:
South Dock; (c) Millwall
Docks, 1865; (d) Royal Albert (c) (d)
Docks, 1876.

Figure 7.4 Foundation using


concrete and steel rails,
c. 1893.

According to a textbook of 1893 a raft would normally be at least 9 in (228 mm)


thick, but 12 in (304 mm) was more usual, and 18 in (457 mm) for larger structures
of the time.50 A modification would be to have the main slab 6 in (152.4 mm) thick,
but design the footings separately continued up above the raft, using the concrete
slab merely as a basement/ground floor level. In practice walls always had foot-
ings courses. Many of the early Hennebique structures followed a similar pattern,

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:49 Page 122

122 Historic concrete

but with the reinforced concrete columns carried on footings below the level of
the basement/ground floor slab (Figure 7.5a). Lion Chambers in Glasgow has foun-
dations of this type.51
When a reinforced concrete slab was used, 4 in (101.6 mm) notched boards would
first be placed to permit the accurate positioning of the (round bar) reinforcement,
with the cross bars being placed next, and wired to the main reinforcement at appro-
priate intervals, and with running boards supported above the reinforcement so it
was not disturbed or deformed (Figures 7.5b and 7.5c). Normally the concrete would
be placed in a relatively thin layer of about 1 in (25 mm) with the reinforcement
being lifted clear, then placed on this layer and the remainder of the concrete would
be placed, well tamped all the time. The floor would then be screeded over with
the exception of those areas where the beam and column supports would be taken
up into the rest of the building. When the slab incorporated pile caps, once the piles
had been driven a box would be made to surround the pile group, the reinforce-
ment positioned, and wooden forms placed for the beams connecting the pile caps,
with the beam reinforcement connected up at its intersection with the piles. These
would all then be concreted up to the underside of the main slab.52

Figure 7.5a Typical Mouchel-


Hennebique column founda-
tion, c. 1907.

Figure 7.5b Foundation slab


reinforcement held in
notched templates, c. 1910.

Figure 7.5c Section showing


pile caps and reinforced
concrete slab, c. 1910.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:49 Page 123

Concrete foundations and substructures 123

It should be noted that even with Hennebique buildings the foundations were
not exclusively of concrete. The concrete columns of the transit sheds for the Num-
ber 9 Dock, Manchester erected 1903–1905 were founded on brick footings,53 and
masonry was again used for the column foundations of Hudson & Kearns print-
ing works in south east London (1904) where the columns were supporting loads
of 2400 lb/in2 (16,547 kN/m2). The reinforced concrete bases to the columns were
5 ft 3 in2 (1.6 m2) and 15 in (381 mm) to 9 in (229 mm) in depth distributing a load
of 22,400 lb/ft2 (1072.5 kN/m2) on the masonry footings.54
The British Reinforced Concrete Engineering Company Limited (BRC), formed
in 1905, expanded rapidly, exploiting the patents of the ‘Paragon’ system of rein-
forcement, which made use of special forms of hoops, stirrups and sectionalized
helical wrapping to produce more rigid reinforcement.52 The firm was taken over
in 1908 and reorganized in 1911, when electrically cross-webbed steel wire fabric
was introduced on the Clinton systems. This proved a great success. Column foot-
ings were pyramidical in form, similar in appearance to those of Mouchel, but with
the base of the pyramid reinforced with BRC fabric, and a similar light fabric
reinforcement was used for ground floor slabs. Heavier reinforcement was avail-
able for road pavements.206
An interesting variation of reinforced concrete foundations was used in the Cot-
tancin system. Here reinforced brickwork walls were used in combination with
reinforced concrete slabs. Footings were built up of parallel reinforced brick walls
at appropriate centres supporting a thin reinforced concrete slab, to form a narrow
caisson. In other cases a series of reinforced brickwork cells were built to support
the basement/ground floor slabs. Knowledge of Cottancin’s work came from reports
to the 1900 Paris exhibition, although only a few Cottancin buildings were erected
in Britain, the best known being Sidwell Street Methodist Church, Exeter.55–57

Pile foundations The use of concrete pile foundations followed rapidly on the development of
the Hennebique system of reinforced concrete in the 1890s. Many ideas used by
the concrete systems had their origins in techniques developed over the previous
century, or earlier. Most obviously timber was replaced by precast reinforced
concrete piles. Even in-situ techniques can be traced to earlier ideas, such as sand
piles58 — a technique whereby timber piles were driven and withdrawn, and the
holes filled with compacted sand. The use of a screw was well established for cast
iron piles,59 and boring techniques were used to sink cast iron cylinders, which
were regularly filled with concrete for bridge foundations. Concrete filled cast iron
columns had been used since the first half of the 19th century.60,61

Precast concrete piles

The earliest systems used in Britain were the Hennebique, Coignet and Considère
systems.55,62–67 The patent (No. 2703) taken out by Philip Brannon in 1871 may never
have been used. Hennebique patented his reinforced concrete piles in 1896 (British
patent 10203 1897) (Figure 7.7a). The piles were of a square section, and a sim-
ilar reinforcement detailing to his columns, although wire cross ties were placed
close together at the head and toe. The tops of the rods were about 2 in (50 mm)
below the head of the pile, and the toe was similar to a timber pile’s cast iron shoe.
Rods were bent in at the foot to bear against the shoe. L.G. Mouchel developed
a screw and a hollow precast pile in 1900 (British patent 4548, amended 1907)
(Figure 7.7b), which was obviously lighter to transport and Mouchel believed had
the same bearing capacity.55,67 It was made with diaphragms containing forked
spacers and wire ties connecting the longitudinal rods. The diaphragms supported
tubular moulds forming the hollow core. Piles were initially cast in vertical forms

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:49 Page 124

124 Historic concrete

Figure 7.6a Driving 43 ft


long Hennebique piles at
Plymouth, 1900.

Figure 7.6b Vertical forms at


Southampton Cold Stores,
1904.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:49 Page 125

Concrete foundations and substructures 125

(a) (b) (c)

Figure 7.7 (a) Hennebique pile, 1897; (b) Mouchel hollow pile, 1907; (c) Considère pile, c. 1908.

within large racks. After about a week’s curing they were stored on the ground,
and normally not driven for a month or more (Figure 7.6b). Horizontal moulds
soon became more common, and other methods of curing such as steam curing
were developed before the First World War.65,68,69
An early British example of the use of Hennebique piles was the GWR grain
warehouse at Plymouth (1900), where 30 ft long (9.144 m) piles 14 in  14 in
(355.6 mm  355.6 mm) supported 12 in  12 in (304.8 mm  304.8 mm) columns,
the ground floor being 6 in (152.4 mm) of mass concrete on rubble fill (Figure 7.6a).67
At Dagenham Docks (1902) A.E. Williams developed a 14 in  14 in
(355.6 mm  355.6 mm) pile with rolled joist reinforcement, bent flat bars to
strengthen the sides parallel to the webs, and loops in the concrete at ‘frequent’
intervals. The point was formed by cutting away the top and forging the flange
to a point.69a

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:49 Page 126

126 Historic concrete

Figure 7.7d Coignet pile:


(i) Early, 1906; (ii) Final
version, 1910.

Figure 7.7e Standard BRC


piles.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:49 Page 127

Concrete foundations and substructures 127

The Armco (Armoured Concrete Construction Company) pile used angled irons
at each corner connected with frequent straps, and a special diaphragm of wire
bent in the form of a spring.62 The Considère system (Patent 14871, 1902)
(Figure 7.7c) used octagonal piles with longitudinal bars surrounded by spirals,
similar to their columns, and was first used for a jetty at Thames Haven in 1907.47,70
Coignet piles were the earliest to be widely used in Britain after the Hennebique
system. Originally patented in France (1894) where experimental piles were dri-
ven at Levallois-Peret in 1894, and then Asnières on a trial basis, they were not
used in Britain until 1906.52,55,71,71a Piles were circular with two flat faces reinforced
with 5⁄8 in diameter (15.875 mm) longitudinal bars, and 3⁄6 in (4.76 mm) rods for
hooping, hooked around the longitudinal bars (Figure 7.7d).
BRC produced 12 in (304.8 mm) square piles for loads up to 40 t, and 14 in
(355.6 mm) octagonal piles for loads up to 60 t, reinforced with longitudinal rods
wrapped with sectional helical reinforcement, with reinforcing hoops at the top
and intervals along the shaft (Figure 7.7e).

In-situ piles

Among the earliest in-situ piles were those of the Raymond Concrete Pile
Company.72–74 The earliest (US) patent was dated 1897, and A. Raymond followed
this with a succession of improvements which made his one of the leading piling
firms of the early 20th century (Figure 7.8). In its original form a tapered cast steel
form or core was used 20 ft long, (6.1 m), 18 in (457 mm) diameter at the butt and
6 in (152.4 mm) at the point; it was in two sections fitted with a dove tail joint and
secured by a key at the head of the pile. This core was fitted with a tapered shell
of thin steel or iron plate. Once the pile had been driven the key was withdrawn
and the core could then be collapsed and withdrawn and the shell filled with con-
crete. The shell was later reinforced with spiralled wire. Piles were also sunk using
a water jet system,74 a method used by Mouchel and other systems. The system
was known in the UK at least from 1901 when its trial in Chicago was reported in
Engineering News, and later described at a Municipal Engineers’ meeting in 1905,63
but its first British application is unknown. In the 1920s it was licensed to
J.W. Stewart.
The Simplex pile (Figure 7.9) invented by Frank Shuman of Philadelphia in
1903, was well known in Britain before the First World War.62,75–79 Early contracts
included Tranmere Bay Development (1905) (Figure 7.7). With Simplex piles the
(originally wrought iron) casing had a cast iron toe which was used to prevent the
head opening when driving. To avoid the loss of these toes an ‘alligator point’ was
developed. The jaws could be opened after the pile had been driven, and with-
drawn once the concrete was poured into position. Problems arose when earth pres-
sure caused the jaws to partially shut. Generally the pile casing would be withdrawn
2 ft (610 mm) and 3 ft (914 mm) of concrete inserted and rammed into position by
a 600 lb (272 kg) drop hammer. Where the diameter was sufficient expanded metal
3 in (76 mm) reinforcing mesh at 5⁄16 in (7.9 mm) could be used, and even precast piles
were inserted through the case. By the use of a heavy mandrel the concrete could,
where necessary, be compressed to form a bulb at the head. In some cases a pre-
cast concrete shoe was used on which the reinforcement rested. An outer casing
was used when the tip was in water-bearing strata. Problems could arise when dri-
ving a group of piles, as driving could interfere with adjacent piles. Although J.W.
Stewart had the original UK licence, Simplex soon set up their own offices.
Of the large number of in-situ systems developed on the continent at this time,
the most important was perhaps the invention of the Belgian Edgard Frankignoul
in 1909.80,81 With Franki piles a dry concrete plug was placed in the casing, and
driven by a drop hammer, the concrete pulling down the casing behind it. Once

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:49 Page 128

128 Historic concrete

Figure 7.8 Steel core


Raymond pile, c. 1901.

the desired depth had been reached the pile was concreted, the shell being grad-
ually withdrawn as successive layers of concrete were rammed in. This system gave
the pile a bulbous head and irregular profile which helped with bearing capacity
in granular soils for which it was ideally suited, although it could be a disadvan-
tage where negative skin friction was a factor. It was soon discovered the system
could be driven at an angle of 25°, and a reinforcement cage could be added. It
was claimed that these piles could be used with advantage where there was a nar-
row band of relatively good ground beneath the surface on which the pile could
bear. The Franki pile was exhibited at the Brussels exhibition at the end of 1910,82
where a pile was test loaded with 70 t, and in 1911 Frankignoul established an

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:49 Page 129

Concrete foundations and substructures 129

Figure 7.9 Simplex pile


system, c. 1906.

international firm, which became world famous. The UK subsidiary was incorpo-
rated in Liverpool in 1931. By 1957 it was the largest piling organisation in the
Commonwealth. In 25 years it installed 400,000 Franki piles, 300,000 after the
war. The first known contract was Codnor Reservoir, Derbyshire, followed by
Gravesend sewage works. In 1935 Franki piles were used for the prestigious job
at John Barnes store, Finchley Road. Their largest immediate post-war job was
Port Talbot steelworks, claimed to be the largest piling contract in the world
(initially 32,000 and eventually 90,000 piles) (Figure 7.10).

Case studies: Second Tobacco Warehouse, Bristol (1906)71a


prior to 1914
The early dominance of reinforced concrete piling L.G. Mouchel and the
Hennebique system was broken in 1906, by the contractor William Colvin and Sons
using Coignet piles for a large tobacco warehouse contract, involving over 600 piles,

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:49 Page 130

130 Historic concrete

Figure 7.10 Installing Franki


piles at Port Talbot after the
Second World War.

Figure 7.11 Raft foundations


at CWS Warehouse,
Newcastle-upon-Tyne, 1900.

in Bristol docks. Mouchel claimed that the Coignet system infringed Hennebique
patents, and took the case to court. Initially it was determined there was a patent
infringement but work was allowed to proceed on the basis that damages would
be paid equivalent to a licence fee if the decision was upheld on appeal.83,84 In
the event the Court of Appeal overturned the verdict,85 and their view was upheld
by the House of Lords, the verdict being that the concept of a reinforced concrete
pile had been anticipated by Brannon and others, and Hennebique/Mouchel could
not claim exclusivity for their patents.86 This may have helped other systems gain
acceptance after this date.

CWS Warehouse, Newcastle-upon-Tyne (1900) (Figure 7.11)

Reinforced slabs and footings were also dominated by L.G. Mouchel and the
Hennebique system prior to c. 1906, the first example of a foundation slab being
at Newcastle-upon-Tyne c. 1900 (see Chapters 4 and 5 by Bussell). Plain concrete
slabs, occasionally combined with timber piles, remained the dominant founda-
tion form, with steel grillages being used after steel frames came in from 1904
onwards; to that extent reinforced concrete was an atypical material.

Extension work at Victoria Station87 (Figure 7.12)

Works to the London–Brighton and South Coast Railway side of the Station
involved extending the Grosvenor Hotel, and a considerable increase in the area
of the station, obtained in part by covering in an area of the Grosvenor Canal. With

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:49 Page 131

Concrete foundations and substructures 131

Figure 7.12 Foundations at


Victoria Station, 1900.

Figure 7.13 Ritz Hotel


foundation grillage, 1904.

the proximity to the river ground was very varied, and the underlying London
clay dipped from around 20 ft (6.1 m) below the surface on the Buckingham Palace
Road side to 40 ft (12.2 m) near the South Eastern Railway station. Overlying strata
included sand and gravel, peat and silt. Generally the foundations were carried
down to the London clay. Where the clay depth did not exceed 35 ft (10.7 m)
7 ft 6 in2 (2.285 m2) concrete (1 : 6) column bases were founded on the clay, beneath
that timber piles of 12 in  12 in (304 mm  304 mm) pitch pile were driven at
3 ft (914 mm) centres, supporting a 1 : 8 concrete slab 7 ft thick (2.134 m). The
average load per pile was about 141⁄2 t (44 kN/m2). Beneath the Hotel annexe and
Eccleston and Elizabeth bridges voids with brick arches and inverts were left,
between 5 ft (1.52 m) and 10 ft (3.04 m) wide.

War Office Buildings, Whitehall88

The contract for the excavation and substructure comprising a 5 ft (1.52 m) thick
plain concrete raft, and plain concrete retaining walls for a two storey basement,
was completed in 1904. The wall footings’ courses were above basement floor level.

Ritz Hotel89 (Figure 7.13)

The Ritz Hotel was probably the first steel framed structure built in London. It
comprised a seven storey steel frame above ground floor level with a lower ground
and basement floor below. The structure was supported on 118 columns up to
ground floor level, founded on cast iron base plates resting on a steel grillage, above
18 in (457 mm) thick concrete footings, about 13 ft (3.96 m) square, sunk in pits
into clay below basement floor level. Part of the frame was supported on rocker
bearings.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:49 Page 132

132 Historic concrete

Figure 7.14 Section through


North-Eastern Railway Goods
Station, Newcastle-upon-Tyne,
1906.

The Institution of Civil Engineers90

The present ICE building erected 1910–13 is another early steel frame structure,
with load-bearing external masonry walls on one elevation. The foundations com-
prise a 4 ft (1.22 m) thick mass concrete raft of 1 : 6 Portland cement : aggregate mix,
containing an asphalt waterproofing layer, with some timber piling at the south west
corner. The frame is supported on steel grillages contained within the concrete raft.

North-Eastern Railway Goods Station and Warehouse,


Newcastle-upon-Tyne91 (Figure 7.14)

This structure, completed in August 1906, was designed as a reinforced concrete


frame, with overall dimensions of the building 430 ft (131 m) long, 178 ft 4 in
(54.35 m) wide, and 83 ft 4 in (25.4 m) high, divided into four floor levels includ-
ing a basement floor which served as the low level goods station. Foundations rested
on boulder clay, the safe bearing capacity of which was estimated at 5 t/ft2
(536 kN/m2). The structure was designed for warehouse loadings, ten rail tracks
of goods traffic, and turntables, cranes, etc. To first floor level the framework com-
prised 70 wall and interior columns in five rows, 33 ft (10 m) apart longitudinally,
and with 25 spans of 37 ft 2 in (11.33 m) and two of 52 ft (15.85 m) centre to cen-
tre between the rows, two of these rows were purely to support the ground floor
goods station. At basement level there were 30 additional columns. The wall
columns were supported on massive concrete retaining walls, which formed the
walls of the lower level station. The footings for the central row of columns were
15 ft 6 in (4.72 m) square, those on each side 7 ft (2.13 m) square, and the outer
bases 14 ft (4.27 m) square, loads on the columns varying between 224 (kN/m2)
and 1105 t. The latter columns were 1600 in in area (40 in  40 in) (1016 mm2).
The footings for these were reinforced with horizontal bars laid in rows at right
angles to each other, connected by vertical loops for resisting shear stresses.

Rowntrees Works, York: Melangeur Block (1907)92 (Figure 7.15)

This block, 105 ft (32 m)  76 ft 7 in, (23.34 m) was 93 ft 6 in (28.5 m) high, built
on six floors including the basement level. The ground being compressible, it was
decided to found the building on a reinforced concrete slab 12 in (304.8 mm) thick
which projected 10 ft (3.048 m) in every direction beyond the outer walls, distrib-
uting the load over an area of 9700 ft2 (900 m2). The slab was reinforced with round
steel bars tied longitudinally and transversely near the top and bottom surfaces.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:50 Page 133

Concrete foundations and substructures 133

Figure 7.15 Section through


Melangeur Block, Rowntrees,
York, 1906–1907.

Columns were supported on bases resting upon the foundation slab. Bases sup-
porting the outer wall columns were 12 ft (3.66 m) square and those supporting
the internal columns were 15 ft 6 in (4.72 m) square, all rising to 2 ft 6 in (767 mm)
above the slab. The reinforcement comprised round steel bars near the base to
resist tension, with vertical stirrups of strip steel to resist shear. Bars forming the
vertical reinforcement of the columns were carried into the base, and connections
were made between the slab and column bases using reinforcement. Earth was
spread over the foundation slab and rammed before a plain concrete slab 9 in
(228.6 mm) deep was placed, to form the basement floor.

J.C. & J. Field’s Factory, Rainham, Essex (1906)93 (Figure 7.16)

These factory buildings had a brick superstructure but, presumably in view of


ground conditions comprising 3 ft (914 mm) of hard silt resting on a peat bed 25 ft
(7.62 m) thick, it was decided to use reinforced concrete for the foundations, and
attempt to reduce the pressure on the ground by distributing the load. The Coignet
system was employed, the second instance of its use in Britain. Foundations were
required for a building 271 ft (8.23 m)  47 ft 6 in (14.48 m), and a boiler house
92 ft 6 in  72 ft 6 in (28.19 m  22.1 m) containing three Galloway boilers.
The boiler house foundations comprised reinforced concrete footings 4 ft 8 in
(1.422 m) wide and 3 in (76 mm) thick beneath two walls, and 43⁄4 in (121 mm) thick
along the third wall, distributing the weight of the building at the rate of 3 cwt/ft2
(16,088 N/m2) over the ground. On these footings a beam 10 in (254 mm) wide and
12 in (304.8 mm) deep was formed to support the brick walls, and reinforced
concrete columns supporting the roof. Reinforcing bars were placed in the lower
half of the slab, connected to bars in the beams. The foundation beam was

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:50 Page 134

134 Historic concrete

Figure 7.16 Section through


foundation beam for boiler
foundations, Rainham
(Coignet system), 1906.

supplemented by two piles in one place and two transverse footings in another
where the loads on the columns were excessive.
The boiler foundations comprised a slab c. 40 ft (12 m) square, to reduce the
load to 3 cwt/ft2 (16,088 N/m2). Care was taken not to disturb the silt strata, and a
crushed brick layer was placed on this followed by 5 in (127 mm) of concrete with
1
⁄4 in (6.35 mm) reinforcing rods placed in the lower third near the external wall,
but in the upper third beneath the boiler. On this slab were placed six parallel
reinforced concrete beams 10 in (254 mm) thick, with their reinforcement looped
into the foundation slab, and these beams supported the seating for the boilers.
The footings for the shed comprised a reinforced concrete slab 4 ft 2 in (1.27 m)
wide and 4 in (102 mm) deep supporting a reinforced concrete beam 10 in
(254 mm) wide and 12 in (304.8 mm) deep. Reinforcing bars were between 1⁄4 in
(6.35 mm) and 1 in (25.4 mm) diameter, with the beam reinforcement looped into
the foundation reinforcement every 6 in (152.4 mm).

Brooklands Motor Track: bridge over the River Wey94

Constructed on the Hennebique system the bridge was 200 ft (61 m) long and 100 ft
(30.5 m) wide, designed for a uniform distributed load of 112 lb/ft2 (5.4 kN/m2) and
a point load of 2 t/ft2 (214.6 kN/m2). The foundations included forty-two 14 in
(355.6 mm) square piles, which were connected continuously with the beams and
columns of the superstructure.

Tranmere Bay Development Works (1905–1906)79 (Figure 7.17)

The engineering workshops of this scheme were largely founded on made ground.
The main erecting and machine shop was 1035 ft (315.5 m) long, with overhead
cranes 60 ft (18.29 m) above floor level, travelling along a 74 ft (22.56 m) span. The

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:50 Page 135

Concrete foundations and substructures 135

Figure 7.17 Photographs of


piling works at Tranmere Bay,
1906.

concrete blocks supporting the column-bases were carried on groups of Simplex


piles — from 4 to 12 for each column. The tubes used for the casing were 40 ft
(12.2 m) long, 16 in (406 mm) in outside diameter, of 1⁄2 in (12.7 mm) thick metal
welded together from three lengths of tubing. This had to be imported from
Germany as the English pattern with one joint and a riveted cover plate was unsat-
isfactory. The casing had an alligator point and was driven with a cast steel cap in
the top of which was an elm or hardwood piece to bear the blows of a 30 cwt
(152.4 kg) iron monkey, falling 6–8 ft (1.8–2.4 m). The casing was driven to a set
of 1 in (25.4 mm) in 4 blows, penetrating 20 ft (6.1 m) of made ground and 10 ft
(3.05 m) of the original strata. The casing was withdrawn using tackle with a capac-
ity of 50 t (500 kN). Concrete (1 : 6 cement : aggregate mix) was tipped sufficient
for 2 ft (609.6 mm) lengths of pile. The maximum load was 20 t/ft2 (2146 kN/m2)
of pile head. Piles were driven at 3 ft (914.4 mm) centres, and in good ground there
did not appear to be disturbance of newly driven piles by driving adjacent piles.
Where a layer of spongy blue clay 6–10 ft (1.8–3 m) thick was present, however,
piles were found to be deformed when exposed, and reduced to 11 in (279.4 mm)
in diameter.95 Here concrete footings were taken down below the layer of clay. Nine
piles were driven a day per machine. Because of the large number of piles used
and the cost of equipment, the foundation cost 15% more than a timber piled
foundation, but it was felt the foundation would be more secure.

Reinforced Retaining walls provide one of the clearest examples of the liberating effect of rein-
forced concrete on civil engineering design. Mass concrete retaining walls of the
concrete type seen in Figures 7.3a–d were designed essentially as gravity structures, whereas
retaining walls96 reinforcement made possible the design of lighter structures of adequate strength

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:50 Page 136

136 Historic concrete

of two main types — cantilever and counterfort, offering a considerable saving in


cost. Methods of design were developed before the First World War.97–99 Numer-
ous examples of retaining walls of these types were erected in the United States
from around 1904 onwards.100–103 In Europe Hennebique again led the way. The
retaining wall erected for the Paris exhibition of 1900 at the Quai Debilly, near
the Trocadero in Paris, is perhaps the best known early example.104,105 Here the
wall was formed of slabs with ribs at the back, both being connected to the foun-
dation slab, which in this case had a rib at the toes at the front (Figure 7.18a).
Other systems aside from Hennebique were used in Britain before the First
World War. At Mappin and Webb’s famous Queen Victoria site in the City (1911)
a complicated construction was installed comprising a base, vertical slab, curved
wall and top slab, reinforced by ‘indented steel bars’ (Figure 7.18b).106,107 A more
normal cantilever wall was built to the same system for the Royal Insurance Com-
pany offices, Piccadilly (Figure 7.18c).106,108 Their office in Lombard Street was
essentially a steel framed building (Figures 7.18d–f ), but the Considère system of
reinforced concrete was used in the foundations which included a cantilever retain-
ing wall with an overall depth of 26 ft 6 in (8.08 m) to the base of its toes, with the
wall varying in thickness between 21 and 10 in (533–254 mm).109 The toe incor-
porated a reinforced concrete beam distributing the loads from the stanchions
supporting the superstructure and was given increased depth to form a reinforced
concrete foundation for stanchions at 14 ft (4.27 m) centres. Comparison of the
thickness of the reinforced concrete with a brick retaining wall suggested the lat-
ter would need a maximum thickness of 13 ft (3.96 m) to give the same stability.
Reinforcement comprised 3⁄4 in (19 mm) vertical rods with horizontal stirrups 1⁄4 in
(6.35 mm) diameter to resist shear.

Figure 7.18a Hennebique


retaining wall at Quai Debilly,
Paris, 1900.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:50 Page 137

Concrete foundations and substructures 137

The Expanded Metal system was used for retaining walls with buttresses or
counterforts.110,111 The walls at Salford public baths were 8 ft 9 in (2.64 m) high,
made of concrete 6 in (152 mm) thick reinforced with No. 10 expanded steel mesh,
with a horizontal bar at the top of the wall; the buttresses were 9 in (229 mm) thick
and 3 ft 6 in (1.069 m) deep at the base of the wall.106 A more substantial wall at
Guildford was 17 ft (5.18 m) high, 6 in (152 mm) thick at the top, 9 in (229 mm) at
the base, with buttresses up to 9 ft (2.74 m) deep at the base, 9 in (229 mm) thick
and at 5 ft 6 in (1.68 m) centres (Figure 7.18f). The wall reinforcement was No. 10
expanded steel mesh, and the base 15 in (381 mm) thick with No. 30 expanded

Figure 7.18b Retaining wall


at Mappin and Webb’s, Queen
Victoria Street (indented bar
system), 1911.

Figure 7.18c Retaining wall,


Royal Insurance offices,
Piccadilly (indented bar
system), 1907.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:50 Page 138

138 Historic concrete

(d)

(d)

Figure 7.18d, e Foundations,


Royal Insurance offices,
Lombard Street (Considère
system), 1911.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:50 Page 139

Concrete foundations and substructures 139

Figure 7.18f Retaining


wall, Guildford, c. 1911.

Figure 7.18g Retaining wall at GPO extension, King Edward Street (Hennebique), 1907–10.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:50 Page 140

140 Historic concrete

Figure 7.18h Retaining wall


at the RAC building, 1910.

Figure 7.18i Retaining wall,


General Accident Assurance
building, Aldwych, 1909.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:50 Page 141

Concrete foundations and substructures 141

steel mesh. This wall was piled on timber piles beneath the buttresses. Reinforc-
ing bars in walls of this type could vary in size from 1⁄2 in (12.7 mm) square at the
top of the wall, to 11⁄16 in (17.5 mm) square at the base and 11⁄4 in (31.7 mm) square
in the back of the counterfort.106
Retaining walls were frequently integrated with the main structure. In the case
of the GPO extension (1907–10) retaining walls were 7 in (177.8 mm) thick at the
top, and 8 in (203.2 mm) at the bottom, with an average height of 30 ft (9.144 m)
above the footings (Figure 7.18g). Reinforced with 7⁄16 in (11 mm) diameter
horizontal rods at 4–10 in (101.6–254 mm) centres on the outside of the wall, and
5
⁄16 in (7.9 mm) diameter at 8–24 in (203–610 mm) centres on the outside, and verti-
cal 5⁄16 in (7.9 mm) diameter rods at 8 in (203.2 mm) centres with stirrups every foot
in height, the counterforts, at 6 ft (1.83 m) intervals were 7 in by 14 in112
(177.8 mm  355.6 mm) with 1 in (25.4 mm) diameter reinforcement. The walls
were stayed by struts and arches at the ground floor and lower ground floor lev-
els. At Newcastle Goods stations the majority of the retaining walls were plain con-
crete, but on the eastern side the ground was unstable, and the retaining wall
comprised stanchions built up from the basement supporting a thin reinforced
concrete wall, and supported by beams at ground floor level and struts in the base-
ment floor between the wall columns and those of the main building structure.
Indented steel bars were used in the reinforced concrete cantilever retaining
walls for the 36 ft (10.97 m) high wall at the rear of the Royal Automobile Club,
Pall Mall (Figure 7.18h).113 The wall varied in thickness from 3 ft 9 in at its base
to 14 in (355.6 mm) at ground floor level, with the toe from 15 to 22 ft (4.57–6.7 m)
wide and 3 ft to 18 in (914.4 –457.2 mm) thick. The reinforcement for the wall com-
prised 11⁄4 in (31.75 mm) and 3⁄4 in (19 mm) indented steel bars, and the concrete
was composed of 1 : 2 : 3 cement : sand : crushed gravel, the latter passing a 3⁄4 in
(19 mm) mesh. The earth was supported by timbering for six weeks before the
formwork was removed.
The reinforced concrete for the General Accident Assurance Building, Aldwych
was provided by the British Fireproof Construction Company, and included a
conventional cantilever retaining wall (Figure 7.18i).114 In the 1930s BRC rec-
ommended, on economic grounds, cantilever construction for walls supporting
up to 15 ft (4.57 m) of earth, and the use of counterforts above this height.115

Piling between the wars: research

Despite early optimism concerning the use of reinforced concrete for piles, prob-
lems soon became evident. A particular problem contractors encountered arose
when driving precast piles through a hard strata to a set in firm ground below.116
Piles were damaged and even failed under these circumstances, and the Federa-
tion of Civil Engineering Contractors asked the Building Research Station to inves-
tigate. The results are reported in the reports of the Building Research Station
in the 1930s, and summarized in papers of 1935 and 1938.117,118
Research revealed most failures were due to excessive compressive stress, with
head failures most frequent, a major factor being compression of the head pack-
ing and unevenness in placing the packing in the head. Failures lower down the
pile were normally the result of hard driving through dense stratum, although
not necessarily at the toe, which might derive lateral support from the ground.
Tensile cracks were found to develop when a hard stratum was encountered rel-
atively close to the surface. Longitudinal reinforcement provided it was sufficient
for the safe handling and transport of the pile, was found likely to be adequate
for safe driving. On the other hand lateral reinforcement, particularly at the head
and toe was important, and it was recommended that for a distance from the
head and toe 21⁄2 –3 times the external diameter of the pile the volume of the

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:50 Page 142

142 Historic concrete

lateral reinforcement should not be less than 1% of the gross volume of the cor-
responding length of pile. External bands at the head were found to be beneficial.
Wet curing of piles had a great influence on impact strength, and for Portland
cement concrete it was recommended it should be of at least 14 days duration
before driving. It was found that the impact strength of concrete was 50–80% of
the cube compressive strength, and a maximum stress of 50% of cube compres-
sion strength was recommended for a driving factor of safety of unity. In view of
the low factor of safety in pile driving it was stressed that care needed to be taken
with driving, and best results were obtained with a heavy hammer and a head-
cushion of lowest resistance, such as of rubber or asbestos fibre.
Considerable effort was also expended on an exposition of the mathematical
theory of stress waves induced in piles during driving. Wave action had been iden-
tified in 1931,119 but these were the first published solutions of the wave equation
applied to pile driving. The full potential of this approach was not realized until
the advent of computers in the late 1950s,120 but is well-recognized today.121
From 1936 the Institution of Civil Engineers became involved in the research
programme, which continued under the auspices of the Joint Committee on Piled
Foundations after the war, when a considerable amount of data was gathered
on pile loading tests, and work published on impact testing of concrete, and pile
group behaviour.122–126

Piling between the wars: practice

In the 1920s British Steel Piling developed their own system of in-situ concrete
piles, known as ‘Vibro’ piles.127 A steel tube, usually 16 in (406.4 mm) in diame-
ter, with the lower rim slightly thickened for tamping the concrete, was placed on
a conical shoe, and driven to a firm base. A dry mix of concrete was placed in the
tube which was then withdrawn by upward blows of the hammer in 11⁄2 in (38 mm)
stages. During withdrawal the tube was subjected to 80 blows a minute by the ham-
mer, tamping the concrete and forcing it down and out against the ground. Rein-
forcement could be placed when necessary before commencing concreting. The
completed piles were normally at least 17 in (431.8 mm) in diameter with a work-
ing load of 50 t (60 t with reinforcement). Over 700 of these piles were used at St
James’ Park Underground Station in the 1920s.
The system was extensively employed on the Royal Docks Approaches improve-
ment scheme which began in 1929, and involved piled foundations for the
Canning Town and Silvertown viaducts128 (Figure 7.19). Canning Town viaduct
was 4000 ft (1220 m) long, and supported on concrete columns resting on 3500
vibro piles. The piles were driven through a layer of ground into the underlying
clay. The majority of the piles were 40 ft (12.129 m) in length, 17 in (431.8 mm) in
diameter, with longitudinal reinforcement of 63⁄4 in (19 mm) diameter steel bars with
spiral binding. They were designed for a working load of 40 t, with a safety
factor of 3. Some of the piles were test loaded, a load of 110 t on a single pile pro-
ducing a settlement of a small fraction of an inch. Some 14 in (355.6 mm) diame-
ter piles were used, with 5⁄8 in (15.88 mm) diameter bar reinforcement, designed
for a working load of 30 t. At Silvertown the average length of the piles was 35 ft
(10.67 m) but some were driven 65 ft (19.8 m) using an 85 ft (26 m) high steel frame.
The working load here was 50 t. Pile cap details are shown in Figure 7.19. BSP
also marketed the Prestcore system, using precast sections grouted together under
pressure in a bored excavation (Figure 7.20).
The (Francois) Cementation Company developed in-situ piling methods
based on the technology of the ‘cementation’ process.129,129a For the ‘Express’ pile
an 18 in (457.2 mm) diameter steel tube with a solid point was driven to a set, a
reinforcement cage placed in position, and then concrete placed and tamped out

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:50 Page 143

Concrete foundations and substructures 143

Figure 7.19 Pile cap details,


Royal Docks Approaches,
1929.

Figure 7.20 Prestcore Pile


Test.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:50 Page 144

144 Historic concrete

below the end of the tube which was gradually removed. These were first used
for a garage foundation in Durham in 1929. In the case of the ‘Bored’ pile ground
was excavated by boring techniques as the tube was sunk, rather than being
displaced. Reinforcement was placed as necessary, and a perforated injection tube.
Ballast was inserted in small quantities and tamped by a drop hammer. Cement
grout was then introduced through the tube under pressure. Bored ‘Francois’ piles
were excavated in a similar manner, but consolidated concrete was placed rather
than using aggregate and the cementation process. The system was first used in
the UK in 1926 at Retford Gasworks. The firm also used ‘Tapered shell’ piles on
a similar basis to the Raymond system. One thousand seven hundred piles of this
type were used for Unilever House in 1930.
West’s Piling Company marketed the Rotinoff system.129,130 Precast concrete
shells in 3 ft (914.4 mm) lengths, with steel bands at the joints, were threaded on
to a steel mandrel, and a shoe fitted at the end. As the shoe was fitted to the man-
drel it took most of the driving stresses. Holes in the shell units were available for
placing reinforcement. When the desired set had been achieved the mandrel was
withdrawn and the pile could be inspected before concreting. Pile diameters of
141⁄8 –24 in (356–609.6 mm) were used.
The ‘Screwcrete’ system was developed by Braithwaite as a concrete alternative to
traditional cast iron screw piles.130–132 A screw or helix made of reinforced concrete,
steel or cast iron, was driven by a steel mandrel rotated by an electric capstan; a
steel or concrete casing was attached to the helix, and could be rotated with it.
When the required depth had been reached the mandrel was withdrawn and the
pile constructed. Thrust bearings connected to the mandrel were required for the
reinforced concrete casing. The screw could be 6 ft (1.83 m) or more in diameter.
For 12–18 in (304.8–457.2 mm) diameter piles the pile was normally precast and
driven by thrust bearings with just a central hole for the mandrel which was con-
creted after the mandrel was removed. Other inter-war developments included
pressure piles, using augers and compressed air, and the ‘Hawcube’ piles made
up of precast units.129,130

Post-war developments: bored piles

In the immediate post-war period research was carried out into the application
of short-bored piles to house foundations in London Clay.133 Further research fol-
lowed, and as the 1950s progressed increasing use was made of bored concrete
piles for more substantial structures in London Clay.134–136
By 1959 Skempton was able to report on results of experience with bored piles at
ten sites in London Clay covering the years 1950–59, and produce conclusions regard-
ing the end bearing and shaft bearing capabilities of such piles, and thus their ulti-
mate bearing capacity, which could be used to establish the working load for a single
pile.137 Most of these examples were relatively small diameter (10–18 in) (254 –
457.2 mm), and to modest depths, but there was already increasing demand for
suitable foundations for tall buildings in the London area, and large diameter bored
piles excavated to a considerable depth in London Clay was one solution.
The use of large diameter ‘pier’ foundations can be traced back to the Chicago
school of foundations developed before the First World War, and the Gow method
of large diameter in-situ pile construction was well known.138 In America these foun-
dations were often taken down to bedrock, which was not an option in London,
and the design of large diameter bored piles in London Clay attracted consider-
able interest in the engineering community.
By 1954 the Benoto system of excavation by hammer grabs had been used in
the Midlands for shaft sinking,139 and in 1956 large diameter piles were installed
by Sir Robert McAlpine at Bradwell.140 The best known early example of large

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:50 Page 145

Concrete foundations and substructures 145

diameter bored piles is that of the Shell Centre on the South Bank development.
Here 4 ft 6 in (1.37 m) diameter bored piles, underreamed at the base to 9 ft
(2.74 m) diameter, were used.141,142 Soon after, in June 1959, work began on 3 ft
(914.4 mm) diameter bored pile foundations, 65 ft (19.8 m) in depth for the 38
storey Millbank Tower.143,144 These piles were not underreamed, and most sub-
sequent bored pile foundations followed the same pattern.
Despite their increasing use engineers still felt concerned about the lack of knowl-
edge about the design, installation, and behaviour of these foundations. Early expe-
riences were shared at a 1961 conference, and an approach was made to the
Institution of Civil Engineers Piling Committee about various industry concerns.145
The BRE were commissioned to carry out research, and other investigations were
carried out by the contractors.146 In 1966 a further conference was held on large
bored piles.147 Much of the experience described was in London Clay, but one
paper discussed the use of large bored piles in rock. These papers show practical
problems involved, particularly when ground conditions were not ideal. Plant was
not always reliable, and when there was a delay in placing concrete after the pile
had been excavated there were concerns over what kind of pile would result in
the presence of water bearing strata or other unfavourable ground conditions.

Prestressed concrete piles148

The potential of prestressed concrete piles was dramatically revealed when


Freyssinet made use of them for his remedial works at Le Havre Harbour149,150
in the 1930s (Figure 7.21). To have driven piles by ordinary methods would have
been difficult without further risk to the foundations of the existing quay struc-
ture, already suffering alarming differential settlement, and Freyssinet decided to
sink 100 ft long prestressed concrete piles using hydraulic jacks. Instead of 2000
conventional reinforced concrete piles, carrying 75 t each, only 475 prestressed
piles were required, capable of carrying 300 t. The piles were hollow cylinders of
24 in (610 mm) external and 15 in (380 mm) in internal diameter, reinforced lon-
gitudinally by 8 wires of 8 mm diameter hard grade steel, with transverse loops
of 6 mm diameter hard grade steel, the concrete mix being 1 : 1.33 : 2.66 using
sulphate resisting Portland cement. The concrete was vibrated, hydraulic pressure
of 285 lb/in2 (1965 kN/m2) applied, and the pile then steam cured and sunk. Stresses
during sinking may have been as high as 7000 lb/in2 (48,263 kN/m2). The success
was spectacular, and after the war a whole range of prestressed concrete piles
became available from manufacturers.

Figure 7.21 Making a


Freyssinet Pile, 1935.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:50 Page 146

146 Historic concrete

Two methods developed in Belgium were published in Britain in 1945.151 By


1950 about 1200–1300 prestressed bearing piles had been driven in Britain, the
largest, about 12 in  12 in (304.8 mm) solid, and 18 in  18 in (457.2 mm) hol-
low.152 At that time Freyssinet, in his lecture at ICE, envisaged future piles being
of tubular precast elements of high quality concrete, with holes in their walls for
cables.153
Perhaps the earliest UK investigations were on a North Thames Gas Board
site at Stanford-le-Hope, Essex, where driving tests for a 10 in  10 in (254 mm 
254 mm)  35 ft (10.67 m) prestressed concrete pile were compared with 14 in
(355.6 mm) and 12 in (304.8 mm) square reinforced concrete piles on the same
site.154,155 Performance was satisfactory, and in early 1950 was followed up by their
application for berthing dolphins at Brentford,154 and then at Beckton155
and Bromley-by-Bow. Solid prestressed piles 12 in (304.8 mm) square and 50 ft
(15.24 m) long were test driven at Portishead B power station at this time,156 and
prestressed piles were also used at the Isle of Grain Anglo-Iranian Oil Refinery.157
Hollow piles appear to have been first used at Falmouth in 1949 — of 18 in
(457.2 mm) octagonal section and 60 ft (18.29 m) long, and again at Beckton for
a coaling jetty in 1955.154 It would appear that, particularly compared to the US,
take-up of prestressed concrete piles was relatively low and only c. 20,000 had been
driven by 1960.154
Typical details for Lee McCalls’ solid and hollow piles c. 1952 are shown in
Figures 7.22a–c, and relevant details are given in Table 7.1.158

Diaphragm Diaphragm walls were developed from the exploitation of bentonite slurries, of
walls159–164 which the most famous early application was in drilling oil wells. The ability of
bentonite, when placed in an excavation, to produce a thixotropic slurry, which
resists overbreak into an excavated shaft, enabled deep wells to be drilled when
other lining methods were impractical, and offered the potential of stabilising an
excavation while material was placed to form a more permanent barrier. Research
began in the United States before the Second World War to investigate the effi-
cacy of bentonite to control seepage through levees on the Mississippi, suggest-
ing it could be used for cut-offs.165–167 In 1945 a cut-off of clay was installed at
Trotters Levee by J.W. Black using a slurry to support the excavation.159
A.D. Rhodes installed a cut-off wall several kilometres long for the Los Angeles
Harbour levees in 1950.159,168 This early example of the ‘slurry trench’ process
involved the placing of selected soil material blended with bentonite slurry, and
by 1968 more than 630,000 m2 of material had been placed by this method by
US and Canadian contractors, the largest single job being the Dead Sea Dikes.163 In
Europe as early as 1934 bentonite may have been used in the installation of a bored
pile wall, and in 1948, 1 m diameter piles were installed after excavation using a
bentonite slurry at Bone in Algeria. 159,168–170
The technique was discussed by Professor Lorenz of Berlin in 1950,171 and
in 1951 he took out his first patent.172 This was largely overshadowed by Veder’s
Italian patent of 1952.173 Veder had carried out his first private tests of structural
diaphragm wall techniques using trenches in 1948, and practical applications
followed in 1950 at Fedala Dam.173 This early Italian work gained the most public-
ity.174 Following disastrous floods in the Po Valley in 1951, in 1952 the Italian
Ministry of Public Works initiated a programme of flood defence works which
involved placing of concrete ‘diaphragms’ within the clay dykes. Initially this was
done by bored pile techniques,174 but in 1954 a bentonite technique was tried out
by the contractors ICOS, advised by Veder, excavating a 500 mm wide trench,
which was filled with bentonite and then concrete placed. The success of the tech-
nique was immediate, and was successfully applied for dam cut-offs, foundations,
basement walls and cofferdams.175 Through the 1950s research continued led by

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:50 Page 147

Concrete foundations and substructures 147

(a)

Figure 7.22 Macalloy pre-


stressed piles, 1952: (a) solid,
(b) hollow; and (c) open-
ended tubular piles.

Lorenz and Veder, and specialist plant was developed for excavating the trenches
and placing the bentonite and concrete. Soletanche developed the Radio Marconi
system at this time, Soletanche being associated with Soil Mechanics Limited in
the UK.169
The earliest British application of diaphragm walling was at Hyde Park Corner
underpass in 1961.176 ICOS were the contractors. The work was closely monitored,
and much interest was attached to this success in London Clay. The same year a
diaphragm wall was installed to take vertical and horizontal loading at an RAF
base.160 By the late 1960s the technique had become well-established in Britain.
The success of the system in London clay saw its widespread use for the support
of excavations and creation of deep basements as at King’s Road (1964) for Boots’
and Sainsburys’ stores.160 The largest application was at Seaforth Docks, Liverpool,
where a major consideration was the extent to which frictional or adhesive forces
between the ground and the wall could be mobilised. The wall here was keyed
into the underlying sandstone. ICOS were again the contractors.177,178
Other early applications included a cantilevered ICOS diaphragm wall for the
A412 widening scheme near Rickmansworth,179 a curved retaining wall for the

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:50 Page 148

148 Historic concrete

Figure 7.22b

Victoria Circus development in Southend,180 Bromley Library/Theatre complex,181


London’s Guildhall precincts redevelopment (1970),181,182 London Central
YMCA (1971),182 Grangemouth pumping station perimeter wall (1970–71),183
Peterhead quay wall (1974),183 and Corsehouse dam cut-off (1973).183 The first
application of concrete diaphragm wall techniques for a British dam was for reme-
dial works at Balderhead dam in the late 1960s.184

Bored piled Precast concrete sheet piles were used by Hennebique in the early days of rein-
retaining walls forced concrete, and first used in Britain for a retaining wall on the Itchen at
Southampton in 1897.67 Between the wars concrete piles were driven either side
by side, or at intervals supporting reinforced concrete slabs to form retaining walls.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:50 Page 149

Concrete foundations and substructures 149

Figure 7.22c

Table 7.1 Macalloy bearing piles

Size Dia. Wt/ft No. of Prestress Maximum Maximum Maximum length for handling (ft)
(in  in) of (lb) bars lb in2 safe B.M. safe load
hole lb/in (t) Maximum •21ᐉ
•3ᐉ
in Initial Final cantilever

Type ‘A’ solid*


10  10 104 1–1” 830 695 107,000 56 12 40 57
10  10 104 1–11⁄8” 1050 882 136,000 48 13 45 65
11  11 126 1–11⁄8 860 723 150,000 67 13 43 61
12  12 149 1–11⁄8 720 600 146,000 88 12 40 59
14  14 204 2–1” 830 700 317,000 112 14 48 70
14  14 204 2–11⁄8” 1035 870 382,000 97 16 53 77
16  14 233 2–11⁄8” 920 770 390,000 121 15 50 72
Type ‘B’ hollow†
12  12 8 96 1–11⁄8” 1100 935 220,000 43 18 60 86
14  14 9 131 1–11⁄8” 806 675 244,000 73 16 54 77
16  16 11 161 1–11⁄8” 655 550 297,000 99 16 54 77
16  16 11 161 2–11⁄8” 1310 1100 594,000 61 23 76 109
18  18 12 214 1–11⁄8” 493 413 330,000 144 14 49 70
18  18 12 214 2–11⁄8” 985 825 660,000 106 20 69 99

* No. of bars at 42 t/in2.



No. of bars at 45 t/in2.

Concrete sheet piles might have a joint such as a tongue and groove connection,
or grouting grooves. The walls might be anchored using a tie rod connecting a
waling to the anchor, or by the use of a slab connecting the retaining wall to rak-
ing piles driven behind the wall.185

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:50 Page 150

150 Historic concrete

As early as 1914 Dyckerhoff and Widmann constructed a retaining wall using


the Strauss method of in-situ pile construction developed by the Russian mining
engineer Anton Strauss from drilling techniques.186 British interest in bored
pile retaining walls grew in the late 1950s alongside the development of large dia-
meter bored piles, Sir Robert McAlpine’s used bored piles in the retaining wall
at the TUC building c. 1955–58,187 and Derrington advocated the use of such
piles for retaining walls in 1961.188 In that year Soil Mechanics Limited installed
24 in (609.6 mm) and 17 in (431.8 mm) diameter bored piles as a continuous
diaphragm for the basement of the Audley Square multi-storey car park using the
Soletanche system (Figure 7.23).189,190 At the Stag Place redevelopment in the early
1960s a Hochstrasser bored pile wall was installed where the western boundary
wall ran alongside the King’s Scholars pond sewer. The piles were 76 cm in diam-
eter and reinforced with secondhand bull-head rails, with a facing of 5 in (127 mm)
of concrete over the inner face to complete the work.191
Bored pile walls were installed to two main patterns — contiguous and secant.192
In the former case piles were installed more or less adjacently, in two alternative
series, to give the concrete in the first series time to gain strength before boring
the adjacent pile. With secant walls the piles were installed by boring and con-
creting at centres of less than two pile diameters, again in two series, creating an
interlocking structure. Lilley Construction developed a method of construction
along these lines in 1968.193 In their case the first series of sites to be installed were
known as ‘female’ piles, and the second series, installed midway between, and cut-
ting a ‘secant’ section from the female pile, were known as ‘male’ piles. For lighter
forms of construction the male piles were reinforced with mild steel bars, but for
heavier work all piles were reinforced with UB sections. Relatively little British
literature was produced on either technique before the mid-1970s. Examples of
contiguous bored pile walls included the NatWest Tower,194 and secant pile walls

Figure 7.23 Bored pile wall


at Audley Square, London
W1, 1961.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:50 Page 151

Concrete foundations and substructures 151

were used for the Heathrow extension of the Piccadilly line,195 and for the British
Library basement.196

Reinforced earth196a1–8

In the 1960s the French engineer Henri Vidal developed ‘Terre armee’ or ‘rein-
forced earth’ as an alternative to heavy retaining walls to stabilize earthworks built
above existing ground level, using galvanized or stainless steel strips to
strengthen or reinforce the soil by mobilizing the friction between the (in most
early applications cohesionless) soil and the reinforcing elements. The concept of
strengthening earth structures was not new, and traditional methods are
described in C.W. Pasley’s early 19th century work on fortification. There is little
evidence however that these ideas had been taken forward in an analytical sense
before Vidal began his investigations, apparently prompted by building sandcas-
tles. British engineers use of the techniques was delayed by a patent dispute
between Vidal and the British Government. Whilst the technique in its origin was
not an application of concrete, as it developed the use of concrete facing units,
whether structural or non structural, became an integral part of the system. The
first application in the UK was in 1973, and research at TRRL and elsewhere pub-
licized the system. Techniques of soil reinforcement developed rapidly, with the
use of grids and geotextiles rather than steel strips for reinforcement.

Crib walls196b1–20

Crib walling is another technique whose origins can be traced back centuries, but
which was reinvented in the last quarter of the 20th as another economic alter-
native to heavy retaining wall construction. At its simplest it takes the form of alter-
native layers of precast concrete elements acting as stretchers and headers. Such
open bottomed timber boxes or cribs, filled with earth or stones, have a long his-
tory of use as a cheap means of slope stabilization where timber was in abundance,
such as the Alpine areas of Europe. Cribwork was used extensively by North Amer-
ican engineers through the 19th century, the term being also applied to open cof-
ferdams for piers and bridgework.(196b1–3) The rapid spread of the railway network
and widespread availability of timber sleepers or ties provided a ready source for
the stretchers and headers from which cribs could be constructed. In the early
20th century, as reinforced concrete beceame increasingly popular, precast ele-
ments replaced timber on American railways. Such elements were in use in North
America by the time of the First World War, some taking the form of open
boxes or bins.(196b4–6)
Cribwalls provided an economic alternative to heavier forms of retaining
wall construction. They were easy and quick to assemble, although there were
height limitations. A considerable research effort in Germany, New Zealand
and particularly Austria where it began in the 1960s and continued into the 1980s,
did much to improve the understanding of crib structures and develop their
design.(196b8–12,15,16) One development, the ‘New Wall’, replaced the stretchers by
anchoring straps, and was intended to address the height issue.(196b13,14)
The open nature of the cribs meant that such techniques could be combined
with biotechical methods of slope stabilization, as seen in the ‘Evergreen’ wall. Some
methods have been developed combining concrete elements with live vegeta-
tion.(196b8,10,19,20)
Crib walls have generally been developed outside the UK. An early British exam-
ple, on the M5 through the Clevedon Hills, won a Concrete Society Award in

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:50 Page 152

152 Historic concrete

1973.(196b7) In the 1980s and 1990s research work was sponsored by the TR(R)L
to encourage use.(196b18)

Design of In Britain geotechnical engineering, in the modern sense of the term, is largely
a post-Second World War development. Some sense of the excitement felt by the
foundations early pioneers can be obtained from Sir Harold Harding’s autobiography.197 When
one looks at the design of foundations by previous generations of engineers one
must bear in mind, therefore, that they lacked many of the methods of site inves-
tigation, sampling, testing, analysis and design which are taken for granted today.
Problems faced by the engineer before the war were highlighted by Terzaghi in
his 1927 paper The science of foundations — its present and future.198 This focused on
specific shortcomings of foundation design at that time — selecting allowable
soil pressure regardless of the area covered by individual foundations and the
maximum permissible differential settlement of the superstructure, calculating
the bearing capacity of piles by the Engineering news formula without regard to
the properties of the soil, and using the bearing capacity of an individual pile as
a guarantee of the bearing capacity of the whole foundation. The discussion on
Terzaghi’s paper provides a fascinating insight into state of soil mechanics of
the time.
The question of an allowable soil pressure for the design of foundations appears
to have developed on an empirical basis through the 19th century (Table 7.2).199–201
One could regard foundation design of the time as a two stage process: having
computed the superimposed load of the superstructure, design foundations of suf-
ficient strength to sustain this load, while selecting the foundation type and dimen-
sions to ensure that the load would not exceed the safe bearing capacity of the
ground. It is apparent there was little consensus in the late 19th century as to what
the safe bearing capacity might be. This dilemma was highlighted by E.L. Corthell
in 1902 when involved in the design of deep foundations at Rosario harbour in
Argentina.202 The experienced contractors Schneider and Hersent proposed a
foundation based on a load of 8 kg/cm2 (7.3 t/ft2). This was rejected by the Board
considering the design, and after considerable discussion an allowable bearing pres-
sure of 3.5kg/cm2 (3.2 t/ft2) was determined upon, with consequent increase in the
cost of the works. Corthell was dissatisfied with the lack of consensus among engi-
neers as to safe bearing capacities of soils, and compiled a large amount of data
to illustrate the situation (summarized in Table 7.3).

Table 7.2 Bearing capacities used in foundation design, 1830–90

Project Ground conditions Foundation type Design loads


(t/ft2)

London Bridge (1831) London Clay/Woolwich Timber piles 5.75*†


and Reading Beds
Nelson’s Column London Clay Concrete block 1.3
Crystal Palace (1850) Compact gravel Concrete footings 2.5
Charing Cross Bridge (1860) Gravel over London Clay Concrete-filled cast 8*
iron cylinders
Hownes Gill Viaduct (1861) Debris over clay or gravel Inverted arches 1
Westminster Bridge (1862) London Clay Timber piles and cast 2
iron cofferdams
Tower Bridge (1886–90) London Clay Concrete and caissons 4

* Settlement known to have been a problem. At Charing Cross it was partially solved by pre-loading the
foundations.

This load was calculated after the bridge was completed. It is unclear what design load the Rennies used.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:50 Page 153

Concrete foundations and substructures 153

Table 7.3 Summary of Corthell’s findings (1907) for deep foundations

Ground type Loads (t/ft2) Average load (t/ft2) Number of examples

Fine sand 2–5.2 4 10


Coarse sand and gravel 2.1–6.9 4.6 33
Sand and clay 2.25–7.6 4.4 10
Soil and alluvium 1.3–5.5 2.6 2
Hard clay 1.8–72 4.6 16

Corthell was not the first to investigate the question.199 In the late 1880s
I.O. Baker had attempted, by examining a group of case studies, to compile some
guidance on safe bearing capacities of various types of ground.203 Even earlier
British engineers in Bengal, confronted with numerous examples of settlement
and cracking of buildings in Calcutta, carried out a series of experiments to
establish the optimum load on the alluvial soil of the area, and the depth to which
foundations should be dug, concluding that to avoid differential settlement the
load should not exceed one ton per square foot (107.3 kN/m2), and in undisturbed
ground the foundation depth should be 4–6 ft204 (1.2 m–1.8 m). In 1893 Sutcliffe50
and Newman published some figures for various types of ground which bear many
similarities to the recommendations of the 1950 Civil Engineering Code of Practice
for Foundations.209 The first statutory regulations appear to be those contained in
the iron and steel frame regulations of the 1909 London County Council (General
Powers) Act.211 Over the next 30 years guidelines were published in various
trade catalogues, some of which were more detailed than the LCC recommenda-
tions.205–212 These various recommendations are summarized in Table 7.4.

Pile driving formulae

Another area discussed by Terzaghi was the value of dynamic pile driving for-
mulae.213 From the early 18th century various formulae were proposed by engi-
neers and scientists to calculate the percussive effect of piling engines, and relating
the force exercised by the ram to the set and the bearing capacity of the founda-
tion. Much was written on the subject, and a large number of formulae are listed
by Chellis.214 Among the earliest formulae to come into widespread use were those
of Woltmann215,216 and Eytelwein.217,218
There is not much evidence to suggest these formulae were used by British engi-
neers in the first half of the 19th century. It is possible that a crude formulae based
on the velocity of the ram as described by Cresy in his Encyclopaedia of Civil Engi-
neering (1847) was used.
In the second half of the 19th century A.M. Wellington developed the Engineering
news formula.219 This was apparently widely used, and continued to be in the early
20th century. All of these formulae were essentially developed before steam ham-
mers were widely used, and were modified accordingly around the end of the
century.
Of the formulae developed in the first half of the 20th century two attracted
most comment. The Hiley formula was developed in the 1920s.220–222 Dissatis-
faction with this and other formulae led Oscar Faber to develop his own formu-
lae, attempting to take account of the difference in behaviour between piles driven
in clay and sand or ballast.223
His formulae attracted much interest at the time, but their value was immedi-
ately questioned particularly with reference to clay. As the science of soil mechan-
ics has progressed and foundation technology changed such formulae have been
replaced by more reliable methods of foundation design.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
154
07.qxd 15/11/2001 15:50 Page 154

Table 7.4 Allowable bearing value in t/ft2 (kN/m2) of soils, 1893–1990

Soil type Baker Sutcliffe Newman LCC Redpath Brown BRC BRC Redpath Brown CECP4 BS
Historic concrete

(1889)203 (1893)50 (1893)202a (1909–30)211 (1913)205 (1918)206 (1932)207 (1938)208 * (1950)209 8004210

Rocks 10–20 Beware of 8–20 16 15 5–30 10–40 (2000–10,000)


(1000–20,000) fissures (860–2145) (1700) (1600) (535–3200) (1070–4300)
Chalk 1–4 3–6 6 6 (640)
(107–430) (320–640) (640)
Dense sand 8–10 4–6 7–9 4 4–8 6–8 4–8 4 4–6 (600)
and gravel (800–1000) (430–640) (750–965) (430) (430–860) (640–860) (430–860) (430) (430–640)
Sand and 2–3 6–7 4 3 4–6 4 2–4 (200–600)
gravel (215–320) (640–750) (430) (320) (430–640) (430) (215–430)
Compact 4–6 5–7.5 6–7 2 4–6 4 (300)
sand (400–600) (535–800) (640–750) (215) (430–640) (430)
Medium 2–4 2–3 3.5–5 2 3–4 3 (320) 2 2–4 (100–300)
compact (200–400) (215–320) (375–535) (215) (320–430) (215) (215–430)
sand

IP: 130.194.20.173
Loose sand 1–1.15 2.5–3 1 1–2 1 (107) 1 1–2 (100)
(107–160) (250–300) (107) (107–215) (107) (107–215)

On: Tue, 17 May 2011 06:20:09


Dry compact 4–6 3–5 5–8 4 4–5 4–6 2–4 3–4 4–6 (300–600)

Delivered by ICEVirtualLibrary.com to:


clay (400–600) (320–535) (535–860) (430) (430–535) (430–640) (215–430) (320–430) (430–640)
(London)
Compact 2–4 2–3 3–6 2 2 2–4 2–3 2 2–4 (150–300)
clay (200–400) (215–320) (320–640) (215) (215) (215–430) (215–320) (215) (215–430)
Moist clay 1–1.5 1.5–2 1 (107) 1–2 (75–150)
(107–160) (160–215) (107–215)
Soft clay 1–2 0.5–0.75 0.25–1 1 1–2 0.5–1 1 0.5–1 (75)
(100–200) (50–80) (25–107) (107) (107–215) (50–107) (107) (50–107)
Alluvial soil/ 0.05–1 0.5–0.75 0.2–1.5 0.5–1 0.5
quicksand (5–100) (50–80) (20–160) (50–107) (50)

Note: imperial units are precise.


*Similar to 1937 LCC Regulations.212
07.qxd 15/11/2001 15:50 Page 155

Concrete foundations and substructures 155

Conclusions This review has covered a period of two centuries of the use of concrete under-
ground, a period when engineers have become increasingly confident in its use
and inventive in its application. For much of this century many of the new devel-
opments have been imported from the continent or the United States, by foreign
born engineers, or enterprising contractors. One might almost say British engi-
neers went to sleep between the wars, to be awakened under the impact of soil
mechanics in the 1940s and 1950s. To an extent this is reflected in the literature,
and there was no British textbook to compare with Patton200 or Fowler224 for 50
years until the works of Henry,225 Tomlinson,226 and Little227 were published. Lim-
itations of space have precluded a discussion of the impact of plant on design and
construction, but many of the post-war developments are a direct result of rapidly
changing plant technology. This is a story which needs to be written.

References 1. Holy Bible. Matthew 7: 24–27.


*2. Skempton, A.W., Landmarks in early soil mechanics. Proceedings of the VII
European Conference on Soil Mechanics and Foundation Engineering: Design
Parameters in Geotechnical Engineering, 1979, 5, 1–26.
*3. Skempton, A.W., A history of soil properties 1717–1927. Proceedings of the 11th
International Conference on Soil Mechanics and Foundation Engineering, 1985,
Golden Jubilee Volume, 95–121.
*4. Heyman, J., Coulomb’s Memoir on Statics. Cambridge University Press, 1972.
5. Glossop, R., The invention and development of injection processes. Part I:
1802–1850. Geotechnique, 1960, 10, 91–100; Part II: 1850–1960. Geotechnique,
1961, 11, 255–79.
6. Glossop, R., The invention and early use of compressed air to exclude water from
shafts and tunnels during construction. Geotechnique, 1976, 26, 253–80.
7. Peck, R.B., History of building foundations in Chicago. University of Illinois,
Engineering Experiment Station, Bulletin Series 373, 1948.
8. Flodin, A., Broms, B., Historical development of civil engineering in soft clay. Devel-
opments in Geotechnical Engineering 20: Soft Clay Engineering. 1981: 28–156.
9. Kerisel, J., Down to Earth: Foundations Past and Present. Balkema: Rotterdam,
1987.
10. Skempton, A.W., Foundations for high buildings. ICE Proc., 4, Part III, 246–69.
*11. Dobson, E., A Rudimentary Treatise on Foundations and Concrete Works. Weale:
London, 1850 (9th edn, Crosby Lockwood, 1903).
12. Law, H., The Rudiments of Civil Engineering, 4 parts in 1 Vol. Weale: London,
1848–52.
13. Hughes, T., A series of papers on the foundations of bridges in Weale, J. Comp.
The Theory, Practice and Architecture of Bridges of Stone, Iron, Timber and
Wire. Weale: London, 1839–43.
14. Skempton, A.W., The Albion Mill Foundations. Geotechnique, 1971, 21, 203–10.
(This foundation can be described as buoyant, but it seems unlikely the concept
was recognised before the latter part of the 19th century. The problem of foun-
dations for large mill buildings was generally avoided by building on rock!)
15. Smeaton, J., A narrative of the building and a description of the Construction
of the Eddystone lighthouse with stone, London, 1791.
*16. Huberti, G. et al., Vom Caementum zum Spannbeton. Band 1. Bauverlag:
Wiesbaden, 1964: 5–36.
17. Belidor, B.F., de. Architecture hydraulique, II Partee, Tome 2. Jombert: Paris,
1753: 178–90.
18. Vicat, L.J., Recherches experimentales sur les chaux de construction, les betons
et les mortiers ordinaires. Goujon Paris, 1818.
19. Vicat, L.J., Note sur un mouvement periodique observé aux voutes du Pont de
Souillac. Annales de Chimie et de Physique, 1824, 27, 70–79 (abstracted in
Mechanics Magazine, 1825, 3, 264–65).
*Particularly useful for general 20. Vicat, L.J., Notice sur le pont et Souillac construit sur la Dordogne.
information. Bulletin des Sciences Technologiques, 1826, 6, 117–19.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:50 Page 156

156 Historic concrete

21. Mary, –. De l’emploi du beton dans la foundation des ecluses. Annales des Ponts
et Chaussees, 1832, 3, 66–105.
22. Semple, G., A Treatise on building in water, Dublin, 1776.
23. Godwin, G., Essay upon the nature and properties of concrete, and its applica-
tion to construction up to the present period. Trans. Inst. Br. Architects, 1835–36,
1, 1–39.
*24. Pasley (Sir), Observations on Limes, Calcareous Cements Mortars, Stuccos and
Concrete (etc.). Weale: London, 1838: 265–73, Appendix pp. 23–33. (This is the
most authoritative source on the early 19th century use of cements and concrete.)
25. Crook, J.M., Sir Robert Smirke: a pioneer of concrete construction. Trans. New-
comen Soc., 1965–66, 38, 5–22.
26. On what materials and their properties are best to form concrete masses for the
foundations of buildings. ICE Min. Conversations, 1830, 264–69 (unpublished).
27. Donaldson, T.L., Handbook of Specifications. Atchley: London, 1857: 416–27.
(The concrete was to be thrown into the trench from a height of at least 10 feet,
a practice condemned by Dobson.)
28. Colvin, H.M. et al., History of the Kings Works, Vol. 6. HMSO: London, 1982:
606.
28a. Clegg, S., Jr., A Treatise on the manufacture and production of coal-gas, 1844,
etc.
29. Chrimes, M.M., Hugh McIntosh (1767–1840): national contractor, Trans. New-
comen Soc., 1994–95, 66, pp? This work draws heavily on the researches of
Dr David Brooke and Professor A.W. Skempton.
30. Taylor, G.L., An Account of the methods used in underpinning the long store-
house at His Majesty’s Dock Yard, Chatham, in the Year 1834. Trans. Inst. Br.
Architects, 1835–36, 1, 40–43.
31. Taylor, G.L., The Auto-biography of an Octogenarian Architect, Vol. 1, 2 vols.
Longmans: London, 1870–71: 167–72.
32. Metropolis Management and Buildings Act, 1878. HMSO: London, 1878.
33. Fletcher, B., The Metropolitan Building Acts: A Textbook for Architects, Sur-
veyors, Builders, etc. Batsford: London, 1882.
34. On concrete. ICE Minutes of Conversation, No. 158, 4 March 1830, 531–32. The
proportions at Archway were 1 cement: 2 sand: 7 gravel.
35. Paxton, R.A., The Influence of Thomas Telford (1757–1834) On the Use of
Improved Constructional Materials in Civil Engineering Practice, MSc thesis,
Heriot-Watt, 1975.
36. Skempton, A.W., Portland cements, 1843–1887. Trans. Newcomen Soc.,
1962–63, xxxv, 117–52.
*37. Vernon-Harcourt, L.F., Harbours and Docks, 2 vols. Clarendon: Oxford, 1885.
*38. Bray, R.N., Tatham, P.F.B., Old Waterfront Walls. Spon: London, 1992.
39. Guillery, P., Building the Millwall Docks. Constr. History, 1990, 6, 3–22.
40. Rickman, J., (ed.), The Life of Thomas Telford. London, 1838: 156–57.
41. Smith, N.A.F., History of Dams. P Davies: London, 1971: 208–13.
42. Skempton, A.W., Historical development of British embankment dams to 1960.
In: Clay Barriers for Embankment Dams. TTL: London, 1990: 15–52.
43. Binnie, G.M., Masonry and concrete dams 1880–1941. Industr. Archaeol. Rev.,
1987, X, 41–58.
44. Bateman, J.F., La Trobe, The History and Description of the Manchester Water-
works. Spon: London, 1884.
45. Wegmann, E., The Design and Construction of Masonry Dams. Wiley: New York,
1888.
46. Gordon, G., Concrete dam for Geelong waterworks. Professional Papers on
Indian Engineering. Second Series, 4(178), 402– 404.
46a. James IGSE
47. Institution of Civil Engineers Engineering Conference, 1907. Trans. Sect. Rail-
ways, Docks Harb., ICE: London, 1907.
48. Shankland, E.C., Steel skeleton construction in Chicago. Min. Proc. Inst. Civil
Engrs, 1896–97, 128, 1–27.
49. The High Court, Calcutta. The Builder, 30 October 1869, 27, 857.
50. Sutcliffe, G.L., Concrete: Its Nature and Uses. Crosby Lockwood: London, 1893.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:50 Page 157

Concrete foundations and substructures 157

51. Twelvetrees, W.N., Concrete-Steel Buildings. Whittaker: London, 1907, 337–54.


52. Jones, B.E., (ed.), Cassell’s Reinforced Concrete. Cassell: London, 1913.
53. Twelvetrees, W.N., Concrete-Steel Buildings. 1907: 8–10.
54. Twelvetrees, W.N., Concrete-Steel Buildings. 1907: 105–07.
*55. Marsh, C.F., Reinforced Concrete. London, 1904. The third edition (1906)
co-written with William Dunn contains minor amendments and updates.
56. Edgell, G.J., The Remarkable Structures of Paul Cottancin. Struct. Engr, 1985,
63A, 201–07.
57. Galbraith, A.R., The Cottancin System of Armoured Construction.
Colwell: Ipswich, 1904.
58. Emploi du sable dans les foundations sur sol compressible. Annales des Ponts et
Chaussees, 1835, 10, 171–214.
59. Mitchell, A., On submarine foundations; particularly the screw pile and moor-
ings. Min. Proc. Inst. Civil Engrs, 1848, 7, 108–132.
60. Jopling suggested the use of concrete filled cast iron columns in 1834. Mechan-
ics Magazine, 1834, 20, 424–28.
61. Concrete filled (square) cast iron piles are described. Architect Engr Surveyor,
1843, 4, 24, 359–62.
62. Walmisley, A.T., Reinforced concrete piles in tidal waters. Concr. Constr. Eng.,
1906–1907, 1, 88–101.
63. Galbraith, A.R., Reinforced concrete piling. Proc. Assn Municipal County Engrs,
1904–1905, 31, 356–75.
64. Reinforced concrete work on the Coignet system in England. Concr.
Constr. Eng., 1907, 2, 320.
65. Gow, C.R., History and present state of the concrete pile industry. Proc. Am.
Concr. Inst., 1917, 13, 174–218.
66. Cummings, R.A., Reinforced concrete piles. Proc. Natl. Assn Cement Users, 1912,
8, 312–25.
67. L.G. Mouchel and Partners. Mouchel-Hennebique Ferro-Concrete, 4th edn.
Mouchel: London, 1909–21.
68. Effect of steam curing on the crushing strength of concrete. Eng. News, 1907,
58, 249–50.
69. Wig, R.J., The Effect of high pressure steam on the crushing strengths of Port-
land cement, mortar, etc., concrete. United States Bureau of Standards, Tech-
nological Papers, 1912.
69a. Reinforced concrete systems: The Williams system. Builders J., 1908, 439–42.
(This lists 13 contracts executed to the system.)
70. Spirally armoured concrete piles. The Engineer, 1908, 105, 476–77.
71. Berger, C., Guillerme, V., La construction en ciment armé. Dunod: Paris, 1902,
336–37.
71a. Edmond Coignet Limited. Reinforced Concrete Systems. Coignet: London,
c.1911. (Coignet’s London office was opened in 1905.)
72. A New system of concrete pile construction. Eng. News, 1901, 45(25), 450–51.
73. Concrete pile foundations at Aurora. Eng. News, 1902, 48(24), 495.
74. Concrete piles for sandy ground. Eng. News, 1903, 49, 275. (James Brunlees first
developed a system of water jetting for the installations of cast iron piles across
Morecambe Bay, with the contractor Philip Brogden in 185).
75. Shuman, C., The Simplex system of concrete piling. Proc. Engrs Club Philadel-
phia, 1905, 22, 347.
76. Mackellar, T., The Simplex system of concrete piling. Assn Eng. Soc., 1907, 39,
266–76.
77. Recent improvements in piles. The Engineer, 1906, 101, 79.
78. Recent improvements in piles. The Engineer, 1906, 101, 382.
79. Ellis, S.H., Tranmere Bay Development Works. Min. Proc. Inst. Civil Engrs, 1907,
171, 127–70.
80. The Franki Compressed Pile Company Limited. Franki Pile: London, 1957.
81. Braatvedt, I.H., Guide to Piling and Foundation Systems. Bramley, Transvaal,
Frankipile, 1975.
82. Der Eisenbetonbau auf der Weltausstellung in Brussel. Beton und Eisen, 1911,
10, 1–4.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:50 Page 158

158 Historic concrete

83. British patent rights in reinforced concrete piles. Concr. Constr. Eng.,
1906–1907, 1, 282–85.
84. Tobacco Warehouse. General correspondence. Bristol Records Office.
85. British patent rights in reinforced concrete piles. Concr. Constr. Eng., 1907–1908,
2, 139–48; Ferro-concrete pile case. Engineering, 1907, 83, 354.
86. British patent rights and reinforced concrete piles. Concr. Constr. Eng., 1909, 4, 4.
87. The Enlargement of Victoria Station. Engineering, 1906, 82, 35–38.
88. Information provided by B.L. Hurst.
89. The Ritz Hotel. Concr. Constr. Eng., 1906, 1, 441.
90. Drawings and specification in ICE Archives.
91. Twelvetrees, W.N., Concrete-Steel Buildings. Whittaker: London, 1907: 72–93.
92. Twelvetrees, W.N., Concrete-Steel Buildings. 1907: 240–52.
93. Twelvetress, W.N., Concrete-Steel Buildings. 1907: 232–39.
94. Copies of drawings, etc., in Concrete Archive at ICE.
95. Problems of pile deformation were well-known before the First World War with
in-situ techniques, see Gow’s paper (ref. 65).
*96. Ketchum, M.S., The Design of Walls, Bins and Grain Elevators. Eng. News, New
York, 1907. (This work was published in London by Constable and available at
ICE from December 1907.)
97. Bone, E.P., Reinforced concrete retaining wall design. Eng. News, 1907, 57(17),
448–52.
98. Bone, F.A., Comparative costs of gravity and reinforced-concrete retaining walls.
Eng. News, 1908, 59(13), 347.
99. Klein, A., Die Form der Winkelstutzmauem aus Eisenbeton mit Rucksicht
auf Bodendruck und Reiburg in der Fundamentfuge. Beton und Eisen,
1909, 16, 384–87; Gilbrin, G., Die Wirtschaftlichste Form der Eisenbeton-
Winkelstutsmauer. Beton und Eisen, 1912, 10, 233–35. (See for German work.)
100. The Subway of the Philadelphia Rapid Transit Co. Eng. Record, 1905, 51(8) 224–28.
101. Graff, C.F., High reinforced concrete retaining wall construction at Seattle,
Washington. Eng. News, 1905, 53(10), 262–64.
102. Graff, C.F., Difficult reinforced concrete retaining wall construction on the Great
Northern Railway. Eng. News, 1906, 55(18), 483–87.
103. American Railway Engineering and Maintenance of Way Association. Retaining
walls and abutments. Proc. AREMWA, 10, Part 2, 1909, 1317–37.
*104. Christophe, P., Beton Armée. Beranger: Paris, 1902: 290–91.
105. Marsh, C.F., Reinforced Concrete. Constable: London, 1904: 66–67.
106. Cantell, M.T., Reinforced Concrete Construction: Advanced Course. Spon:
London, 1912.
107. Reinforced concrete retaining wall in Queen Victoria Street, London EC. Concr.
Constr. Eng., 1911, 6, 956–958.
108. A Retaining wall in Piccadilly. Concr. Const. Eng., 1908, 2, 509.
109. Reinforced concrete at the Royal Insurance Building. Concr. Constr. Eng., 1911,
6, 811–20.
110. Consolidated Expanded Metal Companies. A Handbook of Design. Braddock:
PA, 1919: 189–203.
111. Expanded Metal Company. Expament Pamphlet, No. 5, 2nd edn. Expament:
London, 1923, passim.
112. Watson, T.A., The Reinforced concrete construction of the new General Post Office
Buildings, London. Trans. Jr. Inst. Engrs, 1909–1910, 20, 58–71.
113. Bylander, S., The Architectural and engineering features of the Royal Automobile
Club Building. Trans. Jr. Inst. Engrs, 1910–1911, 21, 243–70.
114. Reinforced concrete in Aldwych: General Accident Assurance Building. The
Builder, 1909, 97, 234.
115. British Reinforced Concrete Engineering Company. BRC Reinforcements, 9th
edn. BRC: Stafford, 1932: 262–65.
116. Detailed records of hard pile driving were published in the US before World War
I. Thompson, S.E., Fox, B., Cast reinforced concrete piles. J. Assn Eng. Soc., 1935,
1, 150–234.
117. Glanville, W.H. et al., The Behaviour of reinforced concrete piling during dri-
ving. J. Inst. Civil Engrs, 1935, 1, 150–234.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:50 Page 159

Concrete foundations and substructures 159

118. Glanville, W.H. et al., An Investigation of the stresses in reinforced concrete piles
during driving. Building Research Station. Technical paper 20, 1938.
119. Isaacs, D.V., Reinforced concrete pile formulae. J. Inst. Engrs, Australia, 1931,
3(9), 305–23.
120. Smith, E.A.L., Pile-driving analysis by the wave equation. ASCE Proc. J. Soil Mech.
Div., August 1960, 86, SM4, 35–61.
121. Goble and Associates. Wave Equation Analysis of Pile Driving: WEAP Program, 3
Vols. US Federal Highway Administration: Washington, DC, 1981.
122. Joint Sub-Committee on Pile Driving. Interim Report 1. ICE: London, 1938.
123. Joint Committee on Pile Driving. Chartered Civil Engineer, May 1950: 16–21.
124. Green, H., Impact testing of concrete. Proc. Conf. Mech. Prop. Non-metallic
Brittle Mater., 1958.
125. Whitaker, T., Experiments with model piles in groups. Geotechnique, 1957, 7,
147–67.
126. Joint Committee on Piled Foundations. Progress report. Proc. Inst. Civil Engrs,
December 1959, 14, N10–15. (In a sense the Committee’s work was continued
by CIRIA.)
127. McCarthy, M.J., Piling in general, with special reference to the vibro concrete
piling system. Trans. Jr. Inst. Engrs, 1927–28, 38, 171–81.
128. British Steel Piling. Royal Docks approaches improvement scheme. BSP pam-
phlet, No.8, 1934. (The licensed contractors were John Gill Contractors.)
129. Faber, O. (ed.), Concrete Yearbook. Concrete Publications: London, 1936 edition.
129a. The Francois Cementation Company Limited. London, c.1934. (The company
set up in England c.1919.)
*130. Dean, A.C., Piles and Pile Driving. Crosby Lockwood: London, 1935: (especially)
16–60.
131. Wilson, G., The Bearing capacity of screw piles and ‘screwcrete’ cylinders. Inst.
Civil Engrs J., March 1950, 34, 4–93.
132. Braithwaite and Company Engineers Limited. Braithwaite: London, c.1950,
Figures 56–73.
133. Ward, W.H., Green, H., House foundations: the short bored piles. Public Works
and Municipal Services Congress, 1952, 373–97.
134. Glossop, R., Greeves, I.S., A New form of bored pile. Concr. Constr. Eng., 1946,
41, 344–51.
135. Meyerhof, G.G., Murdock, L.J., An Investigation of the bearing capacity of some
bored and driven piles in London Clay. Geotechnique, September 1953, 3, 267–82.
136. Golder, H.Q., Leonard, M.W., Some tests on bored piles in London Clay.
Geotechnique, March 1954, 4, 34–41.
*137. Skempton, A.W., Cast in-situ bored piles in London Clay. Geotechnique, 1959,
9, 153–173.
138. Gow, C.R., Concrete piles. J. Assn Eng. Soc., 1907, 39, 255–65.
139. Hunter, L.E., Large diameter piles. Concr. Constr. Eng., May 1954, 49, 165–68.
140. Derrington, J.A., (Discussion) Large Bored Piles Symposium. ICE, 1966, 139.
141. Williams, G.M.J., Design of the foundations of the Shell Building, London. Pro-
ceedings 4th International Conference on Soil Mechanics and Foundation Engi-
neering, 1957, 1, 457–61.
142. Bored pile foundations. Roads Road Constr., July 1959, 37, 215. (This describes
the crawler mounted excavator developed by Economic Foundations, a specially
formed subsidiary of Sir Robert McAlpine’s, to drive a heavy duty auger based
on Texas drilling methods.)
143. Davis, C., Structural engineering aspects of the Millbank Tower Block, London.
Struct. Engr, 40, January 1962, 3–20.
144. Kirkland, G.W., The Millbank Tower. Consul. Engr, March 1962, 21, 290–93.
*145. Symposium on large diameter bored piles, March 1961. Reinforced Concr. Rev.,
1961, 6, 673–726.
146. Frischmann, W.W., Fleming, W.G.K., The use and behaviour of large diameter
piles in London Clay. Struct. Engr, April 1962, 40, 123–31.
*147. Large bored piles: proceedings of the symposium organised by ICE and the
Reinforced Concrete Association, ICE, London, 1966. (Skempton provided some
first approximation design rules.)

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:50 Page 160

160 Historic concrete

*148. Andrew, A.E., Turner, F.H., Post-tensioning systems for concrete in the UK:
1940–1985. CIRIA Report 106, 1985.
149. Boase, A.J., Notes on inspection of structures in Europe. Am. Concr. Inst. Proc.,
1937, 33, 521–26.
150. Freyssinet, E., Progrès pratiques des methodes de traitement mecanique des
betons. La Reprise en sous-oevre des foundations de la Gare Transatlantique du
Havre. Travaux, 1935, 199–227.
151. Magnel, G., Prestressed concrete. Some new developments. Concr. Constr. Eng.,
1945, 40, 221–32, 249–54; 1946, 41, 10–20.
152. New, D.H., Discussion on pile driving in difficult conditions. ICE Eng. Div. Papers
(Works Construction Division), 1950–51, 9, 26–27.
153. Freyssinet, E., Lecture on prestressed concrete 17 November 1949. J. Inst. Civil
Engrs, February 1950, 33, 331–80.
154. Gardner, S.V., New, D.H., Some experiences with prestressed concrete piles.
Proc. Inst. Civil Engrs, January 1961, 18, 43–66. (Discussion), April 1962, 21,
867–91.
155. Concrete piles at Beckton Gasworks. Concr. Constr. Eng., May 1950, 45, 169–70.
156. Morgan, H.D., Haswell, C.K., The Driving and testing of piles. Proc. Inst. Civil
Engrs, Part 1, 1953, 2, 43–75.
157. Records in ICE Piling Committee archives.
158. McCalls, Macalloy Limited. PRC Note 7: Piles, 2nd edn, c.1952.
*159. Boyes, R.G.H., Structural and Cut-off Diaphragm Walls. Applied Science:
London, 1976.
160. Consulting Engineer. Practical design for diaphragm walls. Consul. Engr, Suppl.,
September 1974.
*161. Institution of Civil Engineers. Proc. Conf. Diaphragm Walls Anchorages, 1974.
ICE: London, 1975.
*162. International Society of Soil Mechanics and Foundation Engineering. British
National Committee. Proc. Symp. Grout. Drill. Muds Eng. Practice. Butterworth:
London, 1963.
*163. International Society of Soil Mechanics and Foundation Engineering, 7th Inter-
national Conference, Mexico City, 1969. Proceedings of Speciality Session 14 and
15. Societe de Diffusion des Techniques du Batiment et des Travaux Publics, Paris,
1969.
*164. Xanthakos, P.P., Slurry Walls. McGraw-Hill: New York, 1979.
165. Davis, C.W. et al., Bentonite: its properties … utilization. US Bureau of Mines:
Technical Paper, 609, 1940.
166. Efficacy of bentonite for control of seepage. US Waterways Experimental Sta-
tion. Experiment Station Bulletin, 1938, 2(1), 2–6.
167. US Waterways Experiment Station. Technical Memorandum, 351–1, 1938.
168. Berthier, P., La parois moulée dans le sol. Memoires ICF, September 1964, 9,
33–46.
169. Soil Mechanics — Soletanche Limited. Technical data sheets. Soil Mechanics Lim-
ited, London, c. 1960. (These cover works executed c. 1949–59.)
170. Florentin, J., Les parois moulées dans le sol. Proc. 7th Intl. Conf. Soil Mech. Foun-
dation Eng., 3, 1969, 507–12.
171. Lorenz, H., Ueber die Verwendung thixotroper Flussigkeiten in Grundbau.
Bautechnik, 1950, 27(10), 313–17.
172. Lorenz, H., Erfahrungen mit thixotropen Flussigkeiten im Grundbau.
Bautechnik, 1953, 30(8), 232–36.
173. ICOS. Underground works. ICOS, Milan, 1968.
174. Veder, C., Method for the construction of impermeable diaphragms at great
depths by means of thixotropic seals. Proc. 3rd Intl. Conf. Soil Mech. Founda-
tion Eng., Zurich, 1953, 2, 91–94.
175. Sistonen, H., Montta and Seitakorva hydro power plants. 9th ICOLD Congr.,
1967, 3, Q.34, R.34., 609–27.
176. Granter, E., Park Lane improvement scheme: design and construction. Proc. Inst.
Civil Engrs, 1964, 27, 293–315. (Discussion), 1964, 33, 423–36.
177. Agar, M., Irwin-Childs, F., Seaforth Dock, Liverpool: planning and design. Proc.
Inst. Civil Engrs, May 1973, 54, 255–74.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:50 Page 161

Concrete foundations and substructures 161

178. Cole, P.G. et al., Seaforth Dock, Liverpool: construction. Proc. Inst. Civil Engrs,
May 1973, 54, 275–90.
179. ICOS wall solves ground widening problem. Ground Eng., 1968, 1, 38–39.
180. Diaphragm wall schemes show progress in Britain and France. Ground Eng.,
1970, 3(2), 30–34.
181. Recent diaphragm walls schemes in Britain. Ground Eng., 1970, 3(6), 22–24.
182. Little, M.E.R., An economic appraisal. Consul. Engr. Diaphragm Wall Suppl.,
September 1974, S.23–27.
183. Coats, D.J., Three examples of diaphragm walling. Consul. Engr. Diaphragm Wall
Suppl., September 1974, S.28–31.
184. Vaughan, P.R. et al., Cracking and erosion of the rolled clay core of Balderhead
Dam, and the remedial works adopted for its repair. 10th ICOLD Congr.,
Montreal, 1970, 1, Q.36, R5, 73–95.
185. Wentworth-Shields, F.E., Gray, W.S., Reinforced Concrete Piling. Concrete
Publications: London, 1938: 35–47.
*186. Emperger, F., (ed.), Handbuch fur Eisenbetonbau, 3 Auft. Band 3. Ernst, Berlin,
1922, 312–23 (ill p. 323).
187. Arup job record No. 633. Information supplied by M. Bussell.
188. Derrington, J.A., Concrete cylinder retaining walls. Reinforced Concr. Rev.,
September 1961, 5, 696–98.
189. Soil Mechanics Limited. Continuous bored pile diaphragms in civil engineering
practice. Geotechnical Pamphlet No. 9, 1961.
*190. Bullen, F.R., Notes on the History of foundation engineering. Struct. Engr.,
December 1961, 39, 385–404; 400–401.
191. Mason, J., Frost, A.D., Stag Place development. Struct. Engr., November 1963,
41, 347–365 (Discussion) May 1964, 43, 169–172.
192. North-Lewis, J.P., Lyons, G.H.A., Contiguous bored piles. ICE Conf.
Diaphragm Walls Anchorages, 1974–1975: 184–94.
193. Neal, D., The Effects of concrete mix design on secant piling. Concrete in the
Ground: Proc. Concr. Soc. Sem., 123–32. (Veder had been using techniques like
this in Finland in 1954.)
194. Frischmann, W.W. et al., National Westminster Tower: design. Proc. Inst. Civil
Engrs, August 1983, 74, 387–434.
195. Jobling, D.G., Lyons, A.C., Extension of the Piccadilly line from Hounslow West
to Heathrow Central. Proc. Inst. Civil Engrs, May 1976, 60, 191–218; (Discus-
sion) November 1976: 719–37.
196. Deep foundations for the British Library. Ground Eng., April 1984, 17, 20–26.
196a1. America Society of Civil Engineers. Earth Reinforcement. ASCE: New York, 1979.
196a2. Vidal, H., La Terre Armee Annales ITBTP, 223–24, July/August 1966, 887–938;
259–60, July/August 1969, 1099–1155.
196a3. International Conference on soil reinforcement: reinforced earth and other tech-
niques. Paris: Association Amicale des Ingenieurs Anciens Eleves de l”ENPC,
1979.
196a4. Banerjee, P.K., Principles of analysis and design of reinforced earth retaining
walls. Highway Engr., 1975, 22(1), 13–18.
196a5. Murray, R.T., Research at TRRL to develop design criteria for reinforced earth.
TRRL SR 457, 1977.
196a6. Schlosser, F., Experience on reinforced earth in France, TRRL SR 457, 1977.
196a7. Ingold, T.S., Reinforced Earth. TTL: London, 1978.
196a8. Jones, C.F.P.J., Earth Reinforcement and Soil Structures, 2nd edn. TTL:
London, 1996.
196b1. Bovey, H.T., Cribwork in Canada, Min. Proc. ICE, 1881, 63, 268–72.
196b2. Engineering News, Index 1890–1899, passim.
196b3. Fowler, C.E., Ordinary foundations, including the cofferdam process for piers,
4 eds., title varies, 1898–1920.
196b4. Anon., Precast concrete timber form crib to retain wall. ENR, 1918, 81, 763.
196b5. Anon., Precast concrete cribbing for retaining wall. ENR, 1923, 91, 718–19; 858.
196b6. Anon., Concrete cribbing for railway retaining wall. ENR, 1926, 96, 654–56.
196b7. Payne, D.F., Cribwalling in road construction (M5). DoE Construction, 1973, 7,
6–7.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:50 Page 162

162 Historic concrete

196b8. Schiechtl, H., Bioengineering for Land Reclamation and Conservation, University
of Alberta Press: Edmonton, 1980.
196b9. Brandl, H., Tragverhalten und Dimensionerung von Raumgitterstutzmauern,
Austria Bundesministerium fur Bauten und Technik Strassenforschung, 1980,
141.
196b10. Anon., Changing techniques of cribwall planting, Landscape Australia, February
1981, 91–96.
196b11. Brandl, H., Raumgitter-Stutzmauern, Austria Bundesministerium fur Bauten
und Technik Strassenforschung, 1982, 208.
196b12. Brandl, H., System von Raumgitter-Stutzmauern, 2 parts. Austria Bundesmin-
isterium fur Bauten und Technik Strassenforschung, 1985, 251.
196b13. Brandl, H., Slope stabilization and support by crib walls and prestressed anchors.
3rd Intl Geotech. Seminar, Singapore, 1985, 179–85.
196b14. Germany Forschungsgesellscaht fur Strassen und Verkehrwesen. Merkblatt
fur den Entwurf und die Herstellung vom Raumgitterwanden und Wallen,
1985.
196b15. Brandl, H., Stutzmauersystem ‘New’ und andere Konstruktione … Austria
Bundesministerium fur Bauten und Technik Strassenforschung, 1986, 280.
196b16. Hong Kong. Geotechnical Control Office. Geoguide 1: Guide to retaining wall
design, 1993, 2nd. edn.
196b17. US Army Corps of Engineers. Retaining and Flood Works. ASCE: New York,
1994, Chapter 10.
196b18. Masterton, G.G.T. et al., A Literature and design review of crib wall systems. TRL
Report 131, 1995.
196b19. Morgan, R.P.C., Rickson, R.J., Slope Stabilization and Erosion Control: A Bio-
Engineering Approach. Spon: London, 1995.
196b20. Gray, D.H., Soter, R.B., Biotechnical and Soil Bioengineering Slope Stabilization.
Wiley: New York, 1996.
197. Harding, H.J.B., Tunnelling History and My Own Involvement. Golder Associ-
ates: Toronto, 1981. For an alternative view of the era see R.E. Goodman. Karl
Terzaghi: the engineer as artist. Rston, ASCE, 1998.
*198. Terzaghi, K., The Science of foundations — its present and future. Trans. Am.
Soc. Civil Engrs, 1927, 93, 270–405.
199. Hunt, R., The Supporting power of soils. Assn Eng. Soc. J., 1888, 7(6), 189–97.
APC 1864.
*200. Patton, W.M., A Practical Treatise on Foundations. Wiley: New York, 1893.
201. Smith, J.A., Some foundations for buildings on Cleveland. Assn Eng. Soc. J., 1906,
36, 155–84.
*202. Corthell, E.L., Allowable Pressures on Deep Foundations. Clowes: London, 1907;
Wiley: New York, 1907 (abstract in ICE Min Proc., Vol. 165, 1905–1906).
202a. Newman, J., Notes on Cylinder Bridge Piers. Spon: London, 1893: 24–33.
*203. Baker, I.O., A Treatise on Masonry Construction. Wiley: New York, 1889. (I have
been unable to trace an earlier UK copy than the 3rd edn. of 1890.)
204. Leonard, H., Weight on foundations of buildings in Bengal. Profes. Paper. Indian
Eng., 2nd series, 1875, vol.4, 319–31.
205. Redpath Brown and Company. Handbook of Structural Steelwork. 1913: 289.
206. British Reinforced Concrete Engineering Company. BRC Reinforcements. BRC:
Stafford, 1918.
207. British Reinforced Concrete Engineering Company. BRC Reinforcements, BRC:
Stafford, 1932, 83.
208. Redpath Brown and Company. Handbook of Structural Steelwork. 1938: 485.
209. Institution of Civil Engineers and Others. Civil Engineering Code of Practice 4:
Foundations, ICE: London, 1950.
210. British Standards Institution. BS 8004.
211. London County Council (General Powers) Act 1909, 9th edn, 7 Ch cxxx, 1909.
212. London County Council. Bylaws for the Construction and Conversion of
Buildings and Furnace Chimneys, No. 3319. LCC: London, 1937. (These bylaws
also specified various grades of concrete.)
213. Terzaghi, K., Peck, R.B., Casagrande, A., Discussion on pile driving formulas.
Proc. Am. Soc. Civil Engrs, 68, 311–31.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
07.qxd 15/11/2001 15:50 Page 163

Concrete foundations and substructures 163

*214. Chellis, R.L., Pile Foundations, 2nd edn. New York, 1961.
215. Woltmann, R., Beytrage zur hydraulischen Architektur, 4 vols. Gottingen,
1791–99 (Vol. 4, 1799, 371–89).
216. Woltmann, R., Recherches theoriques et experimentales sur d’effet des
machines et outils … principalment sur d’effet du monton. Dieterich, Gottingen,
1804.
217. Eytelwein, J.A., Handbuch der Mechanik fester Korper, 1801, 1842, etc.
(Chapter v).
218. Eytelwein, J.A., Praktische Anweisung zur Wasserbankunst, 4 vols, 1802–08 (esp.
Vols 2 and 3).
219. Wellington, A.M., Formula for safe loads of bearing piles. Eng. News, 1888, 20,
570–12.
220. Hiley, A., The Efficiency of the hammer blow and its effects with reference to
piling. Engineering, 1922, 119, 657–58; 721–22.
221. Hiley, A., A Rational pile driving formula. Engineering, 1925, 119, 657, 721–22.
222. Hiley, A., Pile driving calculations. Struct. Engr, 1930, 8, 246–59; 278–88.
223. Faber, O., A New piling formula. J. Inst. Civil Engrs, 1946–47, 28, 5–86.
*224. Fowler, C.E., The Coffer-Dam Process for Piers. Wiley: New York, 1898 (2nd edn.,
1907, etc.)
*225. Henry, F.D.C., Design and Construction of Engineering Foundations. Spon:
London, 1956.
*226. Tomlinson, M.J., Foundation Design and Construction. Pitman: London, 1963.
*227. Little, A.L., Foundations. Arnold: London, 1961.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:20:09
08.qxd 15/11/2001 15:51 Page 165

8 The early development of reinforced


concrete shells

Peter Morice and Hugh Tottenham


Synopsis This chapter discusses briefly the engineer’s problem of creating structural systems
to enclose spaces using non-tensile resistant materials and draws attention to
Nature’s solutions to similar problems. The development of reinforced concrete,
a material which can be readily shaped and which has tensile as well as compres-
sive strength, enabled a radical change in possible structural forms, with the thin
shell being one major example. The early development of these forms of struc-
ture took place mainly outside Britain, probably due to our established traditions
of metallic construction. However, the shortages of steel following the Second World
War gave great impetus to their adoption and many interesting structures were
designed and built by British engineers.

Before the industrial revolution, engineers had a limited range of structural mater-
ials of which the most permanent, stone, was available only in small pieces and
had a relatively poor tensile strength. The solution to the problem of spanning gaps
using masonry, one of the structural engineer’s principal tasks, was to adopt the
geometrical form of the arch, which has a primarily compressive internal stress sys-
tem. This two-dimensional structure is readily extended to form the vault, or into
its true three-dimensional form as the dome, with, again, a primarily compressive
stress system. The dome, like the arch and vault, can therefore also be built from
individual small elements with effectively zero tensile strength between the elements.
We are well aware, nevertheless, of the feats of structural engineering achieved from
Roman times to the present day using masonry in these structural forms, which
have indeed had a significant influence on modern shell construction.
A way of visualizing how these structural systems behave, and the shapes which
correspond to applied loads, is gained by reversing the structure to become a ten-
sile form. Thus the hanging string or chain supporting vertical loads is the inverse
of the arch and, as it has negligible bending strength, all the loads are necessar-
ily carried by direct tensile stress. Similarly, a balloon can be considered to rep-
resent the inverse of a dome, although the loading by internal pressure is a little
different from true vertical load and will lead to a slightly different geometry.
Indeed, the fact that a true direct stress system requires a unique geometry is seen
by the changing shape of a hanging string when a concentrated load is applied
or when we apply a concentrated load to an inflated balloon. A flexurally stiff
system will accommodate the tendency to distort under changing load by devel-
oping internal bending and shear stresses. The engineer Gaudi1 used inverse
models to develop the most appropriate geometries for his somewhat unusual
compressive stress roof structures.
An analysis of the membrane stresses induced in shell structures was first given
by Lame and Clapeyron in 1828.2 They showed that loading on shells can pro-
duce a consistent direct and shear stress system; this may lead to tensile as well as
compressive stress resultants in certain regions of the shell membrane according to
the shape and loading conditions. It is to be noted, incidentally, that the membrane
solution is, of course, a state in an elastic shell in which there are deflections result-
ing from the strains corresponding to the stress state.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:21:06
08.qxd 15/11/2001 15:51 Page 166

166 Historic concrete

The invention of reinforced concrete provided the structural engineer with a


material which could be formed into a thin shell of any required geometric shape
appropriate to its principal loading, thereby producing an efficient primary
internal direct stress system. At the same time the reinforced shell membrane had
the capacity to resist tensile as well as compressive stresses. It could also accommo-
date the reasonable bending stresses developed in a structure which had to cope
with a range of loading and support conditions.
In many ways reinforced concrete mimics bone, the basic biological structural
material. The evolution of animals required the development of highly strength to
weight ratio efficient structural forms capable of coping with the range of loading
conditions which arise from movement. Efficiency is achieved by using internal direct
stresses as much as possible and reducing stress variations due to bending. It is
an interesting structural phenomenon that two systems have evolved:
(a) an articulated skeletal structure which can arrange its geometry to establish
primarily direct stress systems under a given load arrangement;
(b) a shell form in which external loading is resisted — primarily by internal direct
stresses — due to its shape.
With the arrival of reinforced concrete, engineers could make use of this latter
form of natural development and apply the structural shell in building construc-
tion with increasing success.
Theoretical work on the bending effects in shells was reported by Love3 at the
end of the last century, and it is from this work that most of the modern analysis
has developed, but initially its practical application was restricted to the design of
spherical domes where mathematical solutions were more easily obtained.
The early shapes of shell tended to follow the forms which had been used for
masonry construction; however, the necessary support to resist spread of the
boundary, previously provided by ring chains, flying buttresses and massive walls,
were now able to be incorporated within tensile reinforced concrete boundary
members. As a result such structures became very much lighter in construction
than their masonry equivalents.
The need for rectangular plan forms with as large as possible column-free areas
for the majority of buildings soon led to the end-only supported cylindrical bar-
rel vault. For modest spans the section of a cylinder could be used with slight thick-
ening at the edges or in the valleys to accommodate tensile stresses (Figure 8.1).
For larger spans, however, the thin shell vault was made integral with a pair of
edge beams spanning the full length (Figure 8.2). In simple terms these both form
hollow beams, the compression ‘flange’ being the crown of the shell membrane
itself and the tensile flange being the valley portion or the edge beams. However,
in both cases there is an incompatibility between the ‘membrane’ stress and strain
state of the shell at its boundaries (which is ‘expecting’ the remainder of the full
cylinder to exist), and that actually provided by the valley connection or the edge
beams. This incompatibility can only be accommodated by edge bending effects

Figure 8.1 Cylindrical shells


with thickened edges (May
and Baker canteen,
Dagenham), 1949.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:21:06
08.qxd 15/11/2001 15:51 Page 167

The early development of reinforced concrete shells 167

Figure 8.2 Cylindrical shells


with edge beams (Karachi
Hangar), 1947.

Figure 8.3 Cylindrical shell


edge effect stress resultants.

in the shell membrane which will also produce shear and torsional influences on
the edge beams themselves (Figure 8.3).
A whole panoply of theoretical studies was made of this problem, starting with
the first practical approximate design analysis provided by Finsterwalder,4

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:21:06
08.qxd 15/11/2001 15:51 Page 168

168 Historic concrete

he assumed simple end support, using a Fourier series, and omitted end moments
and twisting moments. Studies by Dischinger5 extended the analysis to include
previously omitted stress resultants, whilst Schorer6 and Vlasov7 introduced further
simplifications to reduce the extent of design calculations. Jenkins8 showed that
a consistent general theory could be developed in a simple form including all the
stress resultants. In all these cases the shell is described by eighth order partial
differential equations, which accounts for a certain mathematical complexity. (Note
that the standard analysis of an elastic beam requires a fourth order ordinary
differential equation relating its deformation to its loading.)
Concurrently with the use of thin concrete shells for forming roofs, complete
cylindrical shapes were used for liquid containment structures. The theoretical
analysis of such structures was well established early on and is comprehensively
set out in, for example, the work of Timoshenko.9
More exotic shell geometries were also being adopted with the realization that
the primary stresses did not necessarily have to be compressive when using
reinforced concrete. Doubly curved shells of negative Gaussian curvature started
to be used; these are surfaces with the centres of curvature, in two directions at
right angles, on different sides of the surface (Figure 8.4). They were adopted
particularly in the form of the hyperbolic paraboloid, which could be cut into
various ground plan shapes for roofs (Figure 8.5) or in total annular forms for
cooling towers (Figure 8.6). These presented the theoreticians with a new range
of analytical problems to solve. Other geometries such as elliptic paraboloids, cones,
conoids (Figure 8.7) and translational surfaces have also been extensively used.
Another popular arrangement has been the cylindrical shell formed as a north-
light system (Figure 8.8), whilst various shapes of dome can be cut to give a rect-
angular planform (Figure 8.9).
The history of the thin concrete shell in the UK shows a slower initial progress
than in some other countries. There are perhaps three reasons for this. Firstly, by
the time that reinforced concrete had appeared as a practical competitive struc-
tural material, Britain was well into an age of metallic construction. Steel was abun-
dantly available and had a firm grip on major structural work, and its ability to
carry high stresses both in tension and compression in skeletal forms, which could
be readily adapted to virtually all spanning requirements, gave it an undoubted
lead. This is evidenced perhaps by the fact that the Concrete Institute, as a forum
for developing reinforced concrete, was formed in 1908 outside the mainstream
professional body of the Institution of Civil Engineers.
Secondly, the familiarity of steel implied lower costs in construction since
reinforced concrete would require the retraining of the labour force, an increase
in levels of specialist supervision of work, and longer construction periods due to
the need to build complex formwork as well as the fixing of reinforcement and
the casting and curing of the concrete itself. Indeed, despite early work on water
to cement ratio theory and concrete creep, the design of really consistent high
strength concrete only became well understood in the late 1930s.

Figure 8.4 Negative Gaussian


curvature surface.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:21:06
08.qxd 15/11/2001 15:51 Page 169

The early development of reinforced concrete shells 169

Figure 8.5 An arrangement


of hyperbolic paraboloids on
a rectangular planform.

Figure 8.6 Annular


hyperbolic paraboloid
surface.

Figure 8.7 Conoid surface.

Thirdly, structural engineering in Britain during the early part of this century
was not noted for its adoption of advanced theoretical procedures which new forms
of structure required. It is true to say that, up to the Second World War, the teach-
ing of civil engineering structural analysis, although it had encompassed reinforced
concrete, was essentially restricted to two-dimensional analysis of skeletal frames,
and even the theory of flat plates did not form a part of undergraduate studies.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:21:06
08.qxd 15/11/2001 15:51 Page 170

170 Historic concrete

Figure 8.8 Northlight


arrangement of cylindrical
shells.

Figure 8.9 Doubly curved


surface on a rectangular
planform.

This is reflected in the contents of one of the most influential textbooks of the
period by Pippard and Baker.10 Important pioneering theoretical structural
research was indeed taking place in Britain, but this was almost entirely restricted
to aeronautical applications and the tools readily available to civil engineering
designers and consultants were limited; in addition, strict building rules required
full analysis of proposed structures, especially unusual forms. The result was that
novel structures in new materials were few and far between. On the continent of
Europe the steel industry had not acquired such a hold on construction, labour
was generally cheaper and the results of structural research were being applied
more rapidly in the construction industry. Shell development therefore went ahead
much more quickly there.
One of the earliest descriptions of a reinforced concrete shell roof is that of the
Armee-Museum in Munich, by Zollner in 1906,11 which was soon followed by
others at Düsseldorf and Leipzig in the following year. Many more domes were
built in Germany during the next few years, with accompanying theoretical
studies of stress analysis and, in 1915, a study by Zoelly12 in Switzerland of shell
buckling. Although few, if any, reinforced concrete domes were being built in
Britain, a major structure consisting of three shells 15 cm thick, giving a free floor

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:21:06
08.qxd 15/11/2001 15:52 Page 171

The early development of reinforced concrete shells 171

Figure 8.10 Multiple domes


of the hangar at Reval,
Russia, 1917.

Figure 8.11 Dome of the


Frankfurt electricity works,
1928.

Figure 8.12 Octagonal


domes of the Leipzig Market
Hall, 1929.

area of 115 m by 50 m, was designed in Christiani and Neilson’s London office and
built in Russia in 1917 (Figure 8.10). An article describing it appeared in The Builder
in 1920,13 but it did not apparently excite sufficient interest to merit a report in
either engineering institution journal. It is likely that the analysis was carried out
on the basis of membrane stresses. A very large, very thin shell, 26 m in diameter
and 4 cm thick, was built for an electricity works at Frankfurt (Figure 8.11) in 1928,
whilst the octagonal domes of the Leipzig Market Hall (Figure 8.12) of 1929 gave
a free floor area of 237 m by 75 m using shells 14 cm thick.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:21:06
08.qxd 15/11/2001 15:52 Page 172

172 Historic concrete

The Zeiss Dywidag system of shell construction was patented in Germany in


1920, and in 1926 Dischinger and Finsterwalder14 described one of the first long
cylindrical shells built for the Dusseldorf Exhibition. In 1927 the Frankfurt Market
Hall (Figure 8.13) was constructed with long shells spanning 42 m. Thereafter long
cylindrical reinforced concrete shells became a familiar constructional form on the
Continent, with major shell structures being built for the Market Hall at Budapest
and at Rheims railway station. In 1936 Torroja15 built the large Fronton Recoletos
Hall in Madrid (Figure 8.14). This consisted of two cylindrical vaults of different
radii which shared one longitudinal edge but were continuously supported on their
outer edges, providing a free area of 55 m by 35.5 m. The principal spanning was
therefore in the radial direction. It seems it was designed essentially on the basis
of the membrane theory plus the use of a one-tenth structural model which was
tested to failure.
In the early 1930s some interest developed in France in the analysis and design
of hyperbolic paraboloid shells and short cylindrical shells, reported mainly by
Valette16 and Laffaille.17 However, this did not seem to lead to major developments.
It is reported that the first contract in Britain for a thin concrete shell was let in
1936 for Doncaster Municipal Airport.

Figure 8.13 Cylindrical shells


of the Frankfurt Market Hall,
1927.

Figure 8.14 Edge-supported


cylindrical shells of the
Fronton Recoletos, Madrid,
1936.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:21:06
08.qxd 15/11/2001 15:52 Page 173

The early development of reinforced concrete shells 173

In the 1940s the advantages of reinforced concrete shells as a system econom-


ical in its use of materials were seen in South America, where the relative costs
of labour and materials were more favourable. Candela18 adopted the technique
for a wide range of structures of many different geometries. He appears to have
supplemented the membrane solutions with models and full scale tests.
The reconstruction after the devastations of the Second World War required forms
of building which offered economy of material. This gave an enormous boost to
the use of shell roofing in Britain as well as continental Europe, since materials,
particularly steel, were in short supply everywhere. It stimulated further struc-
tural research in British universities, at the Building Research Station and the
Cement and Concrete Association, and caused the rapid adoption of research
results into design office practice. Many simplified design procedures were pro-
posed and textbooks on shell theory and design became available in English.
The Zeiss Dywidag system was licensed to Chisarc and Shell ‘D’, who were
responsible for a number of structures. Twisteel, under the guidance of Hajnal
Konyi, were also major designers and reinforcement suppliers who produced many
hundreds of shells. Blumfeld19 described ‘The development and use of barrel vault
shell concrete’ in one of the first important papers on the subject for the Institution
of Civil Engineers. Ove Arup and Partners made a major contribution both in design,
for example the factory at Brynmawr (Figure 8.15) which had nine rectangular
planform domes each 25 m by 19 m, and in the theoretical analysis.8
Amongst the protagonists of shell construction in post-war Britain was the
architect Mills,20 who was responsible for a range of structures, and the structural
engineers Cousins21 and Snow.22
The mystique which had surrounded shell theory and its application to design
was rapidly dispelled in the late 1940s and early 1950s by the reporting of detailed
analyses of significant new structures. For example, Kirkland and Goldstein23
described the substantial Bournemouth Bus Garage (Figure 8.16) with its clear
span of 45.5 m, and Sexton24 described the earlier Karachi Hangar (Figure 8.17),
with a clear span of 39 m. In both these cases the edge beams were prestressed.
To aid the spread of the technology a number of evening courses for practising
engineers on shell theory and design appeared at this time.
The great post-war interest in shell roofing led the Cement and Concrete Asso-
ciation to hold the first symposium in 1952 devoted exclusively to the subject. This
reports, in considerably more detail than is possible here, what were seen at that

Figure 8.15 Rectangular


planform domes of Brynmawr
rubber factory, 1952.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:21:06
08.qxd 15/11/2001 15:52 Page 174

174 Historic concrete

Figure 8.16 Cylindrical shells


with prestressed edge beams
used for the Bournemouth
Bus Garage, 1951.

Figure 8.17 Cylindrical shells


with prestressed edge beams
used for the Karachi Hangar,
1947.

time as the major achievements to date, the constructional techniques currently


adopted and the state of development and understanding of theory. A fairly com-
prehensive list of publications on shells, up to 1952, is included in the proceed-
ings of this symposium.25 This first specialist meeting was followed by a second in
Norway in 1957,26 which led to the formation of an international organization for
the study and promotion of shell structures, the International Association for Shell
Structures (IASS). This organization continued to hold regular meetings on aspects
of shell studies.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:21:06
08.qxd 15/11/2001 15:52 Page 175

The early development of reinforced concrete shells 175

References 1. Mainstone, R.J., Developments in Structural Form. Alan Lane: London, 1975,
Chapter 5.
2. Lame, G., Clapeyron, E., Memoire sur l’équilibre intérieur des corps solides
homogines. Mem. Pres. Par Div. Savants, (1828), 1833, 4, 465–562.
3. Love, A.E.H., A Treatise on the Mathematical Theory of Elasticity. Cambridge
University Press: Cambridge, 1892.
4. Finsterwalder, U., Die Querversteiften zylindischen Schalangewolbe mit kreisseg-
mentformigen Querschmitt. Ingenieur Archiv., 1933, 4, 43–65.
5. Dischinger, Fr., Die strenge Theorie der Kreiszylinderschale in ihrer Anwendung auf
die Zeiss-Dywidag-Schalen. Beton Eisen, 1935, 34, 257–94.
6 Schorer, H., Line load action on cylindrical shells. Am. Soc. Civ. Engrs Trans., 1936,
62, 767–810.
7. Vlasov, V.Z., Thin walled elastic beams, 1st edn, 1940. General Theory of Shells and
its Application in Engineering, 1949 (Moscow, Stroidzal).
8. Jenkins, R.S., Theory and Design of Cylindrical Shell Structures. O.N. Arup: London,
1947.
9. Timoshenko, S., Theory of Plates and Shells. McGraw-Hill: New York, 1940.
10. Pippard, A.J.S., Baker, J.F., Theory of Structures. Arnold: London, 1936.
11. Zollner, L., The reinforced concrete dome of the Armee-Museum at Munich. Bauzeitung,
1906, Nos 16–17.
12. Zoelly, R., The Buckling of Shells, Thesis. ETH: Zurich, 1915.
13. The Builder, 1920.
14. Dischinger, Fr., Finsterwalder, U., The Dywidag hall at the Hygene Exhibition,
Dusseldorf. Bauingenieur, 1926, 1, 48, 929.
15. Torroja, E., Report on thin slabs in Spain. IABSE Final Report, Third Congress, 1948,
575–84.
16. Valette, R., Thin self-supporting roofs. Genie Civ., 1934, 104(4), 85–88.
17. Laffaille, B., Thin shells in the shape of hyperbolic paraboloids. Genie Civ., 1934,
104(18), 409–10.
18. Faber, C., Candela: the Shell Builder. Architectural Press: London, 1963.
19. Blumfield, C.V., The development and use of barrel vaults concrete. Institution of Civil
Engineers Structural Division, 1948.
20. Mills, E.D., Reinforced concrete shell membrane structures. Archit. Bldg News, 1944,
February, 94–98.
21. Cousins, H.G., Shell concrete construction. RCA Technical Paper No. 6, 1948, 32.
22. Snow, F.S., Shell concrete construction. Struct. Engr, 1947, 25(7), 265–86.
23. Kirkland, G.W., Goldstein, A., The design and construction of a large span prestressed
concrete shell roof. Struct. Engr, 1951, 29, April, 107–27.
24. Sexton, C.G., Prestressed reinforced concrete hanger at the civil airport of Karachi.
J. Instn Civ. Engrs, 1947, 29(7), 109–30.
25. Proceedings of a Symposium on Shell Structures. Cement and Concrete Association:
London, 1952.
26. Proceedings of the Second Symposium on Shell Structures, Oslo. Cement and Concrete
Association: London, 1957.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:21:06
09.qxd 16/11/2001 15:43 Page 177

9 Concrete shell roofs, 1945–65

Robert Anchor
Synopsis In the period 1945–65, considerable numbers of reinforced concrete shell roofs
were designed and constructed for relatively routine jobs. This chapter considers
the commercial background to this activity, and describes the layout of typical struc-
tures. The detailing of the concrete structure and the reinforcement is explained
and examples of typical completed structures are given.

Introduction Shell concrete was used before and during the Second World War for ‘one-off ’
roof structures and also for power station cooling towers, but this chapter is concerned
with the period between 1945 and 1965 during which shell roofs were used widely
as a method of roofing over comparatively routine buildings. In the conditions
prevailing in the UK after the Second World War, steel was only available through
a rationing system devised by the government. This shortage, together with the
need to replace war-damaged buildings, led to the use of concrete shell roofs.
A roof designed as a shell would use less steel than the alternative, a steel truss
roof. Shell construction provided a bonus, as the roof covering was also created.
The marketing of concrete as a material reached unsurpassed heights in this period
due to the efforts of the Cement and Concrete Association. Fashion also played a
part in design, and no self-respecting architect at this time would be without a shell
roof job. Typical uses were roofs for school halls, workshops, canteens, market
buildings, garages, factories and swimming pools.
The description which follows is largely based on the author’s own experiences
as a senior engineer and later as Technical Director of GKN Reinforcements Ltd
(formerly Twisteel Reinforcement Ltd) during the period under review. Other
information has been obtained from Chronowicz1 and British Reinforced Concrete
Engineering Co. Ltd (BRC).2 The diagrams are taken mostly from booklets which
were printed at the time for sales purposes, both by Twisteel Reinforcement Ltd3
and by BRC.2

The postwar It is useful to review the commercial background to the shell era. In the late 1940s,
the design and construction of reinforced concrete building structures was organ-
construction ized rather differently from the present. Traditional firms of structural consult-
industry ing engineers were largely experienced in structural steelwork and masonry design
(with notable exceptions in London). Reinforced concrete design for routine jobs
was substantially in the hands of specialist commercial firms. Among these firms
were the suppliers of steel reinforcement materials (bars and welded fabric) who
operated structural design offices. For any particular project, they offered quota-
tions to architects for the design of reinforced concrete structures, together with
the supply of the necessary reinforcement. Such competitive quotations from nom-
inated suppliers were incorporated into the bills of quantities as prime cost items.
The reinforcement details for the concrete structures were provided by the supplier’s
office, and such firms became the source of highly experienced reinforced concrete
designers and detailers who were in great demand in later years when consulting
firms expanded into reinforced concrete design. Twisteel Reinforcement Ltd, which

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:22:16
09.qxd 15/11/2001 15:52 Page 178

178 Historic concrete

changed its name to GKN Reinforcements (GKNR) around 1955, and BRC were
two such major firms, each with a similar organization. The job records of BRC
have been destroyed, but a copy of the 13th edition of its sales booklet survives.2
Examples of their work can be seen in the illustrations to the book by Chronowicz1
and in the BRC book.2 There were several other reinforcement firms operating
at this time on a more regional basis, but these firms did not design shell roofs in
any quantity. Twisteel established a structural design department in 1934 at New
Malden, Surrey, and the department was finally closed by GKNR in 1979 when spe-
cialism in structural design faded away. In the 1960s, GKNR had works in London,
Birmingham, Wigan, Belfast and Glasgow, but also supplied steel directly from
the rolling mills in Cardiff. Design offices were located in London, Birmingham,
Manchester, Southampton, Belfast, Glasgow, Cardiff and Middlesbrough. At the
time of maximum output, the total technical drawing office staff exceeded 300.
All the drawings prepared by GKNR for both shells and other jobs are still avail-
able on microfiche and have been donated to the ICE archive. There are records
showing that in August 1944 (during the Second World War), a total of 25 shell
designs had been prepared by C.V. Blumfield Consultants Ltd on behalf of Twisteel.
Subsequently, Twisteel traded in the field of shell design as BVR (Barrel Vault Roof )
Designs Ltd, and Dr Hajnal-Konyi was retained for some years from 1945 to advise
on shell design. He later became a consultant in London and designed further
shells on his own account, a particular example being a filling station canopy at
Markham Moor on the A1 trunk road.
Twisteel and its successor (henceforth both referred to as GKNR) made avail-
able a shell design service throughout the UK but with special expertise in offices
in London and Birmingham. In the period between 1948 and 1960, when shell
construction was at its peak, it is believed that about 85% of all shell designs were
prepared by GKNR and BRC. An estimate of the proportion of shell jobs under-
taken by each of the two companies is not possible with any certainty, but, from
discussions with those involved at the time, it is probable that GKNR designed
about 50% of all shells and BRC about 35%.
From the records, it appears that a total of about 2500 shell schemes were
prepared by GKNR. In one period of 6 months during 1949 a total of 212 shell
designs were prepared in their offices, but by 1960 activity had reduced consid-
erably. A study of the references to shell structures in the bibliography obtainable
from the Institution of Structural Engineers Library shows that the period 1951–70
was the peak for articles, journals and books. There was a total of 179 references
in this period. The theory followed the practice.
A second type of expert existed at this time. The general contractors of
the period had little expertise in reinforced concrete frames, and the erection
of the reinforced concrete structures for a building was often sublet to a firm
with specialist experience. Well-known names at this time were Truscon, Laing,
F.C. Construction Co., Peter Lind, Christiani & Nielsen, Tileman, Holst, Caxton
and others. Again, the specialist contractors often had substantial design organiza-
tions, and would tender for the design and construction of reinforced concrete
structures.

Shell layouts The majority of shell roofs designed in the period 1945–65 were of single curva-
ture in the form of part of a symmetrical cylinder or BVR (Figure 9.1). Many jobs
featured multi-bay construction and sometimes continuous spans. The span of the
shell (i.e. in the linear direction) could be up to 150 ft (45.7 m) and the width in
the curved direction was typically half the span providing a column layout with a
ratio of 2 : 1 (Figure 9.2). Typical sizes are shown in Table 9.1. These dimensions were
chosen, together with a suitable radius of curvature, to avoid any tendency for the
shell to buckle under compressive forces at the crown, and to limit the slope of

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:22:16
09.qxd 15/11/2001 15:52 Page 179

Concrete shell roofs, 1945–65 179

Figure 9.1 Barrel vault roof


(BVR).

Figure 9.2 Leading


dimensions of BVR.

Table 9.1 Typical layout dimensions of BVRs*

Span Breadth Rise Radius Thickness

ft m ft m ft m ft m in mm

180 54.9 50 15.2 18 5.5 40 12.2 3.0 76


140 42.7 40 12.2 14 4.3 35 10.7 3.0 76
100 30.5 50 15.2 10.5 3.2 35 10.7 3.0 76
80 24.4 40 12.2 8.25 2.5 35 10.7 3.0 76
60 18.3 30 9.1 6.0 1.8 30 9.1 2.5 63

*The values given in this table are taken from GKNR;3 BRC recommended very similar values.2

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:22:16
09.qxd 15/11/2001 15:52 Page 180

180 Historic concrete

Figure 9.3 Outer bay of


barrel roof.

Figure 9.4 (a) Upstand and


(b) downstand stiffening
beams.

Figure 9.5 Details of:


(a) feather edge valley;
(b) downstand valley beam.

the upper surface of the shell springing to about 40°; otherwise freshly placed con-
crete would tend to move down the slope. Intermediate columns were provided
under the beams at any external edge in a range of shells (Figure 9.3). This had
the effect of preventing undue lateral movement. The chord beams forming the
span supports were usually of solid concrete down to a level near to the column
heads. Alternatively, a frame with a curved beam could be used (Figure 9.4).
The overall depth of construction from the soffit of the valley beam to the crown
of the shell was usually one-tenth of the span. For spans over 100 ft (30.5 m), pre-
stressing was sometimes used. The valley details between adjacent bays of shells
could be ‘feather-edged’, flat, or could incorporate a downstand beam (Figure 9.5).

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:22:16
09.qxd 15/11/2001 15:52 Page 181

Concrete shell roofs, 1945–65 181

Figure 9.6 (a) Upstand edge


beam; (b) dropped edge
beam.

Figure 9.7 BVR showing


typical arrangement of roof
lights.

Similarly, the external edge beams could be upstand or downstand with respect
to the shell (Figure 9.6). Roof lights were commonly provided in BVR jobs, usu-
ally of circular shape (Figure 9.7). It was also possible to provide a continuous
rooflight along the crest of each shell (Figure 9.1). In this case, the two halves of
the shell were strutted apart between edge thickenings.
A popular second type of singly curved shell was the northlight roof (Figure 9.8).
Northlight shells had glazing incorporated on a straight sloping face and shells
connecting the top of one slope to the bottom of the next (Figure 9.9). The upper
and lower shell edges were propped by reinforced concrete posts at about 10 ft
(3.05 m) centres. This arrangement allowed nearly uniform natural lighting in
the work-space below. The column layout was usually in the ratio of between 2 : 1

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:22:16
09.qxd 15/11/2001 15:52 Page 182

182 Historic concrete

Figure 9.8 Northlight shells.

Figure 9.9 Example of


northlight glazing.

and 3 : 2, and spans of 50 ft (15.2 m) to 90 ft (27.4 m) were often used. Typical sizes
as used by GKNR are given in Table 9.2,3 and sizes that were recommended by
BRC are given in Table 9.3.2
A development towards the end of the main shell era was the use of doubly
curved shells, and in particular hyperbolic paraboloidal shapes (see Figure 8.5).4
An advantage of this shape (previously used vertically in power station cooling tow-
ers) was that the formwork could be formed with straight timbers laid at an angle

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:22:16
09.qxd 15/11/2001 15:52 Page 183

Concrete shell roofs, 1945–65 183

Table 9.2 Typical layout dimensions of northlight roofs, as recommended


by Twisteel3

Span Breadth Rise Radius Thickness

ft m ft m ft m ft m in mm

60 18.3 40 12.2 16 4.9 40 12.2 3.0 76


50 15.2 33 10.1 14 4.3 40 12.2 3.0 76
40 12.2 27 8.2 12 3.7 35 10.7 2.5 63
30 9.1 20 6.1 10 3.0 30 9.1 2.5 63

Table 9.3 Typical layout dimensions of northlight roofs,*


as recommended by BRC2

Span Breadth Radius Thickness

ft m ft m ft m in mm

80 24.4 30 9.1 20 6.1 3.0 76


60 18.3 20 6.1 15 4.6 2.5 63
50 15.2 20 6.1 15 4.6 2.5 63
40 12.2 20 6.1 15 4.6 2.5 63

*The rise is not given as it may depend on the window opening.

to the edges. These shells were used singly, with two low corners and provision
for resisting the outward thrust, and also to create a canopy by using three or four
shells supported by a central column. Groups of these elements were used typi-
cally over petrol filling stations.

Building control The concrete code of practice in these years was CP 114: 1948; it had no particu-
lar provisions for the design of shell roofs. Applications for bye-law approval
were dealt with variously. In small towns, the authority would rather not know! In
other areas, a certificate that the design was in accordance with CP 114 was required
and willingly given. To the author’s knowledge, no fire tests were ever made on
shell roofs, and under the regulations no period of fire resistance was required;
but a shell structure clearly has reasonable resistance to collapse in a fire situation.

Design The evolution of mathematical shell design has been dealt with by Morice.4 The
bulk of the single curvature shells designed by GKNR used a system evolved by
Tottenham5 and Bennett.6 Shell designs were originally prepared by a special
group in London, but after about 1954, designs were also prepared in the
Birmingham office. The design tools were then limited to slide-rules, eight-figure
logarithm tables and mechanical calculating machines (which were made for
accounting purposes). Shell structures tended to be designed and detailed by the
more experienced members of the design team. At BRC, shell designs were largely
prepared at the main design office at Stafford under the direction of A.P. (Pop)
Mason and A. Chronowicz.1 The shells designed by both firms were similar, but
in detail family resemblances could be noticed.
The thickness of all types of shells was generally 21⁄2 in (63 mm) in order to accom-
modate the layers of reinforcement with cover of 1⁄2 in (13 mm) on each face and
to minimize the dead load. An allowance of 15 lb/ft2 (0.75 kN/m2) on plan area was
made for imposed loading together with a suitable allowance for any finishes. Occa-
sionally the general shell thickness was increased slightly in order to reduce the
compressive or shear stresses.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:22:16
09.qxd 15/11/2001 15:52 Page 184

184 Historic concrete

At the valleys along the sides of each bay the shells were thickened to 5 in
(127 mm) to assist in resisting the lateral bending moments. The negative lateral
bending moment in a shell is a maximum at the valleys and the maximum posi-
tive moment is at the crown of the shell. At the end stiffening beams, the shells
were thickened to 31⁄2 in (89 mm) on the upper surface for a distance of about one-
tenth of the span from the end beams to enable the shear reinforcement and the
L-bars connecting to the stiffening beams to be accommodated.

Detailing Detailing followed the standard procedures of the time, although double elephant
sized paper (1016 mm  686 mm) would not always accommodate the developed
plan of a shell. Copying drawings and pages of calculations using dye-line machines
was relatively easy and usually part of the new graduate’s induction course — after
he (there were no girl graduates) had learned how to fold drawings correctly. The
reinforcement in a shell generally consisted of two layers of welded mesh placed
at the top and bottom faces and with suitable flying ends to enable overlaps to be
made without increasing the number of layers to be accommodated. Special weights
of welded fabric were available (Tables 9.4 and 9.5) to suit shell construction.
Longitudinal 3⁄8 in (9.5 mm) high-tensile square twisted bars were placed at 12 in
(305 mm) centres alternately in the top and bottom faces inside the layers of
fabric. This provided steel to resist shrinkage cracking. Due to the shape of the
shells, service cracking was not a serious problem. Diagonal bars were placed across
each of the four corners of each bay of a shell to resist shear stresses. The main
reinforcement resisting the tensile force due to the span bending moment was
placed in the downstand beams at the valleys. Cold-worked square twisted bars
were used in the earlier shells, with a transition to ribbed cold-worked bars later
on. The working tensile stress was either 27,000 lb/in2 (189 N/mm2) or 30,000 lb/in2
(210 N/mm2). This reinforcement often occupied between three and five layers in
depth, and as the beam length (which was the span of the shell) could be 100 ft

Table 9.4 Wireweld high-tensile barrel fabrics

Fabric Size of Gauge* of Gauge* Area of wires per Laps (in) Diameter of Diameter
no. mesh longitudinal of cross foot width (in2) longitudinal of cross
(in) wires wires Longitudinal Lateral wires (in) wires (in)
Longitudinal Cross
wires wires

216 12  12 5 5 0.0353 0.0353 12 12 0.212 0.212


217 6  12 6 10 0.0580 0.0129 11 6 0.192 0.128
309 6  12 4 9 0.0846 0.0163 14 9 0.232 0.144
428 6  12 2 8 0.1196 0.0201 16 9 0.276 0.160

*Birmingham wire gauge.

Table 9.5 Wireweld high-tensile barrel fabrics (metric version of Table 9.4)

Fabric Size of Area of wires per Laps (mm) Diameter of Diameter of


no. mesh metre width (mm2) longitudinal cross wires
(mm) Longitudinal Lateral wires (mm) (mm)
Longitudinal Cross
wires wires

216 305  305 75 75 305 305 5.4 5.4


217 152  305 123 27 280 152 4.9 3.3
309 152  305 179 35 356 229 5.9 3.7
428 152  305 253 43 406 229 7.0 4.1

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:22:16
09.qxd 15/11/2001 15:52 Page 185

Concrete shell roofs, 1945–65 185

(30 m) or more, normal spliced overlaps were not possible. The usual detail was
of bars butted together along the span with the positions of the butt joints stag-
gered by at least two lap lengths. Extra bars were added throughout the span to
the theoretical number required so that the required full tensile strength could
be maintained throughout the beam length.

Shell construction In the early days, shell concrete construction required a certain amount of courage,
and in order to give confidence to clients, architects and contractors, a loading
test was carried out by the Building Research Station on a shell roof at Liverpool.
The report said inter alia that: ‘The maximum deflection of the roof at mid-span
under an applied load of 150% of the design live load was 1⁄8 in (3 mm) only. This
is about 1/6000 of the span’. The erection of formwork to a cylindrical shape and
later to a doubly curved shape was not necessarily the most profitable type of work.
However, there was always a feeling of pride (and relief) after satisfactory com-
pletion of a job. Removal of the formwork, starting at mid-span, could be a tense
moment. Formwork was usually of plywood, but occasionally boards were used,
laid on shaped timber bearers in turn supported by metal scaffolding. The verti-
cal tubes were adjusted to height with a screw fixing, and the lateral pre-curved
tubes were often referred to as ‘banana bars’.
The normal concrete for shell roofs was a 1 : 11⁄2 : 3 mix by volume with a required
28-day strength of 3750 lb/in2 (26 N/mm2). The actual shell section had 3⁄8 in
(9.5 mm) maximum-sized aggregate, and the concrete required a workability which
was not too workable at the shell springings where the top surface was at an angle
of about 40°. At the thin portion of the shell greater fluidity was necessary in order
to work the concrete around the reinforcement in the depth available. Concrete
was always mixed on-site, and it was the expertise of the mixer driver which
ensured a satisfactory job. The water content of the mix was determined by eye,
but the results were surprisingly accurate with an experienced and willing opera-
tive. Consolidation of the newly placed concrete was achieved by tamping with
timber poles, which also required expertise and dedication. Vibrators were becom-
ing available but could not be used easily on a shell only 21⁄2 in (63 mm) thick.
Shells were usually finished on the interior face with a layer of insulating board
which could be painted. For open structures (i.e. without walls) the shell soffit was
either painted or left as struck. On the top surface of the roof, two layers of roof-
ing felt were used (Figure 9.10). This provided waterproofing and also allowed
some elasticity against shrinkage movement. Cracking in service was not gener-
ally a problem, and although carbonation of concrete was not seen as a problem
in 1960, the finishes effectively prevented or delayed degradation.

Examples of jobs Illustrations of actual jobs designed by BRC are given by Chronowicz1 and in
the BRC Handbook.2 In particular, the Pannier Market at Plymouth, the Cattle
Market at Gloucester, and the Terylene Works for ICI at Wilton, Teeside.

Figure 9.10 Section of shell


showing finishes.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:22:16
09.qxd 15/11/2001 15:53 Page 186

186 Historic concrete

The following descriptions are of jobs designed by GKNR in the years 1950–65.
This is not to suggest that other firms did not design shell roofs of consequence,
but the information available to the author is entirely from his work with this
company.

Cadbury Bros., Moreton

In 1951, Cadbury Bros. decided to build a completely new factory at Moreton,


Cheshire, to manufacture chocolate biscuits. It was built with concrete shell roof
construction and covered an area of 30 acres (12 ha), and is shown in Figure 9.11.
The factory buildings were of northlight construction in order to prevent sunlight
entering the working area, and the boiler house and the canteen were covered
with normal barrels. The shells were generally of about 50 ft (15 m) span and 25 ft
(7.6 m) width. H block, which was constructed in 1958, was used for printing card-
board packaging. The raw materials were stored at ground level, over which the
printing floor was constructed of a 12 in (305 mm) thick flat slab floor spanning
between columns at 25 ft (7.6 m) centres each way (Figure 9.12). The client
requested a 50 ft (15 m) square layout of columns above first-floor level, to fit in
with the machinery layout. This was achieved by providing a series of 50 ft (15 m)
span frames, upstanding with respect to the shells. These frames were difficult to
analyse with the equipment then available, their reinforcement was difficult to
detail, and they were awkward to waterproof. However, the client’s requirements
were achieved.
In 1962, the last substantial block, for confectionery manufacture, was under
construction. It covered an area of 510 ft (155 m) by 470 ft (143 m) and was a single-
storey construction. The northlight shells again had a span of about 50 ft (15 m) and
a width of 25 ft (7.6 m). Movement joints were introduced by providing double
frames either side of the joint. The factory no longer belongs to Cadbury’s, but is
still in use and in good condition.

Figure 9.11 Aerial view of


Cadbury’s factory at Moreton,
Cheshire, in 1958.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:22:16
09.qxd 15/11/2001 15:53 Page 187

Concrete shell roofs, 1945–65 187

Figure 9.12 Northlight shell


roof of H block, Cadbury’s
factory, Moreton, Cheshire,
with column spacing
50 ft  50 ft (15.2 m  15.2 m).

Figure 9.13 Barrel vault


roofs, Kidderminster Cattle
Market.

Kidderminster cattle market

In 1961, a cattle market was constructed at Kidderminster, Worcestershire, and


roofed with normal cylindrical barrel shells with a maximum span of 100 ft (30 m),
as shown in Figure 9.13. There were three separate buildings. The pig building
was roofed with five bays of shells 35 ft (10.7 m) wide and spanning 100 ft (30 m).
The cattle building was similar. The market building has five bays 30 ft (9.1 m) wide
and is continuous over spans of 65 ft (19.8 m) and 30 ft (9.1 m). The structures still
stand in good condition.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:22:16
09.qxd 15/11/2001 15:53 Page 188

188 Historic concrete

Filling station, Harborne, Birmingham

In 1968, a petrol filling station canopy was constructed in Harborne, a suburb of


Birmingham. It consisted of three identical hyperbolic paraboloidal structures, each
supported by a central column and trimmed with upstanding beams. Each structure
obtained support against wind forces by constraints provided by small props
between the separated edges. The columns were mutually at 55 ft (16.8 m) cen-
tres. The canopy was still in good condition in 1997 when it was demolished in
order to change the layout of the forecourt.

Swimming pool, Hatfield

One of the last major jobs, designed in 1966, was a roof to cover a swimming pool
at Hatfield, Hertfordshire (Figure 9.14). It was constructed of four hyperbolic parab-
oloidal shells in a rectangular layout and contained from spreading with prestressing
tendons at the level of the heads of four supporting columns. The columns were
2 ft 6 in (762 mm) diameter spaced at 69 ft  61 ft (21 m  18.6 m). Shells 1 and 3
were each 90 ft  61 ft (27.4 m  18.6 m) and shells 2 and 4 were each 80 ft  69 ft
(24.4 m  21 m). The pairs of shells were placed so that glazed vertical triangular
spaces occurred between them to provide natural lighting. The overall size of the
roof was 180 ft  160 ft (55 m  49 m). Beams of width 28 in (711 mm) were pro-
vided along the edges of each shell, varying in depth from 24 to 12 in (610 to
305 mm). Secondary columns were provided along the external corners to limit
vertical deflections. The structure was designed for the following loads:

Imposed load 15 0.75


Finishes 12 0.60
3-in Lightweight shell 25 1.25
52 lb/ft2 2.60 kN/m2

Figure 9.14 Details of shell


roof, swimming pool,
Hatfield.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:22:16
09.qxd 15/11/2001 15:53 Page 189

Concrete shell roofs, 1945–65 189

Lightweight concrete was used for the shells in order to reduce the dead load and
also to improve the insulation. A paper on this project was presented to a Sym-
posium of the International Association for Shell Structures at Budapest in 1965.7
The job is particularly interesting in that it has recently been refurbished (1996).
During this work, it was found that the grouted prestressing cables had corroded
and they had to be replaced. However, the shells and supporting structure were
in a satisfactory condition and were retained.

Conclusion Concrete shell construction in the UK ceased on any scale around 1965 when
structural steel had become readily available, formwork costs had increased and
architectural fashion had moved on. The shells described above have survived 35
years, and as far as the author is aware, have not required undue maintainance.
Perhaps experience has a greater part to play in successful design than is presently
acknowledged.

References 1. Chronowicz, A., The Design of Shells — A Practical Approach. Crosby Lockwood, 1959.
2. British Reinforced Concrete Engineering Co. Ltd., BRC Reinforcements, 13th edn. BRC:
London, 1959.
3. Twisteel Reinforcement Ltd., Barrel Vault Roofing. Twisteel: London, 1953.
4. See Chapter 8.
5. Tottenham, H., A simplified method of design for cylindrical shell roofs. Struct. Engr,
1954, 32(6), 161–80.
6. Bennett, J.D., Some Recent Developments in the Design of Reinforced Concrete Shell
Roofs. Reinforced Concrete Association: London, 1958.
7. Anchor, R.D., Shell Roof at Hatfield. Presented in Budapest at Int. Symp. Shell Structs,
1965. Struct. Concr., 1966, 3(2), 99–104.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:22:16
10.qxd 15/11/2001 15:53 Page 191

10 Prestressing

Francis Walley
Synopsis This chapter briefly reviews the early history of prestressed concrete. It discusses
the materials used and the systems of pre- and post-tensioning in use, particu-
larly in the early days. It then describes its uses in the building field and the type
of structures in which it is likely to be found.

The early years The subject of prestressing covers a relatively new process whose inventor and chief
protagonist, Eugene Freyssinet, only died in 1962. Indeed, the major innovations
in prestressed concrete and the notable structures that have resulted, have come
about in the last 50 years.1–4 To put the subject into perspective, it is necessary to
start with the ideas and threads of knowledge that existed long before the suc-
cessful applications of prestressed concrete.
In the first decades of this century, flexural cracks in reinforced concrete played
a prominent part in the history of reinforced concrete in Germany.5 In 1906 a
treatise by Labes6 laid down the following requirement for structures on the Pruss-
ian railways: flexural tensile stresses should offer a safety factor of 1.5 –2.5 against
cracking measured as direct tensile stress. This requirement effectively precluded
the use of reinforced concrete for slabs and beams. At the same time, of course,
it provided engineers the stimulus of an obstacle that had to be surmounted. The
next year, in the same periodical in which Labes had laid down his requirement,
Koenen7 made, for the first time, the practical suggestion of giving the tension
zone a preliminary compressive stress by putting steel rods into tension before
concreting, using a stretching device that was removed after the concrete had
hardened.
As a result of this suggestion, experiments were put in hand by a German Rein-
forced Concrete Research Committee and published before the First World War.
In these experiments, two bars of 18 mm dia. were stressed and held stressed for
45 days up to the date of the test. These clearly showed that cracking was delayed
as compared with similar beams with unstressed rods, that is, an artificial tensile
strength had been given to the concrete. There was, of course, no increase in
ultimate load and those beams that were prestressed but without mechanically
anchored bars failed at about 70% of those with anchored bars. This result has
often been observed subsequently when bond failure occurs.
This early work was not followed up for various reasons. First, no one could
think of a sensible method of introducing the prestressing force and in particular
the bent-up bars that Mörsch8,9 had shown in 1907 to be necessary to resist shear.
Second, Labes,10 in the German Committee for Reinforced Concrete set up in
1907, modified his requirement by substituting the bending tensile stress for the
direct tensile stress of concrete. Third, and more important as time went on, it
was realized that calculations about factors of safety against cracking were largely
illusory since the effect of shrinkage, particularly in the case of heavily reinforced
beams, was of paramount importance. Indeed, to compensate for the shrinkage,
high-tensile stresses, and therefore high-tensile steel, would have been necessary,
and it was not available except at very high cost.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:23:34
10.qxd 15/11/2001 15:53 Page 192

192 Historic concrete

It was also found that, in some of the test beams kept for a considerable time,
the prestressing no longer affected the shape of the load–deflection curve, which
caused some puzzlement at the time before shrinkage and subsequently creep
became better-known phenomena.
Economies were sought by using higher-strength steel and higher stresses.
Unfortunately, although the elastic limit can be increased, Young’s modulus can-
not, so that stresses in the steel have to be limited to avoid large cracks in con-
crete. However, looking back historically, it was Freyssinet’s realization that creep
and shrinkage had to be overcome that enabled him to patent prestressing in 1928:
At that time, so heavy was the intellectual oppression exercised by a hand-
ful of mathematicians, obsessed with their science and blind to reality, that
an unquestioned belief in the constancy of Young’s modulus for concrete
was held, without a valid basis and in spite of many proofs to the contrary,
by all our professors and indeed by all technicians.
As a consequence, the properties of materials seemed to offer no diffi-
culty whatever to the development of my idea — in fact, quite the contrary,
since the fundamental contradiction between the elastic strains of steel and
concrete in reinforced concrete appeared all the more startling, and the
basic theory of this method of construction all the more absurd; from my
very first acquaintance with this material it seemed a very temporary stage
in the technique of associating steel with concrete.11
He had realized or suspected early on in his bridge over the Allier, in the first
decade of this century, when he used prestressed ties in arch bridges, the impor-
tance of shrinkage and creep, but without any figures as to their ultimate values
it was impossible to proceed to a definite conclusion.
By 1928 work had proceeded far at the Building Research Station on this sub-
ject, by Glanville and Faber in particular, which led to the publication of the famous
Technical Paper Nos 11 and 12. This work confirmed Freyssinet’s ideas. He real-
ized that, if he used really high stresses in steel and compressed the concrete as
high as possible, there would be sufficient prestress left in the concrete, after creep
and shrinkage had taken place, to avoid any cracking under service loads; relax-
ation of steel was not yet thought of. The reinforcement was no longer passive,
but active, so that Young’s modulus was not a stumbling block. All earlier work
tacitly assumed that the prestressing of beams would be done by rods and they
would be straight, and hence some considerable attention was paid to vertical stress-
ing of stirrups to reduce principle tensile stresses. Indeed, these were part of
Freyssinet’s patent. Although, in an addendum to the patent, anchoring by bond
was included; this was not a process adopted until exploited by Hoyer, although
Freyssinet had used it for powerline pylons in the 1930s.
The tacit assumption of using straight bars resulted in the need to use vertical
prestressing in beams and also, where possible, the shaping of the longitudinal
profile of a beam to a parabolic form to avoid high-tensile stresses in the upper
fibres at the ends of the beams. There were devices described by Mörsch to stress
additional short bars over the centre of spans, but they were clumsy.
The use of steel wire to form the tendons came in during the early years of the
Second World War, when the Freyssinet jack was invented to anchor simultane-
ously 12 wires first of 5 mm and then 7 mm dia. Magnel developed a two-wire stress-
ing system. These gave an unprecedented degree of design freedom.
One other major contribution was his research into the making of high-quality
concrete — a subject that Freyssinet pursued up to the time of his death.
Wayss and Freytag were the licencees for Freyssinet’s patents in Germany
and, with the close tie-up between universities and industry, it was possible to
integrate research and development work. More importantly, there was a
programme of work for building autobahns so that prestressing could be applied

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:23:34
10.qxd 15/11/2001 15:53 Page 193

Prestressing 193

to autobahn bridges as early as 1938, and some 10 km of bridge beams were


made. During the war years, it became available for the construction of defence
works and an enormous number of beams were used for the building of the
U-boat pens.
The one breakthrough in 1934 that thrust prestressed concrete into the lime-
light was the maritime railway station at Le Havre. This was settling and in
danger of being demolished when Freyssinet said he could save it using prestressed
concrete. The choice was difficult — allow it to collapse or try this unproven
method. The latter was chosen, and, as Freyssinet said, ‘it had the good fortune
to succeed. The risk had been great but I had broken the vicious circle which
encloses all innovators.’

Early post-war use At the end of the war, German cities were virtually totally destroyed, and little
was done for many years except clear up. In France the great need was for resto-
of prestressing ration of communications. In the UK the priority was the rebuilding of cities
and providing accommodation. In France, therefore, the priority was bridges
and five bridges over the Marne made up of precast units stressed together,
were constructed in 1946. Figure 10.1 shows a prestressed water reservoir at Tours
with offices underneath built in the late 1940s.
In Belgium, Professor Magnel was pioneering in both bridges and buildings.
Figure 10.2 illustrates a large hangar at Moelsbruke with a 300 t beam. By 1949
he could boast of some 50 jobs constructed.
In the UK, of course, there were enthusiasts. Mautner, who had escaped from
the Nazis in Germany, was a devotee of the Freyssinet system and did much for
the subject. There was, however, a curious unease, a suspicion of other people, of
infringement of patents (there were some 20 patents). It is difficult these days to
describe the emotions, almost passions, that were raised. Although the Ministry
of War Transport had had designed (by Mautner) some precast, prestressed beams
for stockpiling to replace bridges during the war — these were never used until
years later — no development on bridge works took place because there was no
road programme for many years.

Figure 10.1 Water tower at


Tours with offices below.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:23:34
10.qxd 15/11/2001 15:53 Page 194

194 Historic concrete

Figure 10.2 Hangar at


Moelsbruke.

Figure 10.3 Three-bay


single-storey
workshop.

The major development work took place in the Chief Scientific Adviser’s
Division (later the Chief Development Engineering Division) of the Ministry of
Works. It had its own Development Station, independent then of BRS where it
pressed ahead with construction and testing of various units. It let development
contracts with contractors (Costains) at Childerditch, Essex. It mounted exhibi-
tions and provided lectures all over the UK. At the same time it selected certain
buildings to demonstrate the use of prestressed concrete. Sighthill Stationery Office
was a very early example (1948). This was followed by Kilburn Telephone
Exchange (both being fully framed structures), numerous single-storey industrial
buildings (Figure 10.3) and the providing of what became the temporary office
building programme (Figure 10.4). Another notable example is the hangar at

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:23:34
10.qxd 15/11/2001 15:53 Page 195

Prestressing 195

Figure 10.4 Temporary office


building.

London Airport designed by the Prestressed Concrete Co. (see Figure 2.31,
p. 35).16
It is interesting to note that the annual UK production of prestressed flooring
units rose from zero in 1950 to 1.3 million square yards in 1960.
In tackling any history, it is difficult at certain points to decide which way to
proceed — whether to pick out salient events in succeeding years or to deal with
various technical aspects historically. The latter has been chosen as being less
confusing. The materials already mentioned hold the keys.

Steel Although high-tensile steel wire is said to have been available as early as 1908, most
of the early work used high-tensile steel of up to 14 mm dia., anchored using
wedges in steel plates. The wire before 1940 had a high-tensile strength but a very
variable elastic limit. This latter could be ‘stabilized’ if the wire was repeatedly
stretched and relaxed, and indeed Freyssinet devised a method of doing this
semi-continuously.
By 1940 high-tensile drawn wire was available. This was a high-carbon steel. In
the UK it was ‘patented’ and drawn through dies giving a total reduction of 70%
of area. This was wound off on small coils and gave no special difficulty provided
that it was used for pre-tensioned work. Although Freyssinet had patented
prestressing by bond in his addendum to his patent in November 1928, Hoyer in
Germany succeeded in obtaining a patent for a long-line system of prestressing
provided that his wires did not exceed 2 mm dia. and were not stressed more than
50%. For this type of work and this wire diameter, small coils were not a problem.
Site work was a different matter. In the early days one of the first ‘tools’ or pieces
of machinery to come on site was a wire straightening machine, without which it
was impossible to make up tendons, as the wires always wanted to return to their
small coils. Even as late as 1955, BS 2691 had the phrase ‘when so required by the
purchaser wire shall be supplied in coils of sufficiently large diameter’. It was with
the next revision that large-diameter coils that paid out straight became the norm.
The production of steel for prestressing was a highly cooperative affair, and one
that can be looked back on with some pleasure. After some initial hesitation as to

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:23:34
10.qxd 15/11/2001 15:53 Page 196

196 Historic concrete

whether prestressing was a ‘flash in the pan’, there was an enthusiastic response
from wire drawers, and a creative dialogue took place between the designers and
manufacturers, leading through heat-treated wire on large-diameter coils, to
low-relaxation steels, and then to strands.
The relaxation or creep of wire came into prominence as a result of work by
Professor Magnel at Ghent University in 1946–47.12,13 It was an aspect of
prestressing that Freyssinet, I feel, missed. Indeed, Freyssinet, as late as 1949 in
his lecture to the ICE already quoted from,11 played it down, saying that it was
not important. One of the points to which Magnel drew attention was that, by
tensioning a wire, releasing and tensioning again, the relaxation was reduced, and
on early jobs this was written into the specification. However, since Continental
European steel was made in a different way from British steel, some experimen-
tal work on relaxation was put in hand at the Ministry of Works experimental
station at Thatched Barn in 1950. These gave values for British steel that
could be used in design. They also showed that wire that had been heat-treated
and supplied in large coils had a lower relaxation and that stressing and destress-
ing did not really affect the relaxation. Consequently, this part of the specification
was omitted. Work was continued by BRE, and since that time the effect of
both low and high temperatures and radiation has been investigated in many
places.
Although wire has been referred to at length, this history would be incomplete
if it did not refer to Lee–McCall bars which, once rolled threads had been intro-
duced, gave little trouble. In the early days of cut threads the lengths of bars had
to be calculated precisely and the position of the nut was critical if one did not
wish to risk a snapped bar. As the load required for tendons increased, attention
turned to strand, and one of the first major uses was the bridge in Perth in 1959.
To give some idea of the way in which prestressing increased in use, the weight
of prestressing wire used increased from 4000 t/year in 1950 to 40,000 t/year in
1960.

Concrete The advent of prestressed concrete made it advantageous to use high-strength


concrete. Whereas in reinforced concrete it is not easy to take full advantage of
high-strength concrete, in prestressed concrete there is almost no limit to its use.
In the early days virtually the sole criterion was to obtain an adequate factor of
safety.
It was not long, however, before its other properties of low creep and shrinkage
were also appreciated. There is no doubt that prestressing provided a stimulus to
the production of better and stronger concrete.

Pre-tensioned In the case of pre-tensioned work the stress in the steel has to be transferred directly
to the concrete. In the early years (the late 1940s and early 1950s), following
concrete German practice, 2 mm dia. wire was considered to be the maximum that should
be used. (The early German rule that pre-tensioned units should not be less
than 3 m long or used for rolling loads was not adopted.) For this reason the
Stahlsaitenbeton in Germany, with its multitude of small wires, became the strangbetong
in Sweden. Eventually similar pre-tensioning beds were installed in Britain at Iver
by Holland, Hannen and Cubitt. It was considered that the bond would not be
good enough if a larger-diameter wire was used. However, people began to use
5 mm dia. wire, first with crimps and indentations and then, when these proved
satisfactory, with plain wire. In the late 1950s, the Cement and Concrete Associ-
ation carried out research work to determine transmission lengths, which were
available to the Code Committee. By 1962, 0.7 in (17.8 mm) dia. strand was being
used with a transmission length of 500 mm.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:23:34
10.qxd 15/11/2001 15:53 Page 197

Prestressing 197

Stressing systems The earliest stressing system to be used in Britain was the Freyssinet system which
used a double-acting jack. The first action of the jack was to tension eight or 12
(post-tensioning) wires. The second action was to drive home a cone which consisted of a steel tube
surrounded by high-alumina cement which locked the wires into the anchorage.
The tube was used for injecting grout after stressing had been completed. Although
appearing crude, it was an effective anchoring device. The smaller anchorages
could receive eight, 10 or 12 wires of 0.2 in dia., but there were larger ones which
could anchor twelve 0.276 in dia. wires and even twelve 0.315 in wires. These are
shown in Figure 10.5, and a typical longitudinal section at an end in Figure 10.6.
Figures 10.7–10.9 show a typical anchor block and cone, the spring spacer and a
Freyssinet jack.
These were imported into the UK by the Prestressed Concrete Co., which set
up PSC Equipment Ltd which itself developed systems largely using single wire
stressing but had available a two-wire system which was used for the Intergrid
school system. Table 10.1 shows the systems available in the 1960s.
The Magnel–Blaton system (invented by Professor Magnel of Ghent) came
shortly afterwards and depended on stressing two wires at a time but building them
up into quite large anchorages. Each anchorage was capable of receiving eight wires
but could be built up to eight in number, giving 64 wires. Originally only 0.20 in
dia. wires were used but the anchorages were later modified to receive 0.276 in
dia. wires. The distribution plate to transfer the load to the concrete was of cast
steel and the largest (64  0.276) weighed 95 lb. Details of this anchorage are shown
in Figures 10.10 and 10.11. This system was eventually marketed in Britain by
Stressed Concrete Design, which produced its own anchorage systems (Table 10.2).
The BBRV system was also introduced from Switzerland. In this system for post-
tensioning, the wires had ‘buttons’ formed at their ends to anchor them to a steel
anchorage device. This was not extensively used in the UK.

Figure 10.5 Details of


Freyssinet anchorages.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:23:34
10.qxd 15/11/2001 15:53 Page 198

198 Historic concrete

Figure 10.6 Typical


arrangement of Freyssinet
cables and anchorages.

Figure 10.7 Freyssinet


anchorage block and cone.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:23:34
10.qxd 15/11/2001 15:53 Page 199

Prestressing 199

Figure 10.8 Freyssinet


cables, showing spring
spacer.

Figure 10.9 Freyssinet jack.

The Lee–McCall (Macalloy) system was developed by Donovan Lee in associa-


tion with McCall & Co. Ltd of Sheffield in the early 1950s, and comprised an alloy
steel bar with threaded ends, with a nut which transferred the force in the rod to
a distribution plate at the end of the beam. In the early years the thread was cut,
but later, after isolated failures of bars, was rolled on. Details of earlier anchor-
ages are given in Table 10.3.
The CCL system developed from a small grip they produced for the end anchor-
age in a pre-tensioning system used by the Ministry of Works was used initially in
the Gifford–Udall–CCL systems for post-tensioning. These broke down into the
CCL system (Table 10.4) and the Gifford–Udall and Gifford–Burrow systems
(Table 10.5).

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:23:34
200

Table 10.1 The Freyssinet/PSC system


10.qxd 15/11/2001 15:53 Page 200

Type Initial Number and Duct Anchorage Trade Jack type Jack Retracted Maximum Jack
force size (in) of size size description extension jack length diameter weight
wires or strand (in) (in) (in) (in) (in) (lb)
Historic concrete

(lb  103)

Single wire 5.0 1  0.200 17⁄32  17⁄32 Mono-wire PSC 4 or 6 131⁄2–171⁄4 31⁄2 7–9
9.4 1  0.276 17⁄32  17⁄32 Mono-wire
5
9.9 2  0.200 ⁄8 dia. 13⁄16  23⁄8
5
18.8 2  0.276 ⁄8 dia. 13⁄16  23⁄8
3
19.8 4  0.200 ⁄4  3⁄4 213⁄32  213⁄32
3
37.4 4  0.276 ⁄4  3⁄4 213⁄32  213⁄32
75.2 8  0.276 19⁄16 dia. 41⁄4  41⁄2
112.8 12  0.276 2 dia. 55

12 wire 59.2 12  0.200 13⁄16 dia. 33⁄4 dia. Freyssinet Freyssinet 10 –12 30 9 109
Multi-wire Multi-wire (1953)
112.8 12  0.276 2 dia. 43⁄4 dia. Freyssinet Freyssinet 10–12 30 9 109
Multi-wire Multi-wire (1953)
146.6 12  0.315 2 dia. 6 dia. Freyssinet Multi-wire 8 30 9

IP: 130.194.20.173
5 5
Single strand 10.85 1  ⁄16 ⁄8 dia. Mono-strand Mono-strand
5
14.7 1  3⁄8 ⁄8 dia.
5

On: Tue, 17 May 2011 06:23:34


19.6 1  7⁄16 ⁄8 dia.
5

Delivered by ICEVirtualLibrary.com to:


25.8 1  1⁄2 ⁄8 dia.
58.0 1  0.7 13⁄16 dia. 43⁄4 (int. anch) Freyssinet 8 28 8 200
strand
76.0 1  7⁄8 13⁄16 dia. 55⁄8 (ext. anch)
11 15⁄8 dia.
129.0 1  11⁄8 15⁄8 dia.
103.0 7  3⁄8 2 dia. Mono-strand Mono-strand
137.0 7  7⁄16 2 dia.
181.0 7  1⁄2 2 dia.

Multi-strand 310.0 12  1⁄2 23⁄4 dia. 81⁄4 dia. Freyssinet 113⁄4 38 12 270
428.0 12  0.6 23⁄4 dia. multi-strand
10.qxd 15/11/2001 15:53 Page 201

Prestressing 201

Figure 10.10 Distribution


plate details for 0.2 in dia.
wires (Magnel–Blaton
system).

Figure 10.11 Magnel–Blaton


sandwich plate and wedges.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:23:34
202

Table 10.2 The Magnel–Blaton system


10.qxd 15/11/2001 15:53 Page 202

Type Initial force Number and size (in) Duct size (in) Anchorage Trade Jack type Jack Retracted Maximum Jack
(lb  103) of wires or strand size (in) description extension (in) jack length (in) diameter (in) weight (lb)
Historic concrete

Double 19.7 4  0.200 21⁄2  1 21⁄4  31⁄8 Magnel–Balton Mark I 10 44 67⁄8 70


wire 29.6 6  0.200 11⁄2  11⁄2 3  31⁄2 system Mark II ‘S’ 10 53 87⁄8 112
39.4 8  0.200 2  21⁄8 33⁄4  35⁄8 Mark II ‘L’ 18 69 87⁄8 168
78.8 16  0.200 2.16  1.97 5  5.1
118.2 24  0.200 2.16  2.95 6.15  6.3
157.6 32  0.200 2.16  3.94 7.73  6.7
197.0 40  0.200 2.16  4.92 9  7.1
236.0 48  0.200 2.16  5.90 10.3  7.5
277.8 56  0.200 2.16  6.89 11.4  7.9
317.2 64  0.200 2.16  7.87 12.4  8.3
37.6 4  0.276 11⁄2  11⁄2 34
56.4 6  0.276 21⁄2  11⁄4 4  43⁄4 Mark II ‘S’ 10 53 87⁄8 112
75.2 8  0.276 21⁄2  11⁄4 5  43⁄4 Mark II ‘L’ 18 69 87⁄8 168
150.4 16  0.276 21⁄2  21⁄2 53⁄4  8
225.6 24  0.276 31⁄2  21⁄2 71⁄4  101⁄4
300.8

IP: 130.194.20.173
32  0.276 5  21⁄2 91⁄2  101⁄4
376.0 40  0.276 6  21⁄2 111⁄2  105⁄8
451.2 48  0.276 7  21⁄2 131⁄2  87⁄8

On: Tue, 17 May 2011 06:23:34


526.4 56  0.276 71⁄2  21⁄2 141⁄2  111⁄2

Delivered by ICEVirtualLibrary.com to:


601.6 64  0.276 81⁄2  21⁄2 151⁄2  113⁄4
676.8 72  0.276 91⁄2  21⁄2 171⁄2  121⁄2
902.4 96  0.276 121⁄2  21⁄2 231⁄2  121⁄2

Single 77.4 3  0.5 11⁄4  21⁄2 7  31⁄2 Magnel–Blaton 12 28 37


1
⁄2 in 154.8 6  0.5 21⁄2  21⁄2 86 stress-block 20 361⁄2 47
strand 232.2 9  0.5 21⁄2  21⁄2 81⁄2  81⁄2 system
309.6 12  0.5 31⁄2  21⁄2 9  11
387.0 15  0.5 5  21⁄2 91⁄2  131⁄4
464.4 18  0.5 5  21⁄2 91⁄2  151⁄2
541.8 21  0.5 6  21⁄2 10  171⁄2
619.2 24  0.5 6  21⁄2 101⁄4  193⁄4
696.6 27  0.5 7  21⁄2 101⁄2  213⁄4
10.qxd 15/11/2001 15:53 Page 203

Prestressing 203

Table 10.3 The Lee–McCall (Macalloy) system

Type Initial Number and Duct Anchorage Trade Jack type Jack Retracted Maximum
force size (in) of size size (in) description extension jack length diameter
(lb  103) bar (in) (in) (in) (in)

Bar 60.5 1  7⁄8 11⁄2 55 Macalloy Mark VIII 6 27 71⁄2


threaded 78.5 11 11⁄2 55
101.0 1  11⁄8 11⁄2 51⁄2  51⁄2 (6  6)* Mark VII 12 36 83⁄4
123.0 1  11⁄4 15⁄8 7  6 (7  7)*

*Using wedges.

Ducts and grouting Post-tensioning implies that the tendons are stressed after the concrete has hard-
ened and therefore have to be free to move. Spacers were inserted if the tendon
was not stressed as a whole since the stressing of, say, two wires out of 12 could
lock the whole tendon. Sufficient clearances also had to be left to allow grout to
penetrate the whole length of the duct. Grouting was deemed necessary to
protect the tendon from corrosion and also to enhance the ultimate failing load
of the beam. The ultimate strength of an ungrouted beam was usually of the order
of 70% of a fully grouted one. The complete grouting of tendons has not always
been successful, even if apparently every care has been taken. This has not led to
any great troubles except where units have been exposed to weather, and in
particular to aggressive conditions such as deicing salts and often in association
with bad filling in joints in segmental construction. Where units have been ‘pro-
tected’ and subsequently demolished, the prestressing steel has been found in a
pristine state.
Pre-tensioning as a method of stressing is not discussed at length in this chapter
(units stressed by this method will be distinguished by having no end anchorages),
it is interesting to record that the CCL system, developed by modifying one of their
grips for joining cables to produce a small grip for the end anchorage in a pre-
tensioning system used by the Ministry of Works.

Stressing of Nearly all of the above systems have been modified by using opposed anchorages
to stress circular tanks. The only other system used was the American pre-load
circular tanks system which depended on stressing a continuous wire by passing it through a
die which stressed the wire and reduced its diameter. In these cases no anchor-
ages will generally be found except at the top and bottom of the tank.16

Prestressed This chapter deals only with prestressed units likely to be found in building and
concrete units general civil engineering.
Prestressing, by post-tensioning, was used for many shell structures, particularly
in edge beams but occasionally in the shell itself, and these are discussed in
Chapters 8 and 9. Prestressing both by pre- and post-tensioning has made a major
contribution to bridge building, and details of these are to be found in
Chapter 12. Prestressing as applied to maritime structures is to be found in
Chapter 13.
By far the major contribution to the building programme has been in its use
for flooring units, and these almost without exception have been produced by
pre-tensioning. In the early days (post-war) the incentive was undoubtedly the fact
that prestressing wire, unlike structural steel and reinforcement, was not rationed,
but the units proved to be very reliable and trouble-free. The fact that in general
no tension was allowed in the concrete meant that they were crack-free and
corrosion of the wires was not a problem.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:23:34
204

Table 10.4 The CCL system


10.qxd 15/11/2001 15:53 Page 204

Type Initial Number and Duct Anchorage Trade Jack type Jack Retracted Maximum Jack
force size (in) of size size (in) description extension jack length diameter weight
Historic concrete

(lb  103) wires or strand (in) (in) (in) (in) (lb)

Single wire 9.4 1  0.276 22 CCL compact Mark I wire 5 161⁄8 25⁄8 14
3 1
37.4 4  0.276 ⁄4 or 1 ⁄2 34 plate system 10 211⁄8 25⁄8 18
75.2 8  0.276 11⁄2 45 15 261⁄8 25⁄8 22
112.8 12  0.276 15⁄8 5  51⁄2 5 203⁄4 23⁄8 18
3
37.4 4  0.276 ⁄4 or 11⁄2 21⁄2  21⁄2  31⁄4 CCL spiral Mark II wire 10 263⁄4 23⁄8 24
75.2 8  0.276 11⁄2 3  51⁄2  7 wire system 15 323⁄4 23⁄8 30
112.8 12  0.276 15⁄8 or 2 43⁄4  43⁄4  7 20 383⁄4 23⁄8 36
3
14.7 1  3⁄8 ⁄4 21⁄4  21⁄4 CCL standard Mark II small 10 303⁄8 33⁄4 72
3
25.8 1  1⁄2 ⁄4 33 strand system strand 20 403⁄8 33⁄4 97
58.8 1  0.7 11⁄2 31⁄2  31⁄2
76.0 1  7⁄8 11⁄2 44 Mark II Large 15 31⁄2 101⁄2 380
102.0 11 11⁄2 41⁄2  41⁄2 strand
129.0 1  11⁄8 11⁄2 51⁄2  51⁄2

IP: 130.194.20.173
Single strand 58.0 1  0.7 11⁄2 41⁄2  41⁄2  7 CCL spiral Mark II large 8 193⁄16 75⁄8 95

On: Tue, 17 May 2011 06:23:34


76.0 1  7⁄8 11⁄2 41⁄2  41⁄2  7 strand system strand
102.0 11⁄2

Delivered by ICEVirtualLibrary.com to:


11 41⁄2  41⁄2  7
129.0 1  11⁄8 11⁄2 41⁄2  41⁄2  7
67.4 3  1⁄2 2 41⁄2  41⁄2  7 CCL 3-strand Mark II small 10 303⁄8 33⁄4 72
spiral strand 20 403⁄8 33⁄4 97
103.2 4  1⁄2 2 41⁄2  41⁄2  7 CCL 4-strand Mark II large 8 193⁄16 75⁄8 95
spiral strand
137.00 7  7⁄16 2 68 CCL 7-strand
181.0 7  1⁄2 2 68 system
250.0 7  0.6 31⁄4 99
406.0 7  0.7 31⁄4 10  10
10.qxd 15/11/2001 15:53 Page 205

Prestressing 205

Table 10.5 The Gifford–Udall and Gifford–Burrow systems

Type Initial Number and Duct Anchorage Trade Jack type Jack Retracted Maximum Jack
force size (in) of size size (in2) description extension jack length diameter weight
(lb  103) wires or strand (in) (in) (in) (in) (lb)

Single 75.0 8  0.276 11⁄2 31⁄2 Gifford–Udall Tube 5 141⁄2 21⁄4 10


wire 112.8 12  0.276 2 4 10 191⁄2 21⁄4 13
124.0 8  0.354 2 4
146.6 12  0.315 2 6
186.0 12  0.354 2 6
197.0 21  0.276 2 6
269.0 22  0.315 2 81⁄2
341.0 22  0.354 21⁄2 81⁄2
357.0 38  0.276 23⁄4 81⁄2
Multi- 181.0 7  1⁄2 2 6 Gifford–Burrow 0.5 Small 20 34 3
strand 224.0 4  0.7 2 6 strand 12 22 3
249.0 7  0.6 2 6
310.0 12  0.5 23⁄4 81⁄2 0.7 Small 10 181⁄2 41⁄4
392.0 7  0.7 23⁄4 81⁄2 strand
429.0 12  0.6 23⁄4 81⁄2
Large 58.0 1  0.7 11⁄2 4 Gifford–Burrow 0.7 Small
strand 58.0 1  0.7 11⁄2 5 strand
102.0 1  1.0 11⁄2 4
102.0 1  1.0 11⁄2 5 Large 10 241⁄2 61⁄2 96
129.0 1  11⁄8 11⁄2 4 strand
129.0 1  11⁄8 11⁄2 5

Figure 10.12 Combined


vertical and horizontal spacer
(Magnel–Blaton system).

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:23:34
10.qxd 15/11/2001 15:53 Page 206

206 Historic concrete

The use of prestressed concrete for railway sleepers started during the war and
has continued ever since. It is estimated that over 35 million have been made in
the UK alone. In general these have been made in dedicated factories but often,
alongside, additional beds were laid down to produce beams and flooring units;
flooring units tended after a while to be cast in dedicated factories. Great empha-
sis was placed in the early days on producing the lightest units possible, and Bison
produced a prestressed equivalent of its standard hollow-core unit; outwardly it
could be extremely difficult to say whether a particular unit was made of pre-
stressed or normal reinforced concrete, unless one had access to its ends.
Another unit was an inverted trough which was more often used as a roofing
unit. This was originally developed by the Ministry of Works but subsequently
became the Milbank floor. It would have been difficult to make such a unit in rein-
forced concrete (Figure 10.13).
A popular unit was the X-joist which was originally created as a purlin for farm
buildings, again by the Ministry of Works. This was taken up by the French par-
ticularly, in the first instance, for the rebuilding of Caen. The shape lent itself to
the use of steel pipes for the side moulds which could be used for steam curing,
thus getting a quick turn-around on the stressing beds. This was reintroduced into
the UK to become the Pierhead floor. Unfortunately, back in the UK the pipes
were not used for steam curing and, after a time of using normal cement, high-
alumina cement was used to achieve a quicker turnaround in the moulds until
the high-alumina scare of the 1970s. But undoubtedly many floors constructed
with these joists will still be found. These were used with hollow pots, usually of
concrete (Figure 10.14). One of the difficulties with all these units was the differ-
ential camber which occurred between them. Although strict quality control was
in place, some units hogged more than others under prestress, partly because the
‘small’ section modulus proved sensitive to this effect and partly because early
release from the pre-tensioning beds was an economic necessity.
This problem of differential camber was, in one sense, overcome by what was
known as the ‘Stahlton floor’ which originated in Switzerland in 1945. Originally
it was made up of discrete clay tiles grooved on the top. Pre-tensioned wires ran

Figure 10.13 The Milbank


inverted trough prestressed
unit (shown in a composite
floor). Reproduced by
permission of the British
Constructional Steelwork
Association Limited.

Figure 10.14 Typical


prestressed X-joists.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:23:34
10.qxd 15/11/2001 15:54 Page 207

Prestressing 207

in the grooves and mortar was placed between the ends of the tiles and in the
grooves to achieve bond; being uniformly prestressed and flexible, they did not
suffer from the problem of differential camber. These planks were supported at
5 ft centres and hollow pots were inserted between them. In-situ concrete and any
additional reinforcement was then placed (Figure 10.15). These were introduced
into the UK by the Costain Concrete Co. in 1952. In Scotland, and later in
England, the planks were made in concrete so it would not be easy to recognize
them as a prestressed concrete floor.
This form of construction was extended to form complete flat-slab floors, by
using wide planks up to 4 ft in width for the tension zone of main beams, with the
Stahlton-type planks butting up against them transversely, in-situ concrete with
additional steel for shear and continuity being placed on top.
Samuely used and extended the principle for a large number of schools, using
much larger planks and trough units between them, to achieve spans up to 40 ft,
although later he tended to favour prestressed concrete tees rather than planks
and achieved 1/36 depth to span ratios for floors and 1/45 for roofs (Figure 10.16).
More recently there has been a tendency to use wide hollow-cored prestressed
concrete units which give an immediate working surface if the job justifies
cranage being available. Extruded prestressed concrete slabs were being made in

Figure 10.15 Stahlton floor


units. Reproduced by
permission of the British
Constructional Steelwork
Association Limited.

Figure 10.16 Prestressed


concrete tee units designed
by Felix J. Samuely &
Partners.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:23:34
10.qxd 15/11/2001 15:54 Page 208

208 Historic concrete

Germany during the Second World War; 30 years later they were introduced into
Britain. Annual production is running at about 2.75 million square metres at
present.

Prestressed It will be appreciated that a number of these systems are in fact simply providing
the tensile resistance of a beam or a slab using prestressed concrete or tile instead
concrete trusses of steel reinforcement. This principle has been logically applied to trusses, where
the tension members were made of prestressed concrete and the compression
members of reinforced concrete. Some of the lighter stressed tensile members were
also made of reinforced concrete.
This method was usually employed for one-off jobs, and a truss with smaller
than usual concrete tension members should be considered as a possible prestressed
concrete truss. This method was applied in a more general form in the Intergrid
and Laingspan systems. They were specifically developed for the post-war school
building programme but were applied with modifications to other buildings such
as in the rebuilding of Aldershot. The beams in the system basically consist of short
precast concrete units either with grooves on the upper surface of the lower flange
or with holes in them in which prestressing wires were inserted on site and pre-
tensioned (Figures 10.17 and 10.18).

Prestressed In general these have been confined to simple portals, most often using precast
columns and beams stressed together to give continuity. The likely distinguishing
concrete frames feature in these cases is the slenderness of the top member — the verticals often
carried crane rails so that their slenderness would be masked. These were used in
power stations as well as workshops. Figure 10.4 illustrates a workshop at Pyestock
which also used Milbank units for the roof. As far as the author is aware, only two
framed office-type buildings were erected, both in the London area.

Figure 10.17 Laingspan


system of construction.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:23:34
10.qxd 15/11/2001 15:54 Page 209

Prestressing 209

Figure 10.18 Intergrid


system of construction.

Piles Prestressed concrete has been used for precast piles on a number of jobs, but usu-
ally on sites such as oil refineries where a large number are required, and where
it is economic to set up pre-tensioning beds on site to make them. Most piles in
this situation really only require ‘reinforcing’ for handling and driving, and for
this prestressing is ideal. Occasionally piles were made of precast units stressed
together with Macalloy bars for use in confined situations.

Special applications Containment vessels and high-pressure pipes have often been made of pre-
stressed concrete. In the former category nuclear reactor pressure vessels are a
good example. In these cases the prestressing tendons are often capable of being
withdrawn and inspected for safety reasons. In the latter cases the tangential
prestressing is often induced by winding on prestressing wire under tension —
longitudinal stressing, often by post-tensioning, being provided largely for han-
dling stresses.
Turbine blocks have also been prestressed largely to ensure their integrity under
complex loading, as have large industrial presses.14
Ground anchors which are used extensively are essentially holes drilled into the
ground into which prestressing tendons are inserted — the hole is grouted and,
when sufficient strength has been reached, the tendon is stressed and locked off.15
Test floors have often been made using prestressed concrete, particularly in lab-
oratories where the floors act as the lower half of test frames if access is possible
or required below them. If steel members are not embedded in such floors it is
likely that prestressed concrete has been used. Examples of these are to be found
in university and government laboratories.
A few examples exist of prestressed concrete ‘lift’ slabs where the roof and floors
were cast on top of each other at ground level, with holes left for the columns.
The slabs were then jacked up to their appropriate level and attached to the
columns using steel connections.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:23:34
10.qxd 15/11/2001 15:54 Page 210

210 Historic concrete

Following the successful use of prestressed concrete beams by the Germans


in the construction of their U-boat pens, they have been used on several occa-
sions in the UK in explosive situations, particularly in government research
laboratories.

References 1. Thomas, F.G. (ed.), Prestressed Concrete: Proceedings of the Conference held at the
Institution of Civil Engineers. ICE: London, 1949 (lists 50 patent and 289 references).
2. Walley, F., Prestressed Concrete: Design and Construction. HMSO: London, 1953 (50
references).
3. Walley, F., The progress of prestressed concrete in the United Kingdom. Prestressed
Concrete Development Group (Cement and Concrete Association): Slough, 1962 (41
references).
4. Andrew, A.E., Turner, F.H., Post-tensioning systems for concrete in the UK:
1940–1985. CIRIA: London, Report 106, 1985 (illustrates and describes numerous sys-
tems used in this period).
5. Emperger, F., Handbuch für Eisenbetonbau, Bd 1, Entwicklungsheschichte, 3rd edn.
Ernst: Berlin, 1921.
6. Labes, Die Anwendung des Eisenbetonbaues für Eisenbahnzwecke. Glasers Annln,
1906, 1 December.
7. Koenen, M., Wie kann die Anwendung des Eisenbetons in der Eisenbahnverwaltung
gefoerdert werden? Centralbltt Bauverwaltung, 1907, 520–22.
8. Moersch, E., Versuche uber die Schubwirkungen bei Eisenbetontraegern. Dtsche
Bauzeitung, 1907, 13 April, 207; 20 April, 223; 1 May, 241.
9. Moersche, E., Spannbetontraeger. K. Wittwer: Stuttgart, 1943.
10. Labes, Bestimmungen für die Ausfuhrung von Konstruktionen aus Eisen-beton bei
Hochbauen von 24 Mai 1907, Berlin, 1908. Germany Concrete Regulations.
11. Freyssinet, E., Prestressed concrete: Principles and applications. J. Instn Civ. Engrs,
1950, 33(4), 221–380.
12. Magnel, G., Creep of steel and concrete in relation to pre-stressed concrete. Am. Conc.
Inst. J., 1948, 19, 485–500.
13. Magnel, G., Le flauge des aciers et son importance en beton precontraint. Sci. Tech.,
1945, 2.
14. Prestressed Concrete Development Group. Symposium on the Application of
Prestressed Concrete Machinery Structures. PCDG: London.
15. Hannah, T.H., Design and construction of ground anchors. Report 65. CIRIA:
London, 1980.
16. Preload Ltd. Prestressed concrete tanks. 1965, 1969 edns. London and Northampton.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:23:34
11.qxd 15/11/2001 15:54 Page 211

11 The development of concrete bridges


in the British Isles prior to 1940

Mike Chrimes
Synopsis This chapter traces the use of concrete for bridges in the British Isles from its 19th
century origins to the outbreak of the Second World War. It describes the intro-
duction of reinforced concrete bridges by specialist firms in the early 20th cen-
tury, and some of the more notable bridges designed in the interwar period when
British engineers established a reputation independent of imported systems. The
most important bridge types are identified, and provision for movement and con-
siderations of appearance are discussed. The development of standard bridge load-
ings is traced, and the performance of some of the early bridges is outlined. The
chapter also comments on the economic background, and contrasts progress in
the British Isles with that overseas.

Background Before the Second World War comparatively few bridges in the British Isles were
built of reinforced concrete; at least 75% of the Department of Transport’s con-
crete bridge stock has been built since 1960.1 Reinforced concrete bridges came
to Britain 20 years later than the continent, the first British textbook not appear-
ing until 1913,2 and the majority of the early bridges were built to methods
and systems first developed abroad.3 This backwardness is surprising, as British
engineers had been using concrete regularly since the early 19th century.4

Early British In the early 19th century, British engineers began to use concrete for bridge foun-
dations and substructures.5 This development was largely indigenous and
development Lamande’s use of hydraulic lime concrete for bridge foundations (1802–15) and
Vicat’s important research and concrete foundation work at Souillac bridge
(1818–23) apparently had no immediate British emulator.6 The earliest applica-
tion of Roman cement to a bridge superstructure traced hitherto is for a road
bridge over the Birmingham–Derby Railway line (c. 1837) on the Cliffe Whate-
ley Road. Described as an ‘arch in cement’, the bridge compridsed an arch with
a Roman or proto-Portland cement mortar in combinaton with engineering bricks
as seen in sunivals. Bridges on this line were described as more economical than
those on the earlier London–Birmingham Railway suggesting the motive for the
innovation. The type was common on the Trent Valley line built in the 1840s.6a
The earliest mass concrete bridge was designed by Thomas Marr Johnson, for Sir
John Fowler, on the District Line near Cromwell Road.7 A shortlived affair, the
original lime concrete structure (1867) failed when the centering was struck and
was replaced by a Portland cement concrete bridge, demolished by 1873. Although
Fowler and Baker were later responsible for experiments on expanded metal rein-
forced arches (1895–96) no other concrete bridge designs by them are known.8
However other British engineers began to use plain concrete for bridge super-
structures in the last quarter of the century.
One patentee, Philip Brannon, erected a three span concrete arch at Seaton in
Devon (1877, 50 ft middle span).9,10 Railway engineers were also active.11 Block-

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:30
11.qxd 15/11/2001 15:54 Page 212

212 Historic concrete

work was used on the Callender line (1878) and ‘rubble’ concrete (layers of rub-
ble set in concrete, with alternating layers of rubble concrete and plain concrete)
used for the fill on the Dochart viaduct (1886), which had plain concrete arches.11
The best known extensive use of mass concrete for bridges, on the West Highland
Railway, is surprisingly late (1897–98).12 By this time the London and South
Western Railway had started using mass concrete, at Holsworthy and elsewhere.11
Overseas
developments On the continent Coignet was building substantial concrete arched structures in
the late 1860s.13–15 Monier began experimenting with reinforced concrete arches
from 187316–18 and from the 1880s there were rapid developments in the theory
and application of reinforced concrete on the continent and in the USA.19–22 By
the turn of the century their understanding of reinforced concrete was well
advanced. These developments, although known in Britain through international
exhibitions and engineering literature, were initially overlooked. British engineers’
use of reinforced concrete for bridges was particularly backward. There was
only one isolated instance of the use of iron embedded in concrete for bridges,
Homersfield bridge.
Early reinforced
concrete bridges The bridge over the River Waveney at Homersfield on the Suffolk/Norfolk bor-
in Britain der comprises a wrought iron cage infilled with concrete, and is of uncertain struc-
tural action (Figure 11.1). The 50 ft span bridge was built in 1870 by T&W Philips
using their patented fireproof system.23–25 Hardly an example of reinforced con-
crete, it was more than 30 years before the introduction of foreign systems led to
a rapid expansion in the use of reinforced concrete for bridges in the decade before
the First World War.3
These early bridges were built without design standards, and specifications were
provided by the designers without reference to national guidelines.26 The struc-
tural forms of these bridges had all been developed abroad.27–28 A resume of the
principal types in use before 1940 is to be found in Table 11.1. This reveals strik-

Figure 11.1 Homersfield


Bridge, before restoration
work (photograph ©
E.A. Labnum).

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:30
Table 11.1 Typical bridge types 1900–1940:29,30,99,178,224 although terminology varies, reinforced concrete bridges built before the Second World War can be
classified broadly as given below

Bridge type Section Earliest UK


examples

Beam and slab: parallel longitudinal (a) 1902


beams with flat (a) or haunched (b) 1909
(b) soffits below the roadway
11.qxd 15/11/2001 15:54 Page 213

Beam and slab: parallel longitudinal 1904


beams with soffits curved to
resemble arches

Beam and slab: parapet beams 1906


supporting a transverse spanning
roadway structure

IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:30
Delivered by ICEVirtualLibrary.com to:
Variable depth trusses (a) or (a) 1903–1904
bowstrings (b) with a transverse (b) c. 1925
road structure (note: the terminology
is used loosely in many reports)

Slab vaults with earth or other fill 1904–1905


retained by outer spandrel walls
The development of concrete bridges in the British Isles prior to 1940
213
Table 11.1 (Continued)
214

Open spandrel arch ribs 1904–1905


below the road slab
11.qxd 15/11/2001 15:54 Page 214

Solid spandrel arch beams 1901–1902


Historic concrete

(presumed to have been


designed as arches rather
than beams); secondary
beams below deck slab

Portal frames consciously 1910


designed as such Common c.1928

IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:30
Delivered by ICEVirtualLibrary.com to:
Vierendeel girders 1904
1924
Rare

Balanced cantilevers c.1918


with suspended spans Common c.1935
11.qxd 15/11/2001 15:54 Page 215

The development of concrete bridges in the British Isles prior to 1940 215

ing parallels between bridges in more traditional engineering materials and the
new designs in reinforced concrete. Engineers had been using arch ribs for more
than a century in cast iron. Even more conservative were the earth-filled vault slab
design for arch bridges. With masonry spandrel walls there is little to distinguish
them from the thousands of stone and brick structures, familiar to generations of
engineers. Without access to original drawings it is not easy to establish the struc-
tural principles behind the design of some early bridges, particularly arched bridges
with solid spandrels, which may have been designed as beams rather than arches
in some cases.
Mouchel-
Hennebique The earliest examples of reinforced concrete bridges in Britain were all the work
Bridges29–32 of L.G. Mouchel and Partners, the UK agents for the Hennebique system. By the
end of the First World War they had been involved in over 33 bridges, viaducts
and similar structures, the majority in the period 1907–15 as bridge construction
virtually halted during the war. This was something like 80% of the reinforced
concrete bridges erected in the British Isles at the time.
Some contracts were confined to foundations. Others involved strengthening
or widening existing structures, such as Telford’s cast iron bridges at Stokesay and
Cound (1918), where reinforced concrete arch ribs were added.33 The majority
were road bridges although they designed nearly 30 footbridges. Most bridges were
of modest span (6–18 m). They were characterized by monolithic construction
methods which ensured that all the elements of a structure were combined with
one another, with reinforcement of main girders anchored in adjoining members,
and stirrups used to help bind the structure together. Very thin structural sections
were frequently used for parapet beams, perhaps as little as 6 in with nominal 1 in
of cover over the reinforcement.
Mouchel designed the first ‘modern’ reinforced concrete bridge erected in the
British Isles, Chewton Glenn (1901) in Hampshire, a modest (18 ft) span skew arch
with the concrete ribs concealed behind brickwork.34 It was soon followed by a beam
and slab bridge, on a slight skew, over Sutton Drain, Hull (1902) (Figure 11.2).35
This had a 40 ft span but was 60 ft wide. The first reinforced concrete bridge erected
in Scotland was a 28 ft span road and rail bridge in Dundee (1903), the first British
reinforced concrete railway bridge. It was 40 ft wide, supported on 4 solid span-
drel arched beams, and secondary transverse beams at 4 ft 4 in centres. The first
so called bowstring girder was erected at Purfleet pier in 1903–1904, a 60 ft span
(Figure 11.3).36,37 The Mellor Street bridges in Rochdale, completed in 1905, were
reinforced concrete slab vaults and were Mouchel’s first such structures not to
include ribs or beams in their construction. Bridges of this type had no deck slab.
The Rochdale bridges were the first with abutments and foundations also all in
reinforced concrete.37a

Figure 11.2 Cross-section of


Sutton Drain Bridge, Hull
(1902).

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:30
11.qxd 15/11/2001 15:54 Page 216

216 Historic concrete

Figure 11.3 Truss girder, Purfleet Pier (1903–1904).

Figure 11.4 Berw Bridge,


Pontypridd (1907).

For longer spans Mouchel used open spandrel rib arches. Berw Bridge at
Pontypridd (1907–1909) comprised three spans of 25 ft (7.62 m), 116 ft (35.36 m)
and 25 ft (7.62 m) (Figure 11.4).38
Of similar form was the recently demolished three span (36 ft, 89 ft, 36 ft)
Jackfield Bridge (1909).38a The largest such arch bridge of this period was the
Floriston bridge over the Esk in Cumbria, (1914) with a central arch rib span of
175 ft and two side spans of 147 ft 6 in.38b The arrangement of reinforcement here
(Figure 11.5a) can be compared with that in other span spandrel rib designs
of the period — the Kahn Bridge at Farnworth (1911) (Figure 11.5b) and the
Considere Bridge at Bridgend (1910) (Figure 11.5c) Crewe Park Bridge (1907)
is an example of Mouchel’s solid spandrel arch design. The reinforcement details
(Figure 11.5d) show it comprised an arch rib, with transverse deck beams and a
continuous deck slab which also provided lateral bracing. The reinforcement in
the spandrel walls comprised stirrups connecting the rib reinforcement with the
decking, and dealt with shear, temperature and other stresses.38c Floriston replaced
a cast iron structure of similar spans, a reminder British reinforced concrete bridges
still had not exceeded spans achieved in cast iron a century earlier; they were also
relatively modest compared with contemporary foreign bridges.39
The overall length of some was, however, impressive. A highway viaduct, erected
in 1905 along the River Suir near Waterford, was 720 ft long, supported on 106

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:30
11.qxd 15/11/2001 15:54 Page 217

The development of concrete bridges in the British Isles prior to 1940 217

Figure 11.5 Reinforcement details: (a) Floriston Bridge (1914); (b) Farnworth Bridge (1911); (c) Ogmore Bridge,
Bridgend (1910) and (d) Crewe Park Bridge (1907).

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:30
11.qxd 15/11/2001 15:54 Page 218

218 Historic concrete

Figure 11.6 Elevation and


plan, from below, of Street
Viaduct, Dover (1913–22).

piles.40 Longer still was the 1000 ft Street Viaduct in Dover (Figure 11.6), with asso-
ciated bowstring girder bridges; work began on this in 1913, but, due to the war
was not completed until 1922.41 Both were partly curved in plan.

Other systems Other reinforced concrete systems were in use by 1914.42 The Monier system had
been marketed since 1902 by the Armoured Concrete Construction Company, but
no bridges are known by them. The (American) Kahn System, originally intended
for buildings, soon branched into bridge construction with the UK agents
Truscon.43,44 The earliest contracts were at Lilburn (1907) and Lucker (1906) in
Northumberland, replacing two masonry arches washed away in a flood of 19 May
1906. Both were open spandrel rib arches. Of rather more interest is the 74 ft 9 in
span structure at York Race Course, supported on parapet girders, with four
curved crossbeams over, and cross beams beneath the deck (Figure 11.7).45
Hartlake Bridge (Tonbridge, 1910)46 comprised a 36 ft sloping approach span,
and a slightly arched (12 in rise) river span, with transverse beams to edge beams
utilizing the parapets. Their first multispan bridge was possibly Kings Bridge
Belfast (1909–1910), with four horizontal spans (40/50/50/40 ft).47 The arrange-
ment of the reinforcement for the beams (Figure 11.8a) can be compared with
Mouchel’s near contemporary structure over the Wansbeck at Stakeford (1909)47a
(Figure 11.8b).
The work of the English branch of Considere and Partners reflects the think-
ing of Considere and his partner Caquot in France, and represents perhaps
the first clear break with Mouchel Hennebique methods. The earliest Considère
design was a 17 span beam and slab structure over the Great Eastern Railway at
Angel Road, Tottenham (1908) (Figure 11.9), subsequently incorporated in the
North Circular Road; the spans were 42 ft 9 in long and movement joints were
employed every 200 ft.48–50 A more attractive structure was the 134 ft span arch
over the Mersey at Warrington (1909–1915). This bridge, 80 ft wide, was built in
two sections.51,52
Edmund Coignet & Co. also designed a few bridges in this period.53,54 The most
important early structures were those at Kings Cross over the Metropolitan Rail-
way,55 and two bowstring girder railway bridges at Bangoed.56 The latter were on
a colliery line, the larger being a 56 ft 9 in skew span (Figure 11.10). The Bridge

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:30
11.qxd 15/11/2001 15:55 Page 219

The development of concrete bridges in the British Isles prior to 1940 219

Figure 11.7 York


Racecourse Bridge (1908).

Figure 11.8 (a) King’s


Bridge, Belfast (1909–10) and
(b) Stakeford Bridge (1909).

at Kings Cross was a beam and slab design, 130 ft long in two principal spans of
53 ft and 39 ft, and 60 ft wide, supporting a roadway and two tramway lines. The
bridge was supported on columns at the junction between the principal spans.
The Indented Bar Company were beginning to emerge as an important design
firm just before 1914, and would have been more so if their proposed bridges
on the Metropolitan and Great Central Railways north of Watford had been
executed.56a
Whereas Mouchel made use of specialist approved contractors, it is interesting
to note that D.G. Somerville & Co. were agents for the Kahn, Considere and
Coignet systems, advertising standard arch and beam/slab bridge types which could
be built to any system.57,58

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:30
11.qxd 15/11/2001 15:55 Page 220

220 Historic concrete

Figure 11.9 Angel Road


Viaduct, Tottenham (1908).

Figure 11.10 Bargoed Bridge


(1911).

As articles appeared describing the early system bridges, and textbooks became
available, it became easier for local authority and other engineers to draw up their
own designs.59,59a One of the earliest was Thorverton Bridge, Devon (1907).60,61
This was designed by S. Ingram, surveyor for the north of the county. This was
an 85 ft span structure supported on four arch ribs with secondary transverse
beams supporting the decking. Reinforcement included trussed and plain bars and
expanded metal mesh. Expanded metal27,59a was also used for a simple reinforced
slab 12 in deep of 18 ft span, 40 ft wide at Withycombe Rd, Exmouth (1910); the
reinforcement diagram suggests this was at least unconsciously an early portal
frame (Figure 11.11).62,63 This and a neighbouring bridge were designed by the
local engineer, S. Hutton.
In Somerset a group of 24 reinforced concrete bridges (1909–1914) were designed
by E.J. Stead, Assistant County Engineer.64–67 The majority of these were slab vaults

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:30
11.qxd 15/11/2001 15:55 Page 221

The development of concrete bridges in the British Isles prior to 1940 221

Figure 11.11 Withycombe


Road Bridge, Exmouth (1910).

with masonry spandrels although Donyatt North Bridge was a beam and slab
design with haunched external beams (1909).68 Interestingly Stead had worked
on a reinforced concrete arch bridge in Natal previous to coming to Somerset
in 1908.65 The arches were designed using conventional graphical methods of
the type described by Cain69,70 and Marsh,59 and incorporated a bitumen water-
proofing membrane above the arch ring.

Road bridges Reinforced concrete established itself as a cheap alternative to steel or masonry
for bridge building in Britain before the First World War. It was not used as widely
between the wars as elsewhere partly as the result of a well-developed rail-based infrastructure, and
also because initial concerns over durability restricted the funding arrangements
available from the Local Government Board.71
In the inter-war period the spread of the motor car and the popularity of cycling
encouraged a road building programme which was frustrated by the depression.72
The Road Fund was established in 1920 and the Unemployment (Relief Works)
Act (1920) provided an additional incentive to road building. Concrete bridges
were found to be relatively easy to construct with an unskilled direct labour force
and where a structure was not immediately required this gave the material an
advantage over steel,73 although concrete often lost out for major crossings. By
1930 there were something like 2000 reinforced concrete bridges in the country.
Nevertheless Britain lagged far behind leading continental countries and the
United States in developing its road network, and one scheme approved before
1914, the Great West Road, was still incomplete in 1939. This was symptomatic
of the period and British engineers’ awareness of overseas developments in bridge
building techniques74 and sophisticated road junctions,75 counted for nothing in
the absence of equivalent investment. The road programme and associated bridge
works virtually stopped during 1931–34. There were, however, some outstand-
ing bridges built between the wars and notable designers like Sir Owen Williams
emerged.76
The Royal Commission on Transport (1930) mentioned a target of 1000 bridge
strengthening schemes a year. Although only 200–300 schemes had been approved
by 1939, this was one area where concrete was regularly used. An interesting

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:30
11.qxd 15/11/2001 15:55 Page 222

222 Historic concrete

Figure 11.12 White’s Bridge,


Leek (1930–31).

Figure 11.13 Royal Tweed


Bridge, Berwick (1928).

example is White’s Bridge in Leek (1930–31). Here, because of the skew angle and
restricted channel width, and to avoid the stress concentration at the acute corner
of the abutment, an arch rib and slab solution was preferred to a slab vault. Spe-
cial consideration was given to the connection between the outside rib and the abut-
ment where ‘the steel was carried well back into the abutment’ (Figure 11.12).71
Many early 1920s bridges resembled prewar designs. These include several
major arch rib structures built in the 1920s by Mouchel such as Atcham Bridge
carrying the A5 over the River Severn (1924–27),78 and the Royal Tweed Bridge,
Berwick,79 completed in 1928. Berwick’s scale was remarkable for British bridges
of the time, comprising spans of 167, 248, 285 ft and 361 ft 5 in plus approach
viaducts of 199 ft and 144 ft 6 in. Distinctively the spans increase as the north shore
is approached and the bridge incorporated expansion joints over the piers and
abutments. The arch ribs of the three longer spans were hollow. The bridge encap-

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:30
11.qxd 15/11/2001 15:55 Page 223

The development of concrete bridges in the British Isles prior to 1940 223

Figure 11.14 Chiswick Bridge (1933).

Figure 11.15 Twickenham


Bridge (1933).

sulates all the best characteristics of Mouchel’s work with open spandrel arch ribs
(Figure 11.13).
The two arch bridges designed by Considere and Partners at Chiswick and
Twickenham completed in 1933 are of interest.80 Chiswick comprises a three span
solid spandrel barrel arch, with the 9 in roadway slab resting directly on the vault-

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:30
11.qxd 15/11/2001 15:55 Page 224

224 Historic concrete

Figure 11.16 King George V


Bridge, Glasgow, closing of
central span.

ing at the crown, and on reinforced concrete columns elsewhere (Figure 11.14).
The whole was of cellular construction. At the crown vertical steel plates separated
the half vaults so hydraulic jacks could be inserted to reduce the effects of creep
and shrinkage.81–83
At Twickenham the three hinged barrel arch form of the three main spans was
made explicit (Figure 11.15). The deck was supported on columns over the piers
and abutments, but otherwise supported on longitudinal diaphragms.84
Some features of these bridges had been anticipated by the King George V
Bridge in Glasgow (1924–28), although this was a continuous girder of arched
appearance, 423 ft 6 in in overall length, resting on cast steel roller bearings.85 The
structure comprised 12 parallel girders joined by a vault slab below and the road-
way slab above, with transverse stiffening walls. A central 15 ft strip was omitted
during construction to avoid stresses in the deck slab and concreted monolithically
with the rest of the structure as the final phase (Figure 11.16). The Ellon Bridge
Aberdeenshire (1939–40) also had continuous girders simulating arch ribs, but in
this case the vault slab was omitted, and the deck slab formed a top flange.86

New bridge forms Cantilever bridges

Sir Owen Williams’ replacement of the wrought iron suspension bridge at


Montrose (Figure 11.17) (1928/30) is a twin ‘balanced’ cantilever truss with a
central drop in section, the spans comprising 2 at 108 ft and a centre opening of
216 ft. Its appearance was in conscious imitation of its predecessor. Williams jus-
tified the use of reinforced concrete on cost grounds, although his design was criti-
cized for seeking to imitate a structural form which would have been better
engineered as a steel suspension bridge.76,87,88
Cantilever construction was hardly a novelty with such an example as the Forth
Railway Bridge, and in the late 1930s several cantilever bridges were built. The
balanced cantilever of the type we would recognize today was developed around
1930 on the continent and in Brazil. At Alveley Bridge (1937) cantilevered cen-
tering was used for balanced construction by BRC.89,90 Greenfield Lock Bridge,
Chester (1939–40), a 49 ft 4 in skew span, had a suspended span with 26 ft long
longitudinal beams carried on cantilevers, with balance arms of cellular con-
struction loaded with selected fill.91 The idea of anchoring a shore span, using

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:30
11.qxd 15/11/2001 15:55 Page 225

The development of concrete bridges in the British Isles prior to 1940 225

Figure 11.17 Montrose


Bridge (1930).

counterweights to balance a cantilever arm, and finally inserting a drop-in simply


supported central span, may well have appeared much earlier.
Chettoe and Adams (1933) suggested the use of cantilever and suspended spans,
particularly in multispan girder bridges where there was the possibility of settle-
ment of piers, or for curved viaducts.92 The Lindfield Bridge, Sussex, widening
(1938), was a three span slab with the two side spans cantilevered out over the
pile cap beam supports to support the central slab.93 Inverscaddle Bridge, Argyll
(1939–40) is a three span structure with end slabs counterbalanced at the abut-
ments, cantilevering over the piers, and supporting a suspended central span.94
The Bridge of Orchy, Argyll, (1937–39) was also a three span structure.95 The ele-
vation of its beams was that of a flat arch. There were four main beams to each
span, with the side spans fixed at the abutments, and cantilevered out to support
a 40 ft suspended span.

Rigid and portal frames

Rigid or portal frame type bridges were probably first used for a railway over bridge
at Markersbach in Saxony by the Hennebique agent Max Pommer (Leipzig) in
1902.96 It is unclear when Mouchel first used this form, which had obvious attrac-
tions where beam depth was limited by waterway or traffic clearance.97,98 There
were apparently early British examples at Exmouth (Figure 11.10) and Dugdale
in 1910.62 The latter was designed by the Indented Bar Company.56a Owen Williams
realized their value and designed a bridge of the type at Shepherdslea Wood on
the A2 over the Southern Railway (1927)99 and from the late 1920s such bridges
became increasingly common.100,101
Mouchel designed a bridge at Wisbech with a 92 ft 6 in span (1929–30), the clear-
ance being required for navigation (Figure 11.18).102 Two hinged Portal frame
designs were also used in connection with railway works on the Larne line between
Carrickfergus and Whitehead (1929–34).103 These works also included the first
and the Greenisland loop line flat slab highway bridge in the British Isles,104
designed by Truscon.104a,104b
A two span design was used at Ystradgynlais (Figure 11.19) (c. 1932).105 Steel
portal frames were encased in concrete at Water Eaton, Oxfordshire (Figure 11.20)
(1936).106 An outstanding group of cellular portal frame bridges were designed
by F.A. Macdonald in the late 1930s in Scotland.107–109 That at Aberuthven on the
A9 was a 40 ft wide three hinged structure with 73 ft main span (Figure 11.21).
Either side of the crown hinge the slab was solid for 15 ft, increasing in thickness
from a minimum of 14 in. It then divided into a deck and vault slab connected by

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:30
11.qxd 15/11/2001 15:55 Page 226

226 Historic concrete

Figure 11.18 Wisbech Bridge


(1929–30).

Figure 11.19 Ystradgynlais Bridge, showing reinforcement (1932).

Figure 11.20 Water Eaton


Bridge (1936), (courstesy of
the Structural Engineer).

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:30
11.qxd 15/11/2001 15:55 Page 227

The development of concrete bridges in the British Isles prior to 1940 227

Figure 11.21 Longitudinal section of Aberuthven Bridge (1939).

Figure 11.22 Stow Bridge


(1925–26).

ribs 9 in thick at 9 ft centres. Side flood arches were constructed monolithically to


provide stabilizing moments. The bridge was faced in natural stone.

Other forms

Bowstring girders became more common from the mid 1920s.110 Several early
examples such as Stow Bridge (1925–26) (Figure 11.22) were built by Peter Lind.,
or K. Holst, and may reflect these firms’ overseas origins.
Owen Williams appears to have been the first British engineer to use a
Vierendeel truss, for his collaboration with Ayrton at Findhorn (Figure 11.23)
(1924–26).76,111,112 He used two pairs of beams, 98 ft span, separated over the cen-
tral pier and 36 ft apart. This type had been anticipated by Wayss and Freytag at
Krapiza in Austria–Hungary in 1900 and the idea had attracted some attention
in Germany in the early part of the century.113–116 Mouchel’s early truss at
Purfleet (Figure 11.3) bears a striking resemblance to this type.36
Williams designed a mass concrete arch, for aesthetic reasons, at Wansford
(1930). The bridge was fitted with three temporary steel hinges which were grouted
up when the construction was complete.76,111 Earth filled solid spandrel slab vaults
continued to be used throughout the period, but there is little evidence of the type

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:30
11.qxd 15/11/2001 15:55 Page 228

228 Historic concrete

Figure 11.23 Findhorn


Bridge (1924–26).

of vault arch used on the continent with open spandrels and the deck supported
on columns or transverse walls.117

Waterloo Bridge118 The present Waterloo Bridge (1938–42) in London in many ways exemplifies the
best of British interwar bridge design, although its planning was dogged by con-
troversy. It was designed by Buckton and Cuerel of Rendel, Palmer & Tritton, with
the architect Sir Giles Gilbert Scott. The bridge is a five span box girder structure,
with four main beams, two on each side of the bridge continuous over two spans
and a centre span comprising two cantilever arms and a suspended girder.
The maximum central spans are 250 ft. For architectural reasons the spandrels
were faced with Portland stone, and Scott’s desire to bring lightness to the under-
side of the bridge meant that the main box beams had to be located under the
footways rather than the road pavement, which had to be supported on secondary
T beams and slabs (Figure 11.24). These architectural requirements, including
restraints on the width of each beam, demanded ingenious engineering solutions.
In the end all the main reinforcement was welded, as lapping was impractical. One
notable innovation was the use of prestressing in the shore cantilevers, at the top
of the bearing walls, and around the centre span expansion joints, where high shear
stresses appertained. The bars, with ends upset and screwed, were placed in steel
tubes with projecting end connections. Once the concrete had gained strength the
bars were stressed by passing steam through the tubes, taking up the thermal exten-
sion by turning the end nuts, so that the required stress (30,000 lbs psi) was induced
on cooling. The steam connectors were used to grout up the bars in the tubes.
To deal with differential settlement jacks were built into the piers.
When one considers the tribulations of many concrete bridges in the post-war
period, Waterloo is an outstanding testimony to all involved.

Footbridges Several footbridges incorporating reinforced concrete arches were built over the
London and South Western Railway c. 1904.119 If one excludes links between fac-
tory buildings, Mouchel’s first footbridges were erected in 1908.29 With little road
traffic there are few early examples of pedestrian over road bridges, a notable
exception being the 96 ft arch bridge over the Brighton Road at Reigate Hill

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:30
11.qxd 15/11/2001 15:55 Page 229

The development of concrete bridges in the British Isles prior to 1940 229

Figure 11.24 Waterloo Bridge (1938–42): (a) part-longitudinal view showing part of continuous girder and part of sus-
pended central span and (b) half cross-section at crown.

(1908–1910).120 The 67 ft arch span bridge at Alum Chine in Bournemouth spanned


a steep-sided valley (1908) (Figure 11.25).
A large proportion of footbridges built in this period were over railways, gen-
erally at stations. These structures, because of the load gauge and lateral spacing
could well be considered early ‘standard bridges’, and early precasting techniques
were used.
The concrete girder usually reflected plate girder appearance. Kew Garden
Station footbridge (1911–12) was an exception which used a concrete bow string
for the main truss with diagonal stiffness provided by thin concrete infill panels
(Figure 11.26).121
Footbridges were also required across large railway complexes like marshalling
yards. Their parapets were often 6 ft high and formed the structural member. Such
bridges could have spans of up to 130 ft. That over the Great Eastern Railway at
Enfield Lock had an overall length, including two 63 ft spans and approaches, of
388 ft (1909).121a

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:30
11.qxd 15/11/2001 15:55 Page 230

230 Historic concrete

Figure 11.25 Alum Chine


Footbridge (1908).

Figure 11.26 Kew Gardens


Footbridge (1912).

Railway bridges Some railway engineers were notable exponents of reinforced concrete,122–124 but
its use for underline bridges was restricted. Although few new lines were built,
bridges were being replaced, and there was some concern as late as 1929 that the
potential of concrete was being ignored.125,126
Plain concrete continued to be used. Carrington viaduct (1903),11 attracted com-
ment for its expansion joints (Figure 11.27). In the 1930s precast vibrated con-
crete voussoir blocks were used in the replacement of one of Brunel’s timber
viaducts at Trenance; elsewhere the concrete superstructure was faced in natural
stone.127 The first reinforced concrete railway bridges to be built for main line
traffic were those designed by Mouchel in the Bristol area (1907–1908).29,128
Coignet’s work at Bangoed was more or less contemporary.56 These were excep-
tions, however, and reinforced concrete was normally restricted to overbridges,

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:30
11.qxd 15/11/2001 15:55 Page 231

The development of concrete bridges in the British Isles prior to 1940 231

Figure 11.27 Elevation of


Carrington Viaduct, showing
expansion joints (1903).

Figure 11.28 Construction of


the Viaduct, Valentine’s Glen,
showing a main arch after
steel centring has been struck
(c. 1932).

pedestrian and highway, rather than underline bridges. In the 1930s there was
more widespread use of reinforced concrete, seen in arched viaducts on the LMS
in Ulster (Figure 11.28), one built on a curve,129,130 and increasing use of precasting
techniques.131,132

Precast concrete The first use of precast concrete in bridges is uncertain. Its advantages were obvi-
ous, particularly when access to a site was limited and pressure of time imperative.
One early application was for railway footbridges; allegedly the Southern Region
modelled theirs on that designed by Mouchel’s at Oxshott (1908–1909).133 An exam-
ple is the replacement bridge precast at the Southern Railway’s concrete depot at
Exeter in 1923, and erected in situ by the use of a locomotive and steam crane
(Figure 11.29).133a
The outstanding early use of precast concrete was at Mizen Head, Cork. The
most spectacular of the early non-system bridges (1908–1910), it was a through
arch of 172 ft span. The ribs were precast close by, and cantilevered out from the

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:30
11.qxd 15/11/2001 15:55 Page 232

232 Historic concrete

Figure 11.29 Installation of


precast concrete footbridge
near Exeter (1923).

Figure 11.30 Mizen Head


Bridge (c. 1909).

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:30
11.qxd 15/11/2001 15:55 Page 233

The development of concrete bridges in the British Isles prior to 1940 233

abutments (Figure 11.30).134,135 Mouchel used precast beams for St John’s Hill
Bridge, over the London and South Western Railway in Clapham in 1915
(Figure 11.31), and at Wellesley Park Bridge, Gunnersbury, again over the L &
SW Railway (1922).136,137 ‘L’ beams were used for widening Limeworks railway
overbridge near Doncaster (1938),130 and more unusually precast arch ribs were
used on the North Circular to cross LNER sidings at Neasden (1938).139 The
Midland Railway (Northern Counties Committee, subsequently the LMS (NCC),
used precast beams before the First World War in Ulster, and with the availabil-
ity of heavier duty cranes in the late 1930s were able to make use of longer span
T beams. The expertize of the design team, led by W.K. Wallace, was transferred
to mainland Britain with him after 1933. It was reflected in the use of large pre-
cast units on a rail underline bridge at Northampton (1938) and the development
of prestressed beam units in the war (see Smyth, M.H. Gould, The concrete work
of the LMS NCC, (1997), Innovation in civil and structural engineering, 41–46).
Kirkcudbright Bridge (1927) had raft foundations due to poor ground and the
piers were made of precast reinforced concrete shells with mass concrete hearting.140

Figure 11.31 Placing precast


concrete beams at St John’s
Hill Bridge, Clapham (1915).

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:30
11.qxd 15/11/2001 15:55 Page 234

234 Historic concrete

Appearance The appearance of reinforced concrete bridges was a subject for discussion almost
before they were used in UK.141 Cement manufacturers promoted concrete’s
matters adaptability from early in the century.142 Its mouldability meant that it could be
used to enhance unattractive structural forms. A feature of many early bridges was
the use of moulded architectural features in contrast to today where the lines of
the structure would be used to achieve an aesthetic solution. Decoration could
enable a bridge to blend in with its surroundings. Decorative treatment could either
be carried out in situ143 or with precast elements such as parapets144 supplied by
firms like Empire Stone.145 Precast blocks were also used to simulate natural stone.
Coloured concrete, to blend in with local stonework was achieved by the use of
appropriately coloured aggregate and dyed cement.144,146,147 Natural stone fin-
ishes were simulated mechanically with pneumatic punches and bush hammer-
ing.146–150 Shuttering marks were removed by carborundum bricks.151 White
finishes were created by a final cement wash,152 and from the 1920s by the use of
Snowcrete.142
With all these expedients available it is surprising that natural stone was
employed at all, but on several prestigious bridges, and in specific locations, it con-
tinued to be employed to conceal the reinforced concrete structure.153,154 Well-
known examples of masonry cladding are King George V Bridge, Glasgow, faced
with Dalbeattie granite, and Waterloo Bridge. False arches were a regular feature
of designers of the period, King George V bridge being an obvious example.
The use of selected aggregates, natural stone and other features was encour-
aged by official policy from the mid 1920s. Possibly prompted by the quality of
design which had hitherto prevailed, it was decided to take the appearance of the
bridge into account in all applications for grant aid, a decision which apparently
had the desired effect on subsequent applications.155,156 The Royal Fine Arts
Commission advised on major crossings.
One can possibly see the effects of these changing policies in London’s arterial
roads. The pre-war Angel Road viaduct (Figure 11.8) had no aesthetic preten-
sions,48 but the neighbouring Lea Valley Viaduct (1926)157,158 the associated Lea
Navigation Bridge,159 and the Barking Viaduct (1924–27),160 with functionless
approach towers (Figure 11.32) reminiscent of Francis Thompson’s redundant
pylons at the Britannia Bridge, attracted much favourable comment at the time.
The Lea Valley Viaduct was the work of Owen Williams and Max Ayrton, and their
willingness to experiment with reinforced concrete structural forms attracted
considerable attention. The raked supports for the Wadham road viaduct, are a
welcome alternative to vertical column supports of the beam and trestle type.161

Movement matters Permanent hinges had been used in metal bridges through the 19th century, and
were introduced for masonry arches in Saxony in 1880.162 The idea was devel-
oped rapidly on the continent in the next 20 years for both stone and concrete163
but hinged arches were rarely used before the 1920s in Britain. The oldest type
used in concrete was a lead strip,164 coated in copper to avoid contact with the
concrete, placed to provide contact in the middle of the section and taking the
line of thrust, with a filler of bituminous felt or cork at each side. Dowel bars were
necessary to anchor the arch rib to the abutment. More efficient hinges were devel-
oped of curved contact type, made of cast iron or steel, with dowel bars permit-
ting some rotation. Additional reinforcement was necessary transverse to the line
of the hinges to spread the load concentration in both the arch and abutments.
Where individual ribs were involved a pin bearing in a saddle was used.
A two hinged arch was designed over the Thames near Datchet in 1924.165 This
was a two span Truscon structure of unequal span. There were concerns about
stresses from uneven loadings and temperature changes particularly on the centre
pier. A number of three hinged arches had been designed in reinforced concrete

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:30
11.qxd 15/11/2001 15:55 Page 235

The development of concrete bridges in the British Isles prior to 1940 235

Figure 11.32 Barking Viaduct


by L.G. Mouchel and
Partners (1927).

by this time, and hinges were also employed in a long viaduct where settlement
was anticipated, dividing the viaduct into rigid frames and short cantilever spans,
thus obviating the need for expansion joints.166
The Considère type of hinge was in reinforced concrete with a view to permit-
ting the arch to act as a three hinged determinate structure during construction,
and finally be closed up to become a fixed arch for live loads.50 It was intended
to deal with much of the rib shortening due to dead load before closure, much of
the shrinkage, and some plastic yield. By placing the hinge eccentrically in larger
structures an initial bending moment could be created in the arch rib to coun-
teract some of the severest moments likely to be experienced under service con-
ditions. Considère had advocated their use for the relief of secondary stresses
in arch bridges before the First World War167,168 and his English subsidiary used
them at Warrington (1909–1915) and Ogmore, Bridgend (Figures 11.4, 11.33)
(1910).49 Scott, one of Considère’s engineers, publicized their use further in his
standard text book,169 and in an article published in 1924.170 Hydraulic jacks were
later used to compensate for shrinkage and creep, as at Chiswick, to avoid the
necessity for springing hinges which would have been difficult to leave open with
the masonry facing.83,171
Aside from the specific circumstances of arch bridges, there was need to pro-
vide for movement in bridges and viaducts generally due to shrinkage, tempera-
ture effects, and possible foundation movements.92,169,172 Various types of bearings
were used such as sliding plates, not particularly suitable for heavy concrete struc-
tures, segmental cast iron rockers well suited for deck girder and through girder
bridges, heavier iron/cast steel designs, and cast steel roller bearings for heavier
structures (Figure 11.34). Considère made use of reinforced concrete rocker
bearings (Figure 11.35).173

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:30
11.qxd 15/11/2001 15:55 Page 236

236 Historic concrete

Figure 11.33 Temporary


reinforced concrete hinge at
Ogmore, Bridgend (1910).

Figure 11.34 Cast steel roller


bearings.

Figure 11.35 Reinforced


concrete rocker bearings.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:30
11.qxd 15/11/2001 15:55 Page 237

The development of concrete bridges in the British Isles prior to 1940 237

The suitable application of each type is described by Chettoe and Adams.174 With
girder bridges space was left between the girder and abutment to permit move-
ment (Figure 11.36), with copper and steel sliding plates on the bearing surfaces,
and some method of covering the space beneath the pavement. With a concrete
pavement steel protection angles were used. On longer bridges expansion joints
had to be provided. In multiple spans this would normally be done over piers.
Some filler could be placed in the wearing surface, but the slabs themselves would
have a plate sliding over steel angles (Figure 11.37). The five span bowstring girder
bridge at Kirkcudbright (1927) had bitumen filled expansion joints, with the
girders resting on cast steel rockers to accommodate settlement.175 In some cases
interlocking joints were used, as on Barking viaduct.160

Figure 11.36 Movement


joint for a concrete girder
bridge.

Figure 11.37 Cross-section


of expansion joint in
the concrete spans,
Kincardine-on-Forth Bridge.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:30
11.qxd 15/11/2001 15:55 Page 238

238 Historic concrete

Loading Prior to 1910 there were no nationally agreed design standards and highway bridge
loadings were specified by local authorities and usually related to ‘normal traffic
of the district’.176,177 The structure’s acceptability was verified by test load, gen-
erally steam rollers. The variety of early design loads is striking. The beam and
slab bridge at the Sutton Drain was designed for four wagons carrying 25 t on two
arches 8 ft apart while that at Kings Cross was designed for a uniformly distrib-
uted load of 0.2 t/ft2, and two moving loads of 8 t each at 6 ft centres, or 16 t on
each axle. Footway loads were calculated for a uniformly distributed load of
0.0625 t/ft2. Kings Bridge, Belfast was designed for a uniform distributed load of
0.05 t/ft2, and a moving load of a traction engine and three tractors, whereas
Wansbech Bridge, Stakeford was designed for a uniform distributed load of
0.05 t/ft2 but a moving load of a 30 t steam roller or two 15 t steam rollers. To
address this situation the Concrete Institute and local government engineers set
up a joint committee on highway loads. Their recommendations were published
in 1918.178 In 1922 the Ministry of Transport defined a standard train of loads
which comprised a 20 t tractor with three 13 t trailers plus a 50% impact factor
(Figure 11.38).179,180,181
These minimum standard loadings included a recommended standard width
of 10 ft for each lane of traffic. Their application to bridge design was by no means
straightforward. The justification for such heavy loading criteria, unlikely to be
met in practice, was to produce bridges which would last 100 years without need
for strengthening due to unforeseen traffic loads. The absurdity of the rigid appli-
cation of such thinking was seen at Fingringhoe, Essex, 1924 (perhaps the earliest
use of high alumina cement in bridges) where the local country road was too
narrow to contemplate driving the standard train along to test load the bridge.182
In some areas, notably Liverpool and Glasgow,183 the local regulations stipu-
lated even more severe loadings to allow for the transport of boilers and other
heavy engineering plant. Some engineers, such as W.L. Scott and Owen Williams
urged more realistic loadings should be adopted, and in his standard textbook
Scott reproduced the French and American regulations as alternative bases for
design.169 The French regulations were used by BRC as the basis for the design
of a bridge at Stretford (1925) where a loading of 70 t weight on two axles had
been specified.184
The Ministry of Transport themselves recognized that some guidance was
required on the standard loadings, and their application was discussed at the 1923
Public Works Congress.185 Williams examined the implications of the loadings on
the economic design of beam and slab concrete highway bridges.186 He concluded
that longitudinal beams were unnecessary for short spans (15–20 ft) and if they
were required transverse beams should be avoided. Williams was critical of the
loadings themselves and called for research on more realistic loadings, which would
permit ‘more and better engineering’ and more attractive bridges.187
Influence lines were regularly used as a method of determining live loads.169,188
For arch bridges with fixed supports Carpenter adopted a method based on an
assumption of uniformly distributed live loads.189 Arch design was also the subject
of an extensive series of articles by G.P. Manning (1930) and subsequent textbook.190

Figure 11.38 Standard


loading train (1922).

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:30
11.qxd 15/11/2001 15:55 Page 239

The development of concrete bridges in the British Isles prior to 1940 239

Figure 11.39 Equivalent loading curve (1931).

Scott promulgated Pigeaud’s method of analysing concentrated loads in thin


slabs.191,192
In 1931 the Ministry of Transport produced its equivalent loading curve
(Figure 11.39), and introduced knife edge loading, including an allowance for
impact.193 Perhaps in recognition of the rather unhelpful nature of the standard
loading train the curve was accompanied by an article illustrating its application
by Hargreaves,194 and given extensive treatment by Chettoe and Adams in their
standard textbooks, which were the highway engineers’ bible in the 1930s.92
E.A. Scott discussed the application to slabs.195 At the same time Westergaard (1930)
published his classic paper on wheel loads on slabs.196
BS 153,197 first issued in 1922, was essentially concerned with steel girder bridges.
The section dealing with loads (1923) included an impact factor inversely pro-
portional to the span. It was amended in 1925 to include loading trains for rail-
way and highway bridges, the latter being acceptable for MOT purposes. This
amendment was incorporated in the 1937 edition; it was not until 1949 that abnor-
mal and normal loads were specified although they were taken into account for
the design of major bridges like King George V.
Railway bridge loadings were the subject of extensive research in this period,
and previous assumptions were challenged.198–201

Load tests Early load tests were both simplistic and varied as shown in Figures 11.7 and 11.40.
The Kahn Bridges in Northumberland were test loaded with two 15 t traction
engines run side by side and in tandem,44 whereas the York Racecourse Bridge
was test loaded with 126 t of earth (Figure 11.7).45 Farnworth Bridge was test loaded
with water enclosed by the parapet girders on each side and clay dams at each
end.44 The Mellor Street bridges in Rochdale were designed for a trailer carry-
ing a Lancashire boiler, estimated load 25–30 t, and test loaded with 25 t of pig

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:30
11.qxd 15/11/2001 15:55 Page 240

240 Historic concrete

Figure 11.40 Load testing at


Ballingdon Bridge, Suffolk,
before the First World War.

iron on each pavement, together with a 12 t steam roller and an 18 t bogie of pig
iron, and a 16 t traction engine pulling a bogie of cast iron weighing 32 t. The max-
imum observed deflection was 0.04 in.37a
However appropriate test loads might be for steel or iron bridges they penal-
ized concrete bridges. If the bridge were designed for MOT loadings it was neces-
sary for the concrete to have gained sufficient strength to sustain the standard
test load after 90 days. This would necessarily involve overdesign as the concrete
of that era could be presumed to continue to gain strength after that period,
moreover the quite exceptional nature of the standard train loading could
impose severe loads, probably unrealistic in terms of service, while the bridge was
still gaining strength. In 1931, Chettoe urged that loading and deflection tests
should be postponed until the end of the maintenance period, one year after
construction.202

Service life Despite the extensive body of literature produced in recent years on bridge assess-
ment and repair, relatively few case studies have been published on the perform-
ance of pre-1940 concrete bridges.203,204 The Institution’s Panel for Historical
Engineering Works have compiled data on some of the earliest bridges, collated
as Table 11.2. Many of the bridges which have been demolished have disappeared
because of the redundancy of the structure rather than its unserviceability. Slab
vault structures such as those designed by Stead have performed well, certainly
better than his beam and slab bridges.64 The early Mouchel examples in Rochdale
have been recently assessed.37a Load tests and analysis resulted in both achieving
full 40 t assessment capacity. Most early bridges which remain show some evidence
of spalling or inadequate cover. In structures of this age this is unsurprising, but
some are known to have needed attention after a relatively short time. The Dundee
harbour bridge, in a marine environment, was first gunited in 1932 and needed
attention again in 1989. Jackfield bridge had a history of repairs dating back
to the 1930s. Recently found to have severe carbonation/corrosion problems, it
was inadequate for modern traffic loads, and despite its listed status has been

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:30
11.qxd 15/11/2001 15:55 Page 241

The development of concrete bridges in the British Isles prior to 1940 241

Table 11.2 Reinforced concrete bridges built in the British Isles 1870–1914

System/principal Number of Earliest Fate Still Visual signs Repaired Repaired Demolished/
designer known bridges known known exist of spalling pre-1940 post-1940 derelict

Mouchel 300 1901 96 59 20 2 18 37


Considère 14 1908 3 2 1 1
Coignet 9 c.1910
Expanded metal 20 [1902–4?] 1 1
Railway engineers 20 1902
Kahn/Truscon 12 1906 7 7 2 3
Local authority 27 1907 24 23 1 1
Other 24 1870 6 4 1 3 2

replaced.205 The Truscon bridge at Woodbridge (1912) has a weight restriction.206


Strengthening works made use of the Kahn reinforcement system. Although there
was much spalling due to inadequate cover, there was no evidence of chloride
attack. The Mizen Head bridge has survived well. Despite chloride ingress and
corrosion problems the concrete is generally dense and of high strength, and cover
(0.8–1.8 in).207 Considering the relatively small number of surviving examples
of bridges predating 1915 a case could be made for a comprehensive survey of
their current condition combined with archival research into the original design
specifications, to give a reasoned assessment of their performance. Listed status
of selected structures could be combined with regular monitoring to give long term
service records of reinforced concrete bridges.
There is more information available on later bridges. Lougher viaduct, a 14 span
structure of 1922 had to be extensively strengthened and gunited in 1950, and
expansion joints, omitted from the original structure, were introduced at that time.
Priory Bridge, Taunton, a Mouchel structure of similar vintage, has recently been
condemned and replaced.208 The Lea Valley viaduct has also been replaced, as
part of the North Circular improvements.209 The majority of its problems related
to leaking expansion and construction joints, although there was some honey-
combing; the load assessment was generally positive as it was at Wadham Road.
There in the absence of waterproofing the camber had provided valuable excess
cover. Jordanstown Bridge (1931), a variable depth beam and slab structure of
three spans has been strengthened with an overslab and high tensile steel link rein-
forcement in the deck beams to increase the bending and shear capacity of the
structure.210 Hampton Court Bridge, a three span arch (1931), had local weak-
nesses in the top slab which spanned between columns and onto the arch, and
has recently been strengthened.210 The more aggressive marine environment at
the Royal Tweed Bridge, Berwick, and leakage of deicing salt through joints has
caused cracking and spalling, especially where concrete was porous or honey-
combed.211 Montrose Bridge has recently been strengthened. It was suffering from
alkali aggregate reaction.212 Problems generally are said to arise from thin mem-
ber sections and lack of reinforcement cover, although Maunsell’s report revealed
that the latter problem could occur with badly built modern structures.203 Clearly
time related phenomena like carbonation depth and chloride penetration can cause
problems in older bridges, but data are lacking from which general conclusions
can be drawn on the durability of such structures. It is alleged that Considere spec-
ified a dryer mix than Mouchel, and it would be of interest to know what impact
this has had on performance.

Conclusions Most early British reinforced concrete bridges were based on foreign engineers’
designs and systems. A tremendous debt is owed to a generation of largely for-
gotten engineers who worked for specialist companies before the Second World

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:30
11.qxd 15/11/2001 15:55 Page 242

242 Historic concrete

War — T.J. Gueritte of Mouchel, Scott of Considere, Stroyer of D.G. Somerville,


Mason of BRC, Legat of F.A. Macdonald. Alert to overseas developments they
ensured that there was progress in concrete bridge design in Britain between the
wars. British work, however, remained largely derivative, and was generally ignored
by foreign commentators. The most innovative British designer of the period,
Owen Williams, did not warrant a mention in the New York Metropolitan Museum
of Modern Art publication on bridge architecture in 1949; the only British con-
crete bridges to feature were Berwick and Waterloo.213
The design standards and analytic solutions developed in this period in many
areas formed the basis of post-war codes. However many of the properties and
problems associated with reinforced concrete only became fully understood with
the post war work of the Cement and Concrete Association. Until well after the
Second World War concrete was believed to be impervious and waterproofing was
often not used. Many bridges imitated earlier metal and masonry forms;
advantage was not always taken of the potential of the new material. In retrospect
detailing was carried out without sufficient consideration of maintenance or of the
effect of shape on weathering and deterioration. Nontheless many bridges
represented here are worthy of the pride expressed by those concerned with the
work.
Space has precluded discussion of abutment and substructure design, consider-
ation of bridge hydraulics, and developments in construction techniques.92,214–216
These all merit further consideration.

Acknowledgements The author wishes to acknowledge the help received from R.J.M. Sutherland,
W.A. Smyth, David Greenfield and Clive Melbourne in developing this chapter,
and Philip Andrews for his original draft.

References 1. There are local exceptions to this pattern. Surrey for example has a relatively
large number of older 20th century bridges. Some of these works are described in
Robinson, W.P., A few descriptive notes on concrete roads in Surrey. Proc. Instn
Munic. Engrs, 1928–29, 55, 229–35.
2. Rings, F., Reinforced Concrete Bridges. Batsford: London, 1913.
3. See Chapter 4.
4. See Chapter 3.
5. See Chapter 7.
6. Ecole [Royale] des Ponts et Chaussees. Collection lithographiques, Paris, 1827,
Vol. 2.
6a. PRO Rail 38/16; Oxford Archaeology Unit (1999).
7. Chrimes, M.M., Sir John Fowler — engineer or manager. ICE Proceedings, Civil
Engineering, August 97, 135–43.
8. Tests with expanded metal. Concr. Constr. Eng., 1908–1909, 3, 265–67 (these tests
were allegedly related to work on the Central Line).
9. Early examples of reinforced concrete. Builders’ J., 20 May 1908, 438–39.
10. Bosticco, M., Early concrete bridges in Britain. Concrete, September 1970, 363–66.
*11. Wood-Hill, A., Pain, E.D., On the construction of a concrete railway viaduct. ICE
Min. Proc., 1904–1905, 160, 1–61.
12. Wilson, W.S., Some concrete viaducts on the West Highland Railway. ICE Min. Proc.,
1906–1907, 170, 304–307. See Newby for illustrations.
13. Coignet’s artificial stone. Engineering, 1869, 8, 274, 277 (Coignet’s work was widely
publicised at the 1867 Paris exhibition. While there is no direct evidence of any influ-
ence on Fowler and his colleagues, it is a remarkable coincidence that Fowler’s work
followed the exhibition in the same year).
*Important general references 14. Coignet, F., Bétons agglomeres appliques a l’art de construire. Paris, 1861 (also British
are indicated with an asterisk patent 2659, 1855).

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:30
11.qxd 15/11/2001 15:55 Page 243

The development of concrete bridges in the British Isles prior to 1940 243

15. Beckwith, L.F., Report on Beton-Coignet, its fabrication and uses. USGPO:
Washington, 1868 (report on Paris Universal Exhibition).
16. Wayss, Freytag. Das System Monier. Wayss and Freytag: Berlin and Vienna, 1887.
17. Koenen, M., Fur die Berechnung der Starke der Monierschen Cementsplatts. Cen-
tralblatt der Bauverhattung, 1886.
18. Beer, W., The Monier system of construction. ICE Min. Proc., 1898, 133, 376–92.
*19. De Courcy, J.W., The emergence of reinforced concrete 1750–1910. Struct. Engr,
1987, 65a, 315–22.
*20. Hamilton, S.B., A note on the history of reinforced concrete in buildings. National
Building Studies, Special Report, 1956, 24.
*21. Haegermann, G. et al., Vom Caementium zum Zement, Vol. 1. Bauverlag: Wiesbaden,
1964.
*22. Taylor, F.W., Thompson, S.E., Concrete: plain and reinforced, 1st–3rd edn. Wiley:
New York, 1905–1917.
23. Labrum, E.A., PHEW Newsletter, 1996, 69, 4.
24. New Civil Engineer, 25 January 1996, 4.
25. Phillips, W and T., Architectural Iron Construction, London, 1870 (gives no clue to
its application with concrete).
*26. Institution of Civil Engineers. Preliminary and Interim Report of the Committee on
Reinforced Concrete. ICE: London, 1910.
27. Twelvetrees, W.N., Reinforced concrete bridges. Concr. Constr. Eng., 1906–1907, 1,
171–80; 261–68; 341–319; 417–30 (Most examples are foreign, but Reedsmouth
Bridge, Northumberland (1904), comprised a concrete slab reinforced with
expanded metal and concrete encased RSJs (174–75).)
*28. Christophe, P., Beton arme et ses applications. Beranger: Paris, 1902 (describes many
early bridges).
*29. Much of what follows is taken from: L.G. Mouchel and Partners. Mouchel-Hennebique
Ferro-Concrete (4th, edn., 1921). Mouchel: London, 1909.
30. Mouchel-Hennebique Ferro-Concrete. List of works 1897–1919. Mouchel: London,
1920 (there is some uncertainty about the dates given, some being date of design
rather than completion/opening).
31. Cusack, P., Francois Hennebique: the specialist organisation and success of ferro con-
crete, 1892–1909. Trans. Newcomen Soc., 1984–85, 56.
32. Hennebique’s work is recorded in the house journal ‘Beton arme’.
33. Davis, A.J., Historic Shropshire bridges strengthened with Ferro-concrete. Ferro-con-
crete, 1919, x, 249–62.
34. Otter, R.A., Civil Engineering Heritage: Southern England. TTL: London, 1994:
149–50 (HEW 172) (This was essentially a repair to a brick arch design which had
gone horribly wrong).
35. Ferro-concrete bridge over the Sutton Drain, Hull. Engineering, 1903, 75, 14, 16.
36. Ferro-concrete bowstring bridge at Purfleet. Engineering, 1904, 78, 582–83.
37. Pont en beton arme a Surfleet (sic). Genie Civil, 1905, 47, 221.
37a. Platt, S.S., On some of the more recent municipal works of Rochdale. Proc. Instn
Munic. Engrs, 1913, 39, 396–408. For a recent analysis see Clapham, P.J., Young,
B.K., Supplementary load testing of Mellor Spodden concrete arch bridges. Bridge
Management, Vol. 3; Proc. 3rd Int. Conf. Bridge Manage., Guildford, 1996, 675–83.
38. Lowe, W.E., Berw Bridge, Pontypridd. Ferroconcrete, 1910–11, 2, 211–15.
38a. Brear, B., The new ferroconcrete bridge at Ironbridge, Coalbrookdale. Ferro-
concrete, 1910, 1, 19–23.
38b. Ferroconcrete bridge at Longtown, Cumberland. Engineering, 1914, 98, 341,
plate 23.
38c. Ferroconcrete bridge in Crewe Park. The Engineer, 1908, 105, 346–48.
*39. Emperger, F. von., Handbuch fur Eisenbetonbau, Vol. 3, Part 3: Bruckenbau. Berlin:
Ernst, 1908, 216–17 (lists 69 bridges with spans over 30 m the largest being that over
the Isar at Grunwald with two 70 m spans). Newby illustrates several continental
bridges.
40. Information from Dr R. Cox. HEW 3208.
41. New reinforced concrete viaduct and bridges at Dover. Concr. Constr. Eng., 1922,
17, 399–406.
42. Progress during quarter of a century. Concr. Constr. Eng., 1926, 21, 111–49.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:30
11.qxd 15/11/2001 15:55 Page 244

244 Historic concrete

43. Reinforced concrete systems: XII. The Kahn Trussed Bar. Builders’ J. Concr. Steel
Suppl., 27 March 1907, 38–40.
44. Gould, M. Early Kahn Bridges. PHEW Newsletter (ICE), 61, March 1994, 1–2. (Other
early bridges were those at Great Ayton (c. 1909–1911), Woodbridge (Guildford,
1912) and Wergins (1913). The first bridges with which Owen Williams was involved
were Aberavon and Carlisle. The Kahn bridge at Farnworth is described in Engi-
neering, 1912, 93, 286, 288.)
45. York Racecourse. Concr. Constr. Eng., 1908, 3, 487–89.
46. Concr. Constr. Eng., 1910, 5, 596–98.
47. Ferroconcrete bridge over the River Lagan. The Engineer, 1913, 115, 493–94.
47a. Ferroconcrete bridge over the River Wansbeck. The Engineer, 1909, 108, 444–45.
48. Steinberg, H.E., Twenty-one years’ development in reinforced concrete design. Concr.
Constr. Eng., 1926, 21, 93–98.
49. The Considere System of Reinforced Concrete Design. Considere: London [1912]
(also List of works, c. 1935).
*50. Considere, A., Experimental Researches on Reinforced Concrete, 2nd edn. McGraw
Hill: New York, 1906.
51. Details of this are available on microfilm in the Concrete Archive at ICE. The first
contract was awarded in 1910.
52. Other pre-First World War structures were at Dunblane footbridge, Sale, (widening,
1910), Pont Hiw (1910), Bexhill Railway Footbridge (1911), Pontycryft (1911),
Holmwood Station (strengthening, 1911), Dorking Station, Debenham (1912), Biel
(1912), Stavenston (1912) and Leeming (1913). Some foundation works for bridges
were also supplied.
53. Workman, G.C., Some recent works in reinforced concrete. Concr. Inst. Trans., 1912,
4, 18–60.
54. Edmund Coignet and Company. Reinforced concrete construction. London, c. 1913.
These included footbridges over railway tracks at Heigham-Hellesden (Norwich) and
Erith and two walkways connecting warehouses. A 180 ft long beam and trestle road
bridge in 5 spans (max span 40 ft) was erected at Mauld near Inverness, including
concrete fenders in the piers. Two 27 ft span beam and slab bridges were erected at
Saltley, Birmingham, almost certainly using parapet girders.
55. Reinforced concrete bridge at King’s Cross Station. Concr. Constr. Eng., 1912, 7,
780–84.
56. Reinforced concrete bowstring bridge for the Powell Duffryn Coal Company. Concr.
Constr. Eng., 1912, 7, 150–52.
56a. Indented Bar Company Bridges. The Company: London, 1913.
57. D.G. Somerville and Company. Handbook of standard steel and reinforced concrete
construction. The Company: London, 1908.
58. Reinforced concrete bridge at Bosmere, Suffolk. Concr. Constr. Eng., 1911, 16,
387–88.
*59. Marsh, C.F., Reinforced Concrete. Constable: London, 1904: 205–207, 377–408,
461–94.
59a Meik, C.S., Reinforced concrete in engineering structures. Proc. Instn Munic. Engrs,
1908–1909, 35, 11–46. Meik was the engineer for Purfleet pier.
60. Otter, R.A. (ed.), Civil Engineering Heritage: Southern England. TTSL: London,
1994: 82 [HEW 1087].
61. Expanded steel for reinforced concrete construction. Surveyor, 1912, 41, 864–67.
62. Reinforced concrete bridge at Exmouth. Concr. Constr. Eng., 1910, 5, 845–46.
63. Hutton, S., Exmouth’s decade of progress. Proc. Instn Munic. Engrs, 1912, 38,
235–36 I.
64. Information from David Greenfield, MICE. Three were widening schemes.
65. Candidate’s circular, ICE archives.
66. Stead, E., Reinforced concrete lining to bridges. Surveyor, 1911, 39, 865.
67. Stone, R.M., Bridge building on Sedgemoor. Surveyor, 1912, 41, 864–67.
68. HEW reports on Donyatt North (HEW 1516) and Bottle Bridge (HEW 1515).
69. Cain, W., Theory of Solid and Braced Elastic Arches. Van Nostrand: New York, 1908
etc.
70. Cain, W., Theory of Steel-Concrete Arches. Van Nostrand: New York. 2nd, edn.,
1902; 5th, edn, 1909.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:30
11.qxd 15/11/2001 15:55 Page 245

The development of concrete bridges in the British Isles prior to 1940 245

71. Ferro-concrete and the Local Government Board. Engineering, 1910, 89, 207;
editorial, Concr. Constr. Eng., 1910, 5, 1.
72. Jeffreys, R., The King’s Highway. Batchworth: London, 1949.
73. A Reinforced concrete bridge. Concr. Constr. Eng., 1921, 16, 645–646.
74. Gueritte, T.J., A Study of the views of M. Freyssinet, designer and constructor of
the ferro-concrete viaduct at Elorn-Plougastel. Societe des Ingenieurs Civils de
France. British Section, 34th ordinary meeting, 1931 (also reported in the Structural
Engineer.
75. Knight, H.S.L., Modern trends in road junction design. Roads Road Constr., 1937,
15, 294–99.
*76. Williams work is described in detail in: Cottam, D., Sir Owen Williams. Architectural
Association: London, 1986.
*77. Public Works Congress and Exhibition. British Bridges. The Congress: London,
1933: 249 (hereafter British Bridges).
78. British Bridges, 1933, p. 269.
79. Ferro-concrete (special number). Royal Tweed Bridge. Mouchel: London, 1928.
80. Three new Thames bridges. The Engineer, 1933, 156, 17.
81. Chiswick bridge: a new method of jacking open arches. Civil Engineering, June
1932, 36.
82. The New bridge over the Thames at Chiswick. Engineering, 1932, 133, 665.
83. Scott, W.L., Construction of Chiswick Bridge. Concr. Const. Eng., 1931, 26, 39–40;
1933, 29, July (Suppl.), 32–46.
84. Twickenham Bridge. Civil Engineering, March 1933, 88–91.
84a. Scott, W.L., Twickenham Bridge. Concr. Constr. Eng., 1933, 28, July (Suppl.), 14–20.
85. Somers, T.P.M., George the Fifth Bridge, Glasgow. ICE Min. Proc., 1928–29, 227,
155–86.
*86. Road bridges in Great Britain Concrete publications: London, 1939, 93 (reprinted
from Concr. Constr. Eng., January–April 1939). (Referred to hereafter as Road bridge
in Great Britain, 1939.)
87. Williams, E.O., Montrose bridge. The Engineer, June 1931.
88. British Bridges, p. 435.
89. Mason, A.P., Alveley bridge: cantilevered centering. Concr. Constr. Eng., 1937, 32,
453–59.
90. BRC appear to have begun bridge work just before the First World War, with around
14 bridges built by 1923 (BRC Structures, Manchester; BRC; 1923). Later BRC struc-
tures are described in The Concrete Way (1928–33), 5 vols.
91. Road Bridges in Great Britain. 26–29.
*92. Chettoe, C.S., Adams, H.C., Reinforced Concrete Bridge Design. Chapman & Hall:
London, 1933. (2nd edn., 1938).
93. Road Bridges in Great Britain. 1939: 45–46.
94. Road Bridges in Great Britain. 1939: 105–107.
95. Road Bridges in Great Britain. 1939: 137–41.
96. Emperger, F. von., Handbuch fur Eisenbetonbau, Vol. 3, Part 3: Bruckenbau. Berlin:
Ernst, 1908.
97. Smith, E.C., Influence line diagrams for portal girder bridges. Structural Engineer,
1934, 12, 27–40 (refers to many Mouchel designs). Bottesford Bridge (1910) was a
very rigid structure (Ferroconcrete, 1910, 2, 75–76).
98. Cement and Concrete Association. Rigid Frame Bridges. C&CA: London, c. 1937.
99. British Bridges, p. 144.
100. The design and construction of a new canal bridge. Proc. Instn Munic. Engrs,
1930–31, 57, 21–35.
101. Heywood, R., Bridge activities in Lancashire. Proc. Instn Munic. Engrs, 1939–40.
66, 172–200.
102 Reinforced concrete bridge at Wisbech. Concr. Constr. Eng., 1930, 25, 609–611.
103. McIlmoyle, R.L., Reinforced concrete railway bridges. Concr. Constr. Eng., 1930,
25, 37–45.
104. The design of flat slabs was described in Hill, A.W. Reinforced concrete flat slab
bridges. Roads Road Constr., 1938, 16, 294–99.
104a. Gould, M.H., The concrete work of the LMS NCC. Innovation in Civil and Struc-
tural Engineering, 1997, 41–46.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:30
11.qxd 15/11/2001 15:55 Page 246

246 Historic concrete

104b. Hill, W.A., The rigid frame bridge, Civil Engineering, 1932, 27, 40–43.
105. Mason, A.P., Rigid frame bridges in reinforced concrete. Struct. Engr, 1933, 11,
478–502.
106. Leeming, J.J., Some portal frames bridges in Oxfordshire. Struct. Engr, 1937, 15,
146–59; 307–402.
107. Aberuthven Bridge, Perth and Kinross. Road Bridges in Great Britain, 1939, 151.
108. Bridge of Allan. Road Bridges in Great Britain, 1939, 156–59.
109. Aboyne Bridge, Aberdeenshire. Road Bridges in Great Britain, 1939, 88–89.
110. British Bridges, 1933, 204–205 (Stowe Bridge, 1925–26); 437 (Etive Bridge, 1931–32)
Concr. Constr. Eng., 1926, 21, 126 (free Bridge, Kings Lynn); 1927, 22, 82 (Welsey
Bridge); 416 (Kirkudbright Bridge).
111. Recent bridges by Sir Owen Williams. Concr. Constr. Eng., 1929, 24, 281–88.
112. Bruce, R., The Great North Road over the Grampians. ICE Min. Proc., 1930–31,
232, 113–30.
113. Marsh, C.F., Reinforced concrete (1904). Fig. 468, p. 494.
114. Morsch, E., Der Eisenbetonbau: Der Bruckenbau Eisenbeton, 5th edn. Stuttgart: K
Wittner, 1933.
115. Patton, Professor. Pfostenfackwerk, Zentralblatt der Bauwerwaltung, 1907, 558.
116. Emperger, F. von., Handbuch fur Eisenbetonbau, Vol. 3, Part 3: Bruckenbau. Berlin:
Ernst, 1908, 255–57.
117. Possible early examples of deck stiffened vault slabs as favoured by Maillart (Fig. 4,
Newby) are Considère’s design at Dunblane, and the Cruit Island footbridge, County
Donegal, designed by Mouchel, both c. 1911. For a later example see Donald, D.A.,
Glasgow–Edinburgh road (Newbridge Bridge) Proc. Instn Munic. Engrs, 1930–31,
57, 1001–1006.
118. Buckton, E.J., Cuerel, J., The New Waterloo bridge. ICE J., 1943, 20, 145–201.
119. Palmer, P.H., Armoured or reinforced concrete. Proc. Instn Munic. Engrs,
1904–1905, 31, 343–55. A letter to Ferroconcrete 1909, 1, 132 states the first ‘fer-
roconcrete’ footbridge was at Copnor near Portsmouth, over the joint L & SWR and
LB&SCR line in 1902, but the author has been unable to verify this. It is possible
the early bridges were reinforced with expanded metal.124 (This was certainly used
for precast arched segments for three pinned arch footbridges before the First World
War. Railway Engineer, July 1914, 35, 209–213.)
120. Reigate Hill bridge. Ferro-concrete, 1912, 3, 30–31. Records in concrete archive at
ICE.
121. Railway footbridge: Kew Gardens Station. Ferro-concrete, 1912, 3, 339–41. Records
in concrete archive at ICE.
121a. Railway bridge over Enfield Lock. Ferro-concrete, 1909, 1, 102–104.
122. Institution of Civil Engineers. Engineering Conference, 1907. Section I: Railways;
Section II Harbours and docks. ICE, London, 1907.
123. Ball, J.B., Tests of reinforced concrete structures on the Great Central Railway. ICE
Min. Proc., 1915, 199, 123–32; 145–229.
124. Ball, J.D.W., Reinforced Concrete Railway Structures. Constable: London, 1913.
125. Railway underline bridges. Concr. Constr. Eng., 1929, 24, 91–92.
126. Progress during a quarter of a century. Concr. Constr. Eng., 1926, 21, 135.
127. Reconstruction of Trenance viaduct. Civil Eng., July 1939, 260–62.
128. Ferroconcrete railway structures. Railway Engr, 1908, 29, 257–60.
129. Reinforced concrete viaducts at Valentine’s Glen, Northern Ireland. Struct. Engr,
1933, 11, 199–232.
130. McIlmoyle, R.L., Reinforced concrete viaducts near Belfast. Struct. Engr, 1933, 11,
430–43.
131. Follenfant, H.G., A precast reinforced concrete underline railway bridge. ICE J,
1939., 7, 25–34 (see references 133, 136–39 below).
132. Fabricated concrete railway work. Engineering, 1932, 133, 538–40.
133. Railway over-bridge Oxshott. Ferro-concrete, 1909, 1, 99–101. This was apparently
not precast, although girders at Enfield (121a) were precast on staging above the
line and lowered into position.
133a. Precast reinforced concrete footbridge. Concr. Constr. Eng., 1924, 19, 19–22 (from
Railway Gazette 1923, 39, 580–82).
134. Information from Dr R. Cox, Trinity College (HEW 3026).

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:30
11.qxd 15/11/2001 15:55 Page 247

The development of concrete bridges in the British Isles prior to 1940 247

135. Mizen Head Bridge. Concr. Constr. Eng., 1910, 5, 847–50 (some reinforcement was
expanded metal, and part was to Ridley and Cammell patents).
136. Combination pre-cast and in-situ reinforced concrete bridge. Concr. Constr. Eng.,
1923, 18, 169–175.
137. Wills, E., Some further notes on Chiswick. Proc. Instn Munic. Engrs, 1921–22, 48,
707–709.
138. Limeworks railway bridge. Road Bridges in Great Britain, 1939, 3–7.
139. Dog Lane Bridge, Neasden. Road Bridge in Great Britain, 1939, 60–61.
140. Wolff, W.V., Reinforced concrete bridge at Kirkcudbright. Concr. Constr. Eng., 1927,
414–20.
141. Husband, J., The aesthetic treatment of bridge structures. ICE Min. Proc., 190, 145,
221–43.
142. The Cement Marketing Company’s Everyday uses of Portland cement, 5 eds
(1909–1930) records the development of its various forms, colours and textures.
143. Stephen’s Road Bridge, Bournemouth. Concr. Constr. Eng., 1922, 17, 523–24.
144. New Dorking to Reigate Road. Concr. Const. Eng., 1927, 22, 375–76.
145. Reinforced concrete bridge at Hastings. Concr. Const. Eng., 1922, 17, 537.
146. Colouring a concrete bridge. Concr. Const. Eng., 1925, 20, 666.
147. Bridge in coloured concrete. Concr. Const. Eng., 1927, 28, 36–37.
148. British Bridges, p. 105.
149. British Bridges, pp. 474–75.
150. British Bridges, p. 457.
151. British Bridges, p. 437.
152. Reinforced concrete in 1926. Concr. Constr. Eng., 1927, 22, p. 68.
153. Abbey bridge, Leicester. Concr. Constr. Eng., 1931, 26, 249–50.
154. British Bridges, pp. 464–65.
155. Ministry of Transport. The Design of road bridges. Circular No. 224, Roads. Report
on the Road Fund 1924–1925, 43.
156. Report of the Road Fund, 1925–1926, 19–20. Reports on progress on the issue of
the quality of bridge design.
157. Concrete in 1926. Concr. Constr. Eng., 1927, 22, 1.
158. British Bridges, pp. 106–107
159. British Bridges, p. 192.
160. Welch, G., Viaduct and bridge at Barking. Concr. Constr. Eng., 1927, 22, 243–51
(this viaduct was designed by Mouchel’s, with Ayrton responsible for the architec-
tural treatment).
161. Wadham Road viaduct. Concr. Constr. Eng., 1930, 25, 107–109.
162. Emperger, F. von., Handbuch fur Eisenbetonbau, Vol. 3, Part 3: Bruckenbau,
Chapter 1. Other sources suggest Erlanch in 1873.
163. Molitor, D.A., Three-hinged masonry arches; long spans especially considered. ASCE
Trans., 1898, 40, 31–85.
164. Marsh, C.F., Reinforced Concrete, 1904, 205–206.
165. New reinforced concrete bridge over the Thames. Concr. Constr. Eng., 1924, 19,
775–78.
166. Stroyer, R.N., Hinges in reinforced concrete structures. Concr. Constr. Eng., 1924,
19, 207–212.
167. Considere, A., Spirally armoured concrete. Engineering, 1910, 89, 578–79.
168. Ferro-concrete bridge at Chateau Thierry. Engineering, 1910, 90, 327, 329.
*169. Scott, W.L., Reinforced Concrete Bridges. Crosby Lockwood: London, 1925–31.
(3 eds.)
170. Scott, W.L., Secondary stresses in reinforced concrete arched bridges. Concr. Const.
Eng., 1924, 19, 9–11.
171. Chiswick Bridge: lifting the arch. Roads Road Constr., 1934, 12, 22. The idea was
taken from Freyssinet.
172. Emperger, F. von., Handbuch fur Eisenbetonbau, Vol. 3, Part 3: Bruckenbau,
Chapter 1.
173. Scott, W.L., Reinforced concrete bridges, 1928, 331–33.
174. Chettoe, C.S., Adams, H.G. Reinforced concrete bridge design, 1933, 344.
175. Wolff, C.V., Reinforced concrete bridge at Kirkcudbright. Concr. Constr. Eng., 1927,
22, 415–20.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:30
11.qxd 15/11/2001 15:55 Page 248

248 Historic concrete

176. Henderson, W., British highway bridge loading. ICE Proc., 1954, 3, 2, 325–73.
177. Wheel loads and tyre widths. Surveyor, 1911, 39, 556–59.
178. Concrete Institute. Loads on highway bridges; report of a joint committee,
London, 1918 (2nd edn., 1926 by Institution of Structural Engineers).
179. Ministry of Transport. Standard load for highway bridges, MoT: London, June
1922.
180. Ministry of Transport. Report on the Road Fund 1921–22. London, 1923.
181. Reinforced concrete road bridges. Concr. Constr. Eng., 1923, 18, 721–33.
182. Weaver, L.T., Fingringhoe bridge, Essex. Aluminuous cement in bridge construc-
tion. Concr. Constr. Eng., 1924, 19, 213–16.
183. King George V Bridge, Glasgow. Concr. Constr. Eng., 1928, 23, 2.
184. Warwick Road Bridge, Stretford. Concr. Constr. Eng., 1925, 20, 510–12.
185. Hawkins, J.F., Mitchell, C.G., General construction of bridges. Public Works, Roads
and Transport Congress, 1923, 218–35.
186. Williams, O.E., Beam and slab concrete highway bridges. Instn Munic. Engrs J., 1926,
report in Concr. Constr. Eng., 1926, 21, 359–363.
187. Williams, O.E., Letter on the Bridge Stress Committee. Concr. Constr. Eng., 1929,
24, 180.
188. Mason, A.P., The Design of reinforced concrete arch ribs: an accurate shortcut
method. Concr. Constr. Eng., 1924, 19, 143–50.
189. Carpenter, H., Design of arched bridges with fixed supports. Concr. Constr. Eng., 1930,
25, 612–22; 673–84.
190. Manning, G.P., Reinforced Concrete Arch Design. Pitman: London, 1933 (originally
published in Concr. Constr. Eng., 1930–31).
191. Scott, W.L., Design of reinforced concrete slabs. Concr. Constr. Eng., 1930, 24, 167,
221–93.
192. Pigeaud. Recherches sur les plaques rectangulaires appuyees a leur pouvoir. Annales
des Ponts et Chaussees, 1921, 5–47.
193. Ministry of Transport. Standard Loading for Highway Bridges. HMSO: London,
1931.
194. Hargreaves, G.H., The application of the equivalent loading curve for bridges. Concr.
Constr. Eng., 1931, 26, 661–69.
195. Scott, E.A., Tables for slabs designed to Ministry of Transport load and stress require-
ments. Struct. Engr, 1934, 12, 382–92.
196. Westergaard, H.M., Computation of stresses in bridge slabs due to wheel loads.
Public Roads, 1930, 11, 1.
197. BS 153: British Standard Specification for Girder Bridges: Parts 1 & 2 — 1922: mate-
rials and workmanship; Parts 3, 4 & 5 — 1923: loads and stresses, details of con-
struction, erection; Parts 3, 4 & 5 — 1923: appendix no. 1 (1925): tables of British
Standard unit loadings; Parts 1 & 2 — 1933; Parts 3, 4 & 5 — 1937
198. DSIR. Bridge Stress Committee. Report. HMSO: London, 1928.
199. Gribble, C., Impact in railway bridges with particular reference to the Report of the
Bridge Stress Committee. ICE Min. Proc., 1928–29, 228, 46–153.
200. Inglis, C.E., Impact in railway bridges. ICE Min. Proc., 1931–32, 234, 350–444.
201. Foxlee, R.W., Greet, E.H., Hammer blow impact on the main girders of railway
bridges. ICE Min. Proc., 1933–34, 237, 239–418.
202. Chettoe, C.S., Testing concrete bridges. Concr. Constr. Eng., 1931, 26, 5–7.
203. Wallbank, E.J., The Performance of Concrete in Bridges. HMSO: London, 1989.
204. Mallett, G.P., Repair of Concrete Bridges. Thomas Telford: London, 1994.
205. Thomas, W.H. & Partners, Free Bridge Jackfield. Condition report. March 1986.
Ref 928/85/SJP/JM
206. Cogswell, G., Herbert, A.P., A25 Woodbridge (old) Guildford Bridge. Highways and
Transportation, September 1991, p. 7.
207. MacCraith, S., Performance of an 80-year-old reinforced concrete bridge in an
extreme environment. Proceedings of the 3rd International Conference on Corro-
sion of Reinforcement in Concrete Construction (Society of Chemical Industry) 1990,
p. 188.
208. Information from David Greenfield, MICE. The original bridge is described
in: New reinforced concrete bridge at Taunton. Concr. Constr. Eng., 1923, 18,
20–24.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:30
11.qxd 15/11/2001 15:55 Page 249

The development of concrete bridges in the British Isles prior to 1940 249

209. Elliott, D.W.C., Inspection, load assessment and repair of Lea Valley viaduct and Wad-
ham Road viaduct. Bridge Maintenance 1993, 2, 625–33.
210. Lockwood, S.E., and others. Strengthening concrete bridge decks-increasing the shear
capacity. Bridge management 3: Proc 3rd Intl Conf Bridge management, 1996.
173–79.
211. Palmer, J., Cogswell, G., Bridge strengthening in practice. Bridge maintenance 2,
1993, 912–20. This was a Mouchel structure, described in a paper by T.J. Gueritte
in Concr. Constr. Eng., 1933, July (Suppl.), 28, 21–31.
212. Wood, J.G.M., Angus, E.C., Montrose bridge: inspection, assessment and remedial
work to a 65-year-old bridge with AAR. Structural Faults and Repair 95, Vol. 1,
pp. 103–108.
213. Mock, E.B., The Architecture of bridges. Museum of Modern Art, New York, 1949.
214. Adams, H.C., The Design of bridge substructures. Public Works Congress, 1931,
187–241.
*215. Legat, A.W., and others, Design and construction of reinforced concrete bridges. Con-
crete Publications, 1946 (has much on erection techniques). This was based on arti-
cles originally published in Concrete and constructional engineering 1933.
216. Wynn, A.E., Design and Construction of Formwork for Concrete Structures. Con-
crete Publications: London, 1926 (2nd edn. 1939).

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:30
12.qxd 15/11/2001 15:56 Page 251

12 UK concrete bridges since 1940

William Smyth
Introduction This chapter on concrete bridges since 1940, concentrates particularly on the early
postwar period and deals with more recent bridges briefly and mainly from an
evolutionary point of view. A paper on prestressed highway bridges in the UK,
published in the Proceedings of the Institution of Civil Engineers in 1989,1 deals more
fully with the more recent bridges as do some of the other references appended
to the chapter. As well as the paper referred to above, an indispensible reference
is the book Modern British Bridges2 which contains brief descriptions, often with draw-
ings and photographs, of many bridges built or under construction between the
end of the war and 1964.

Concrete bridges The outstanding fact about concrete bridges after the Second World War is the
way in which prestressed concrete rapidly became the dominant material for all
after the war but the smallest bridges or those which were so large that they had to be built of
steel, and even at this end of the scale the spans achievable in concrete have been
increasing all the time.
The advent of prestressing meant that construction techniques formerly only
used for steel bridges became feasible in concrete and were quickly made use of
on the Continent, where large numbers of bridges had to be rebuilt after the land
wars which had been fought there. By 1950 bridges by Freyssinet and Magnel had
been built by assembling precast segments joined with concrete or mortar,
and also Finsterwalder’s first in-situ concrete box girder bridge built by cantilever
construction. During the early 1960s the first incrementally launched concrete
bridge was built in Germany, and the first concrete bridge to combine cantilever
construction with precast segments jointed with epoxy resin, in France. In each
case it took some time before the technique was used for bridges in the UK,
possibly because of the lack of opportunity.*
Prestressing techniques underwent significant development during the period
and, together with increasingly higher strengths of concrete and steel and further
developments of construction techniques, have led to substantial economies in
materials and costs and have also produced changes in the character of bridges.
Other significant changes which have affected the design of bridges, although in
less obvious ways, are a considerable increase in understanding of the behaviour
of the ground on which bridges are founded and in the ability to predict its behav-
iour, and improved understanding of structural behaviour, as well as methods of
calculating the way in which a structure will behave. The advent of computers and
later of sophisticated electronic calculators increasingly enabled more and more
*Box girders were certainly not complex calculations to be made with greater and greater ease. Ancillary elements,
new to the UK and a reinforced such as bearings and expansion joints, have also changed over the period. The
concrete bridge was built by
balanced cantilever construction in changes have not generally been dramatic, but a series of evolutionary steps adding
1936.3 up to considerable changes over time.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:54
12.qxd 15/11/2001 15:56 Page 252

252 Historic concrete

Figure 12.1 Reconstruction of bridge over London and North Eastern Railway in Yorkshire (1943). The second
prestressed concrete bridge deck in Britain.

Early prestressed Very few bridges were built in Britain during the war and not many immediately
after it, and it was not until the late 1950s that bridges were being built in any
bridges in numbers. The government had anticipated that large numbers of bridges would
Britain — the be destroyed and in 1940 the Ministry of War Transport had a stock of emergency
1940s and 1950s bridging beams made of prestressed concrete. These beams were either I or
box section in various lengths, made by the long-line process. When the military
situation improved, it was decided to use some of them for urgent permanent
bridgeworks. In 1943 they were used for the reconstruction of two bridges
carrying roads over railways, one in Lancashire using 44 ft long beams of box
section, and one in North Yorkshire using 54 ft beams of I section (Figure 12.1),
and plans for reconstruction of another such bridge were being considered.4
A stock of these beams remained after the war and at least four other uses are
recorded.5*
The emergency bridging beams were designed by Dr Mautner of the Prestressed
Concrete Company which had been established in the UK in 1938 as a licensee
of the Freyssinet organization and a subsidiary of L.G. Mouchel and Partners.
Mouchel as consulting engineers and Dr Mautner as designer of the prestressing
were the designers of Nunn’s Bridge (Figures 12.2 and 12.3), built in 1948 near
Boston, the first in-situ prestressed road bridge in Britain.6 When Dr Mautner
died in 1949, A.J. Harris, who had been working with Freyssinet in France,
succeeded him.
*According to Sir Alan Harris The Freyssinet system and the Hoyer system for long-line pre-tensioning were
these beams were so cheap that it
was impossible to compete with
the first to be used for bridges in England, but other systems were in use by the
them, which implies a significant early 1950s. Walley7 mentions that ten years earlier the only British one was the
number. Lee-McCall system, and records several others which were in use by 1962.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:54
12.qxd 15/11/2001 15:56 Page 253

UK concrete bridges since 1940 253

Figure 12.2 Nunn’s Bridge,


Fishtoft, the first in-situ
prestressed concrete road
bridge in Britain.

Figure 12.3 Nunn’s Bridge, general arrangement. The transverse beams are typical of early post-war bridges.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:54
12.qxd 15/11/2001 15:56 Page 254

254 Historic concrete

Figure 12.4 Rhinefield


Bridge in the New Forest.

Figure 12.5 Rhinefield Bridge,


general arrangement and
prestressing.

A paper of 19498 includes comprehensive lists of prestressed bridges built up


to that time and an extensive bibliography. The British references are to the
two bridges of 1943, Nunn’s Bridge and ‘the Adam viaduct, near Wigan … the first
railway under-bridge to be built of pre-stressed concrete’ using precast I beams;
also to designs for several slab under-bridges using the Freyssinet system, whose
construction was to start later in the year for British Railways.
The early prestressed bridges were mostly fairly simple and small. Several
interesting ones are mentioned in a paper of 1952 on bridges in Hampshire.9 The
earliest of these is the Rhinefield Bridge (Figures 12.4 and 12.5) carrying a minor
road over a stream and the first of a number of bridges in the New Forest
constructed of precast, post-tensioned units using the Freyssinet system and
stressed together transversely. The slight vertical curvature of the Rhinefield Bridge
made the longitudinal cables straighter which reduced friction and produced a
very attractive little bridge which still serves its purpose well. Figure 12.6 is from
the same paper.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:54
12.qxd 15/11/2001 15:56 Page 255

UK concrete bridges since 1940 255

Figure 12.6 Proposed


reconstruction of A30 bridge
over the Test at Stockbridge.
The bridge seems not to
have been built to this
design.

Because of the shortages of materials after the war and the consequent require-
ment for licences to use reinforcing steel there was an incentive to use prestress-
ing wherever possible and transverse stressing of bridge decks was quite common.
In some of these transversely stressed bridges the surfacing was laid directly on
the precast units. Segmental construction was also used, not in large cross sections
as in later times but for individual beams or portions of a slab, and there were
interesting combinations of precast segments with in-situ concrete. A small bridge
at Martinhoe in North Devon was destroyed by floods in 1952 and had to be quickly
rebuilt on a site not accessible for precast beams. The deck was made from precast
trough units side by side and end to end, erected on falsework (Figure 12.7) and
stressed together longitudinally with unsheathed cables. Sheathed cables were also
provided so that the deck could again be stressed longitudinally and also laterally,
after the troughs had been filled with concrete.10 Freyssinet prestressing was used,

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:54
12.qxd 15/11/2001 15:56 Page 256

256 Historic concrete

Figure 12.7 Martinhoe


Bridge over River Heddon at
Hunter’s Inn, N. Devon,
replaced after the floods of
1952. Tensioning the main
cables.

Figure 12.8 Cross-section of


Barbrook Bridge, N. Devon.
Another bridge which was
replaced after the floods of
1952.

but as the deck was designed by Gifford and built by Udalls it may have been a
forerunner of the Gifford-Udall system. Other bridges reconstructed because of
the floods are described by Criswell (Figure 12.8).11
Some interesting footbridges were built. The first prestressed fixed arch in the
world was claimed to be a portal frame footbridge across the Cherwell at Oxford
(Figures 12.9 and 12.10) in which the only precast elements were the prestressed
concrete planks forming the walking surface (since replaced by an in-situ concrete
slab).12 Others were a continuous prestressed concrete footbridge, constructed for
the Festival of Britain,13 which was tested to destruction after the Festival ended,14
and the St James’s Park Bridge (Figure 12.11), a three span continuous beam
bridge constructed with the end spans initially acting as cantilevers.15 The Eel Pie
Island Footbridge had a mainly precast structure consisting of cross beams and
segments of main beams at the same level, stressed together through dry packed
mortar joints.16
Two service bridges made from precast segments should be mentioned. A pipe
bridge at Gunthorpe near Nottingham (since demolished) consisted of an arch
made from segments stressed together and tied by a deck, also made from pre-
cast units, in the form of a trough; the support at one abutment had Freyssinet
hinges top and bottom.17 The other one is a trough shaped beam bridge consist-
ing of two beams made from precast segments of I section post tensioned by the
BBRV system, with precast cross-beams and slabs forming the bottom of the U.18
A few reinforced concrete bridges were completed during the war or built a few
years after it.19,20 By far the most notable was Waterloo Bridge, completed in
1942.21

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:54
12.qxd 15/11/2001 15:56 Page 257

UK concrete bridges since 1940 257

Figure 12.9 Portal frame footbridge over the Cherwell at Oxford. Layout of prestressing cables on what was possibly
the first statically indeterminate prestressed bridge in Britain.

Figure 12.10 Footbridge


over the Cherwell.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:54
12.qxd 15/11/2001 15:56 Page 258

258 Historic concrete

Figure 12.11 St. James’s


Park Bridge.

Early prestressed Railway civil engineers realized very early the advantages of prestressed concrete.
Not only were the first prestressed bridges in Britain railway overbridges, but
railway bridges the first railway underbridge in prestressed concrete, the Adam Viaduct,22 was
built in 1946, and the first partially prestressed bridge ever to be constructed
was a railway overbridge built in 1949. Partial prestressing* was a controversial
subject. Freyssinet objected to it, and (possibly somewhat later) the UK Ministry
of Transport did not allow tensile stresses in prestressed concrete at working
loads. Its chief advocate was Dr P.W. Abeles of the Eastern Region of British
Railways.
A paper of 195123 deals with work being done by British Railways on railway
bridges.
Three kinds of railway underbridges were being pursued:

1. Simple constructions comprising precast, prestressed beam and slab units.


A range of these was being developed and some had already been built. The
design for a bridge at Barmouth24 using 12-wire cables is illustrated and one
for a bridge at Ystalyfera using Macalloy bars is mentioned.
2. Precast, prestressed floor units for use in association with steel main girders,
to provide a shallow form of construction, lending itself to rapid site erection,
in replacement of existing bridges. Steel end plates were used for anchorages
and to allow for connection to the main girders. One of the two designs
illustrated has provision for stressing in the longitudinal direction of the
bridge (Figure 12.12). A later paper shows developments of this structural
type.25
3. A proposal for precast, prestressed concrete channel type units, for single track
bridges, to be site assembled and post-tensioned to form bridges with com-
pletely composite main beams and floor.

Prestressed overbridges designed by British Railways are also covered in the


paper. These included slab bridges with pre-tensioned slab units transversely post-
*That is limited tensile stress
allowed under live load and tensioned together without any topping (Figure 12.13), and partially prestressed
passive reinforcement provided. structures.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:54
12.qxd 15/11/2001 15:56 Page 259

UK concrete bridges since 1940 259

Figure 12.12 Prestressed


concrete railway underbridge
deck slab for use with welded
plate girders.

Figure 12.13 Prestressed


slab construction for road
bridges over railways (1951).

A later paper deals with a bridge at Manchester made from precast box units
assembled on falsework with in-situ diaphragms and then stressed.26 Other refer-
ences are noted.27–29
A 1959 paper gives more information on partially prestressed railway
bridges.30 The deck of the first one in 1949 consisted of inverted T beams and the
in-situ concrete infill contained mild steel reinforcement (Figure 12.14). A similar
deck built in 1950 incorporated untensioned wires in the precast beams instead
of the bars in the in-situ concrete. Somewhat later high-voltage electrification of
railways required the depth of overbridge decks to be reduced to the absolute min-
imum; a new design of deck, partially prestressed and prestressed in two stages
was developed which allowed depth to span ratios as low as 1/30 (Figure 12.15).
Two such bridges had been built before 1959. In the second and more developed
of these groups of precast inverted T beams, including unstressed as well as pre-
tensioned wires, were assembled fully supported near the bridge; duct formers
and some transverse reinforcement were placed; in-situ concrete infill/topping was

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:54
12.qxd 15/11/2001 15:56 Page 260

260 Historic concrete

Figure 12.14 Details of road


over rail bridge at Buck Lane
(1949). This was the first
partially prestressed bridge.

Figure 12.15 Details of road


over rail bridge at Colchester
about 1958. The ‘wafer slab’
design was developed to
allow exceptional span/depth
ratios (up to 1/30) for railway
electrification. The precast,
partially prestressed inverted
T-beams were propped dur-
ing casting of the in-situ con-
crete, and the whole was
then stressed using the Mag-
nel-Blaton system.

placed to make a slab unit, Magnel-Blaton cables were inserted and tensioned and
the slab units were lifted into place by crane. British Railways seems to have lost
interest in partial prestressing after 1963.31
Railway bridges were also built at industrial sidings. A trough-shaped through
bridge built at a Rotherham steelworks in 1952, using the Lee-McCall system, has
a skew of 58.5° and a span of 160 ft over the River Don and was thought at the
time to be the largest span railway underbridge in the world32 (Figure 12.16). See
also Ref. 33.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:54
12.qxd 15/11/2001 15:56 Page 261

UK concrete bridges since 1940 261

Figure 12.16 Railway bridge


at Rotherham steelworks,
over River Don (1952). At the
time this bridge was thought
to be the largest span railway
bridge anywhere (span 160ft.,
skew 58.5°. Lee-McCall
system.

Railway bridges Minimum disruption to train services has always been a major criterion for rail-
way engineers, and one which is particularly difficult with underbridges. One of
after 1960 the main techniques developed is the construction of a new deck alongside the
existing track, followed by sliding it into position. A paper of 1961 describes three
interesting bridges: one consisting of four separate prestressed box girders rolled
in to position; one of transverse prestressed slabs post-tensioned together; the third
is a cast in-situ box girder whose top flange is a platform, with cantilevers of the
bottom flange carrying tracks on either side.34
Sliding in of concrete railway bridges and some larger and interesting box
girder and arch bridges built in the 1960s are described in other references.35,36
One of the bridges was recently reassessed.37
The first incrementally launched bridge in the UK was a railway overbridge
(1977).38 In the early 1980s a unique prestressed concrete cable stayed railway
underbridge was built across the M25 at a considerable skew.39

Larger bridges — The first substantial prestressed concrete road bridge was the replacement for
Northam Bridge over the River Itchen at Southampton (1954, Figures 12.17
1954 onwards and 12.18).40 This bridge combined up-to-date technology with a style belonging
to well before the war. It is typical of its time in that the main deck structure con-
sists of beams rather narrowly spaced and has transverse diaphragms. The beams
were precast on site using deflected cables and the deck was made continuous
for live and superimposed loads by means of in-situ diaphragms between the
ends of the beams and precast prestressed slabs clamped between the tops of
the beams by transverse stressing over a length where the flanges of the tees were
omitted.
After the Northam Bridge prestressed concrete was used for larger and larger
bridges, using a variety of forms and construction techniques. Cavendish Bridge
(1956), Clifton Bridge41,42 and Bridstow Bridge43 were all cantilever and suspended
span bridges using precast beams for the suspended span. Clifton Bridge
(Figure 12.19) was probably the first postwar concrete bridge in the UK to use
box girders and free cantilever construction. The anchor spans and cantilevers

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:54
12.qxd 15/11/2001 15:56 Page 262

262 Historic concrete

Figure 12.17 Northam


Bridge, Southampton, the
first major prestressed bridge
in Britain.

Plan and longitudinal section from and including the south abutment to and including pier 2

Cross-section at mid-span

Figure 12.18 Northam Bridge.The frequent diaphragms are typical of the period. Near the supports substantial parts
of the flanges of the T-beams were removed to allow the junction (continuity) slabs to fit between them. Freyssinet
system. Main beams precast, pre-tensioned on site, with deflected tendons. Post-tensioned through diaphragms after
junction slabs placed.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:54
12.qxd 15/11/2001 15:56 Page 263

UK concrete bridges since 1940 263

Figure 12.19 Clifton Bridge, Nottingham. Again the frequent diaphragms can be seen, even in the box-girder. (a) Typi-
cal cross-section. (b) Longitudinal section.

are box girders and only the cantilevers were built by free cantilever construction.
The anchor spans and cantilevers of the Medway Bridge (1963),44,45 also a
cantilever and suspended span box girder bridge, were built by free balanced
cantilevering, and its main span of 152 m was by far the largest concrete span in
the UK until the Orwell bridge (1982, 190 m span).46 It may also have been the
first to omit diaphragms except at supports.
Other bridges of interest are: Donnington Bridge,47 with closely spaced portal
frames, partly precast, each leg formed by a tie and a strut. Winthorpe Bridge48
with three span continuous multiple boxes; Taf Fawr Bridge49 with precast seg-
mental I beams built by cantilever construction and connected by in-situ flanges
to form a three-celled box;50 and Wentbridge Viaduct51,52 a three-span box girder
with raking legs which have concrete hinges at top and bottom.
In 1963 the Ministry of Transport held an open competition for the design of
a bridge to carry the M1 over the River Calder, near Leeds. The crossing was on
a considerable skew and the bridge had to be capable of coping with mining
subsidence. Three of the four prize winning designs had prestressed concrete
decks and the other was prestressed concrete except for its composite steel and
concrete suspended span.53,54
Before the end of the 1960s prestressed concrete had superseded reinforced
concrete for all but small bridges and arches, and the box girder had become a
dominant structural form for large bridges in concrete and steel, because of its
structural efficiency and economy. The scale of bridges increased, not only in span
but also in cross section, large single cell box girders with wide edge cantilevers

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:54
12.qxd 15/11/2001 15:56 Page 264

264 Historic concrete

replacing the earlier type with several small cells and smaller edge cantilevers, or
none in the earliest bridges. Fewer cantilever and suspended span bridges were
built and more of the larger bridges were continuous beam structures.

Reinforced concrete bridges


Particularly in the 1950s some larger bridges were still built wholly or partly of
reinforced concrete. For obvious reasons a number of them were large arches, gen-
erally with open spandrels and often with twin arch ribs, typified by the Lune
Bridge carrying the M6 (1959).55 The reinforced concrete arches of the Nant Hir
and Taf Fechan Bridges46 were built a few years later by cantilever construction
using temporary cables, as described by Hansen.47 The beam bridge carrying the
A1 over the River Wharfe56 has cantilevers and anchor spans of reinforced con-
crete although the suspended span uses precast post-tensioned I beams. The long
approach viaducts either side of the Queenhill Bridge (1961) were reinforced con-
crete boxes.57

Long motorway-type viaducts


In the 19th century long viaducts were mostly railway viaducts. From the end of
the 1950s a number of long road viaducts were built, many in prestressed
concrete. Some of the ones built from precast segments are mentioned elsewhere
in this chapter, from the Hammersmith Flyover onwards. The Chiswick-Langley
Viaduct58 has tee-headed piers carrying precast pre-tensioned inverted tee beams
with an in-situ concrete slab, and others have been built from other types of
precast beams. Others were constructed in situ using span-by-span construction,
such as Gateshead Highway (1971),59 a serious attempt to visually integrate the
junctions of the slip road ramps with the main structure. More recently cantilever
construction has been used, mostly as post-tensioned box girders of various
configurations, some multi-cell and some as groups of single cells.

Larger precast segmental bridges


The Chiswick Flyover (1959) was built from concrete beams one third of the span
in length, assembled on temporary supports and post-tensioned.60 Precast
segmental construction gained a new dimension with the construction of the
Hammersmith Flyover (1961) (Figure 12.20a, b), the first spine box with edge can-
tilevers, and a bridge of unique character. Alternating spine segments and
diaphragm/cantilever bracket units were assembled on falsework with in-situ con-
crete between them before stressing.61 Mancunian Way62 and Westway63 (Figure
12.20c) which followed used the same construction technique with simpler arrange-
ments of precast segments incorporating the spine box with the edge cantilevers
and with diaphragms only at supports. London Bridge64,65 used a similar tech-
nique, but with units suspended from a gantry. Epoxy resin joints were first used
on a British bridge, for joining precast segments which were assembled on false-
work, at Rawcliffe Bridge (1968).66 More recent bridges have used match cast seg-
ments, with a thin layer of epoxy resin in the joints and built without falsework
by cantilever construction. The first two British bridges to use match cast segments
in cantilever construction were Byker Viaduct67, a railway bridge for the Tyne and
Wear Metro, which used both balanced and progressive cantilevering, and the
M180 bridge across the Trent.68,69

Incremental launching
The first incrementally launched bridge in the UK was Shepherd’s House Bridge
carrying the A4 over a main line railway (1977),32 and several others have since
been completed.70

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:54
12.qxd 15/11/2001 17:24 Page 265

UK concrete bridges since 1940 265

(a)

(b)

Figure 12.20
(a) Hammersmith flyover:
double cantilver precast units
59 ft long supporting 24 ft
carriageways. (b, c) Westway:
double cantilever precast
units 94 ft long supporting 41
ft carriageways. (c)

Smaller bridges Although many quite small bridges were built of prestressed concrete, many
were built of reinforced concrete,71 for example the Kingsgate Footbridge, which
since 1957 integrates its structural form and construction method with its appearance (1963).72
Since road construction started seriously in 1957 vast numbers of bridges have
been built over grade separated roads. Some of them carry important roads
or railways, but many are carrying minor roads, farm accesses, footpaths or
bridleways.
On the Preston Bypass, the first section of M6 which was started in 1956 the mix
of bridges is shown on Figure 12.21.73 By 1959 the bypass had been completed
and work was under way on the M5 in Worcestershire,74 the A1 in the West
Riding of Yorkshire,75 the Maidstone Bypass in Kent.76 The general picture was
of a mixture of prestressed and reinforced concrete and sometimes composite steel
and concrete bridges. However this was not always the case as can be seen from
Figure 12.22 which shows standardized solutions to typical motorway bridging
problems on various motorways, but mainly the M1.77 Eleven types of overbridge
are shown in the original figure. One of them (representing four bridges) has
universal beams and lightweight concrete. The other ten types (representing 178
bridges) are all of reinforced concrete cast in situ.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:54
12.qxd 15/11/2001 15:57 Page 266

266 Historic concrete

Figure 12.21 Bridges on Preston Bypass. A mix of types and materials, as on many early motorways.

Some of these reinforced concrete bridges on the M1 are unusual. The first one
in the figure is the anachronistic-looking propped portal frame slab with mass
concrete legs/abutments which is such a feature of the southern part of the motor-
way; the deck thickness is stated to be 1/36 of the span. The following four appear to
be a type with hinges in the deck, some of which on the M6 were recently
subjected to assessment.78 On the early motorways a great variety of forms was
used, simply supported or continuous beams, portal frames, occasional arches.
Cross-sections might be beam and slab, solid slabs or voided slab and various com-
binations of precast and in situ, reinforced and prestressed. Other papers on early
motorway bridges may also be of interest79–81 and Modern British Bridges contains a
number of early overbridges.2

Precast bridges Smaller concrete bridges often had decks consisting of precast concrete slabs,
others of in-situ slabs supported by and working compositely with concrete beams,
and standard either reinforced or, if prestressed, often precast and pre-tensioned. Precast beams
beams were originally designed ad hoc but towards the end of the 1950s and early 1960s

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:54
12.qxd 15/11/2001 15:57 Page 267

IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:54
Delivered by ICEVirtualLibrary.com to:
Figure 12.22 Standardized solutions to typical motorway bridging problems. This table shows a number of interesting reinforced concrete bridge types,
UK concrete bridges since 1940

and indicates that some designers saw no place for prestressing as late as 1971.
267
12.qxd 15/11/2001 15:57 Page 268

268 Historic concrete

the Prestressed Concrete Development Group, under the aegis of the Cement
and Concrete Association, prepared standard sections. These and subsequent
developments have been well covered by Sriskandan82 and Taylor.83

Estuarial crossings Most steel bridges incorporate significant concrete elements. Of the four major
estuarial road crossings by suspension bridge which were built in Britain after 1940,
two have concrete towers. The Tamar Bridge84 also has a concrete deck slab. The
tapering slip-formed concrete towers of the Humber Bridge are themselves major
constructions.85

Ancillary items During the postwar period there have been considerable changes in such things
as expansion joints, bearings and hinges; involving the use of new materials to
produce more effective and often cheaper designs. Laminated rubber bearings
were available by 1957 when they were used on Pelham Bridge.86 Pot bearings
were developed in 1959 and sliding bearings using PTFE were first used at the
beginning of the 1960s.87 Older types also were improved, with rollers of increased
surface hardness and the use of stainless steel.
Concrete hinges were used on several bridges, and a prototype for Wentbridge
Viaduct48 was tested at the Cement and Concrete Association (C&CA) (Figure 12.23).
A type of triple hinge was developed for Wichert Truss bridges in Yorkshire
designed to cope with mining subsidence.88 Mesnager type hinges were also some-
times used, for instance in the decks of some motorway overbridges75 and between
the deck and main piers on the Medway Bridge (Figure 12.23).41
Expansion joints with more effective sealing became available. However the prob-
lems of leaking joints and the importance of detailing in the vicinity of joints were

Figure 12.23 Two types of concrete hinge. The ones for Wentbridge Viaduct were precast and those on Medway
Bridge were cast in situ.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:54
12.qxd 15/11/2001 15:57 Page 269

UK concrete bridges since 1940 269

generally not appreciated as fully as in more recent times. Many bridges consisted
of several (sometimes many) spans of simply supported beams with a joint over
every support and these have often leaked very badly. Sometimes new materials
or methods (e.g. epoxy nosings) seemed promising in trials but were less successful
in general use.

Structural analysis The history of structures since the 18th century has been paralleled by the
history of structural analysis, and this was also true in the postwar period. The
tools available for analysis also underwent dramatic changes.

Tools
Up to and for some time after the war, the engineer’s chief calculating tool was
the slide rule which during the 1970s was gradually superseded by increasingly
sophisticated electronic calculators. For calculations requiring greater accuracy
logarithmic tables and mechanical calculators, often hand operated, were used until
the 1960s. Before the advent of computers the solution of simultaneous equations
with more than four unknowns was a long winded process, rapidly approaching
the impractical as the number increased. By the end of the 1950s computer
bureaux could provide rapid solutions to sets of simultaneous equations, although
it may be that many engineers were slow to take advantage of them. By the mid-
1960s some structural programmes were available and design offices had started
to equip themselves with computers. In the late 1970s and early 1980s, pro-
grammes became available specifically for analysing prestressed concrete bridges
as well as programmes which enabled very demanding construction calculations,
such as controlling the alignment in cantilever construction, to be carried out
rapidly. The result was to enable engineers who were using innovative methods
to do so more easily and to encourage others to follow.

Analysis and the new techniques


New structural techniques needed more sophisticated analyses. Prestressed con-
crete is more demanding and less forgiving than reinforced concrete, where high
local stresses can redistribute themselves without causing failure. Types of struc-
tures such as box girders also require more complex analyses to realize their poten-
tial. After the advent of computers the proportion of statically indeterminate
bridges increased.
Perhaps the most important analytical development of the 1950s was that by
the C&CA on methods of load distribution, following on from the work of Guyon
and Massonet on the Continent.89
Another important development was an increasingly better understanding of
how box girders actually worked. The paper on Clifton Bridge38 refers to aero-
nautical engineering literature being used to give an idea of the effects of skew.
The work of Vlasov,90 based on thin shell theory, became available in the early
1960s, introducing the idea of the ‘bimoment’, self-equilibrating moments asso-
ciated with warping and distortion of the cross-section. For British engineers the
papers of Richmond91 from the mid 1960s made the subject more comprehensible,
and computers made the calculations feasible. Also during this period thermal
stresses began to be calculated as a matter of routine and the ability to carry out
more complex calculations was generally followed by demands for them to be car-
ried out.
For bridges which were difficult to analyse, physical models, of perspex and
sometimes of micro-concrete, were tested, many at the laboratories of the C&CA.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:54
12.qxd 15/11/2001 15:57 Page 270

270 Historic concrete

Clifton Bridge was possibly the first, others were Mancunian Way and Gateshead
Highway. Following model tests at the C&CA the Medway Bridge was built with-
out diaphragms in the spans, showing the way for later concrete box girders.

Problems with Concrete bridges have not been immune to the problems experienced by concrete
structures generally due to alkali silicate reaction and the use of calcium chloride.
prestressed Bridges are particularly exposed to the elements and the salting of road bridges
bridges has produced more corrosive environments. Some of the earliest prestressed
bridges still seem to be in good condition; however problems have been experi-
enced with others, particularly corroding tendons where bridges have been salted
and ducts were not completely filled with grout. It was a number of years after
the war before suitable ducts were found and satisfactory techniques for grouting
them. Cables external to the concrete sections were used as early as Clifton Bridge
(completed 1958, detailed design 1953) and in a number of subsequent bridges.
However corrosion of tendons was experienced on some of these and as a result
in the early 1980s the Department of Transport banned the use of external cables
on its bridges. In 1985, a segmental bridge with grouted tendons failed, due
to corrosion at the joints between segments,92 and corrosion has been found in
some other bridges with grouted cables. In 1992 the Department placed a ban on
grouted cables in new bridges being constructed for it, and for some years only
external cables were acceptable. A Concrete Society working party was set up
to find means of producing grouted cables with acceptable reliability93–95 and after
some four years the ban was lifted.

The changing The changes in bridges since 1940 were not driven by style, but by the search for
more effective and economical methods of construction. The changes in the char-
character of acteristic appearance of bridges may have a cultural component, but it is hard to
bridges isolate it, particularly as the architecture of the early postwar period emphasized
function as the basis for style. Some of the larger bridges were designed with advice
from architects and may have had an influence on others. Bridges of this period
are simpler in appearance than their predecessors, larger in scale, lacking in detail.
This is just as true of steel bridges as it is of concrete ones. Large modern concrete
and steel bridges are often surprisingly similar in appearance, particularly where
the concrete has a smooth surface.
Typical bridges of the 1950s may be simpler in appearance than those of the
1930s, but they still do not have the distinctively modern character which started
to appear in the 1960s. A typical 1950s beam bridge often has a number of beams
very close together without any cantilever over the edge other than a slightly pro-
jecting fascia. It also has cross beams or diaphragms to stiffen the cross section
and distribute loads. These are not present in modern bridges which have had
more sophisticated analyses and a different economic climate for construction. Even
box girders had frequent internal diaphragms, and Medway Bridge must have
been one of the first to do without them. A typical modern road bridge has few
main members, possibly only one or two box girders, often with edge cantilevers
which are often quite large, and diaphragms only at the supports. Bridges are still
evolving, as always.

Postscript In February/March 1996 English Heritage put forward, for public consultation,
65 postwar buildings recommended for listing. Eleven of these were bridges (nine
of concrete) and the only ‘buildings’ recommended Grade I were the Severn Bridge
and the Kingsgate Footbridge. In 1998 these two were listed as Grade I, Winthorpe

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:54
12.qxd 15/11/2001 15:57 Page 271

UK concrete bridges since 1940 271

Bridge as II* and the following as Grade II:


• English and Welsh sections of the Wye bridge;
• Rhinefield Bridge, near Brockenhurst;
• Footbridge over the River Cherwell at Parson’s Pleasure Punt Rollers, Oxford;
• Garret Hostel Bridge, Cambridge;
• Wentbridge Viaduct, near Selby;
• Swanscombe Footbridge, A2 near Ashford;
• West Footbridge, London Zoo.
All but the Severn and Wye Bridges are concrete. The Adam Viaduct was listed
Grade II in 2001.

Acknowledgements The starting point for this chapter was a study of postwar bridges for English
Heritage. It could not have been written without the help of Michael Chrimes and
the libraries of the Institution of Civil Engineers, the British Cement Association
and Ove Arup and Partners; and would have been even more imperfect than
it is without discussions with Dr George Somerville, Sir Alan Harris, Dr E.W.H.
Gifford, Alfred Goldstein and Dr Francis Walley, and help from too many others
to mention by name. The help of John Henry was invaluable in finding out how
to get to bridges considered for listing.

References 1. Sriskandan, K., Prestressed concrete road bridges in Great Britain: a historical survey.
Proc. ICE, 1989, 86, 269–302.
2. Henry, D., Jerome, J.A., Modern British Bridges. CR Books: London, 1965.
3. Mason, A.P., Alveley Bridge — cantilevered centering. Concr. Constr. Eng., 1937, 32,
453–59.
4. Paul, A.A., The use of pre-stressed concrete beams in bridge deck construction. J. ICE,
1943, 21, 19–30.
5. Anon., An exhibition of prestressed concrete. Concr. Constr. Eng., 1949, 44, 129.
6. Anon., Reconstruction of Nunn’s Bridge, Lincolnshire. Concr. Constr. Eng., 1948, 43,
239–44.
7. Walley, F., The progress of prestressed concrete in the United Kingdom. Lecture given
at the Institution of Civil Engineers, 1962.
8. Thomas, F.G., Pre-stressed concrete. Proceedings of Conference on Prestressed Con-
crete held at Inst. Civ. Engrs, 1949.
9. Gifford, E.W.H., Recent developments in highway bridge design in Hampshire. Proc.
ICE, Part 2, 1952, 1, 461–85.
10. Anon., A prestressed bridge in North Devon. Concr. Constr. Eng., 1954, 49, 169–70.
11. Criswell, H., Prestressed concrete bridges in North Devon. J. Inst. Highw. E., 1956,
3, 60–66.
12. Goldstein, A., Design and construction of a prestressed concrete arch footbridge at
Oxford. Concr. Constr. Eng., 1950, 45, 347–57.
13. Anon., Concrete structures at the festival of Britain. Concr. Constr. Eng., 1951, 46,
199–206.
14. Anon., Test of a prestressed concrete footbridge. Concr. Constr. Eng., 1952, 47, 185–88.
15. Walley, F., St. James’s Park Footbridge. Proc. ICE, 1959, 12, 217–21.
16. Prestressed Concrete Development Group, Prestressed Concrete Footbridges, Pre-
stressed Concrete Development Group, 1962.
17. Anon., A prestressed pipe bridge. Concr. Constr. Eng., 1955, 50, 229–32.
18. Anon., A Swiss system of prestressing. Concr. Constr. Eng., 1958, 53, 431–39.
19. Anon., Two reinforced concrete bridges at Boroughbridge. Concr. Constr. Eng., 1947,
42, 217–22.
20. Anon., Boonshill Bridge, Rye, Sussex. Concr. Constr. Eng., 1947, 42, 271–72.
21. Buckton, E.J., Cuerel, J., The New Waterloo Bridge. J. ICE, 1943, 20, 145–201.
22. Anon., A pre-stressed concrete railway bridge near Wigan. Concr. Constr. Eng., 1947,
42, 305.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:54
12.qxd 15/11/2001 15:57 Page 272

272 Historic concrete

23. Dean, A., Pre-stressed concrete applied to the construction of railway bridges and other
works, Proc. ICE, Rly. Div., 1951, 44, 14–33.
24. Anon., A railway viaduct in pre-stressed concrete. Concr. Constr. Eng., 1952, 47, 385–87.
25. Berridge, P.S.A., Prestressed concrete slabs for railway bridges. Concr. Constr. Eng.,
1954, 49, 283–88.
26. Anon., A prestressed bridge at Manchester. Concr. Constr. Eng., 1958, 53, 395–98.
27. Note in Concr. Constr. Eng., 1949, 44, 37.
28. Anon., Prestressed concrete bridges in Yorkshire. Concr. Constr. Eng., 1950, 45, 99–100.
29. Anon., A new type of prestressed bridge deck. Concr. Constr. Eng., 1958, 53, 145–46.
30. Sadler, R.E., Development in overhead electrification of railways as it affects the civil
engineer. Proc. ICE, 1959, 12, 125–51.
31. Abeles, P.W., Partial prestressing and its suitability to limit state design. Struct. Engr,
1971, 49, 67–86. Discussion 529–41.
32. Anon., Prestressed concrete railway bridge at Rotherham. Concr. Constr. Eng., 1953,
48, 19–21.
33. Anon., Test of a prestressed concrete railway bridge. Concr. Constr. Eng., 1951, 46, 186–87.
34. Turton, F., Three prestressed concrete railway bridges. Proc. ICE, 1961, 20, September,
1–18, discussion 1962, 22, July, 317–330.
35. Mann, F.A.W., Developments in the construction of concrete railway bridges. Concrete,
1968, 2, 373–79, 426–29.
36. Mann, F.A.W., Railway Bridge Construction: Some Recent Developments. Hutchinson
Educational, 1972.
37. Blackler, M.J., Cooke, R.S., Besses O’ Th’ Barn Bridge: inspection and testing of a seg-
mental post-tensioned railway bridge. Proc. ICE Structs Bldgs, 1995, 110, 19–27.
38. Best, K.H. et al., Incremental launching at Shepherd’s House Bridge. Proc. ICE, Part
1, 1978, 64, 83–102.
39. Kretsis, K., Lyne railway underbridge. Proc. ICE, Part 1, 1982, 72, 585–610.
40. Wooldridge, F.L. et al., The New Northam Bridge, Southampton. Proc. ICE, Part 1,
1955, 4, 269–89.
41. Finch, R.M., Goldstein, A., Clifton Bridge, Nottingham: Initial Design Studies and
Model Test. Proc. ICE, Part 1, 1959, 289–316.
42. Finch, R.M., Goldstein, A., Clifton Bridge, Design and Construction. Proc. ICE, Part
1, 1959, 4, 317–52
43. Henry, D., Jerome, J.A., Modern British Bridges. CR Books: London, 1965.
44. Kerensky, O.A. et al., Medway Bridge: design. Proc. ICE, 1964, 29, 19–52.
45. Kier, M. et al., Medway Bridge: construction. Proc. ICE, 1964, 29, 53–100.
46. Lewis, C.D. et al., Orwell Bridge — design. Proc. ICE, Part 1, 1983, 74, 765–78.
47. Henry, D., Jerome, J.A., Modern British Bridges. CR Books: London, 1965, 144–45.
48. Anon., Bridges on the Newark Bypass Road. Concr. Constr. Eng., 1964, 59, 351–55.
49. Coombs, A.S., Hinch L.W., The Heads of the Valleys Road. Proc. ICE, 1969, 44, 89–118.
50. Hansen, F., A contractor’s view of design and its influence on construction. Proc. Meet.
Design Concr. Bridge Struct., Concrete Society: London, 1967.
51. Sims, F.A., The design of the Wentbridge Viaduct. Struct. Concr., 1963, 1, 573–80.
52. Markham, R.B., The Construction of the Wentbridge Viaduct. Struct. Concr., 1963,
1, 553–65.
53. Anon., Prize-winning designs for a motorway bridge. Concr. Constr. Eng., 1964, 59, 113–18.
54. Gifford, E.W.H. et al., The design and construction of the Calder Bridge on the M1
motorway. Proc. ICE, 1969, 43, 527–52. Discussion 1970, 46, 355–71.
55. Henry, D., Jerome, J.A., Modern British Bridges. CR Books: London, 1965: 97.
56. Henry, D., Jerome, J.A., Modern British Bridges. CR Books: London, 1965: 176–77.
57. Gibb, M.E., Tansley, F.J., Queenhill Bridge over the River Severn. Proc. ICE, 1962,
23, 545–63.
58. Henry, D., Jerome, J.A., Modern British Bridges. CR Books: London, 1965, 160–61.
59. Brown, P.A., Prestressed concrete bridges 1970–74. Concrete, 1974, 8 (April) 45–49.
60. Henry, D., Jerome, J.A., Modern British Bridges. CR Books: London, 1965, 108–109.
61. Rawlinson, J., Stott, P.F., The Hammersmith Flyover. Proc. ICE, 1962, 23, 565–600.
62. Bingham, T.G., Lee, D.J., The Mancunian Way elevated road structure. Proc. ICE,
1969, 42, 459–92.
63. Baxter, J.W. et al., Design of Western Avenue Extension (Westway). Proc. ICE, 1972,
51, 177–218. Also paper by Nundy on construction 219–50.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:54
12.qxd 15/11/2001 15:57 Page 273

UK concrete bridges since 1940 273

64. Brown, C.D., London Bridge: planning, design and supervision. Proc. ICE, 1973, 54,
25–46.
65. Mead, P.F., London Bridge: demolition and construction. Proc. ICE, 1973, 54, 47–69.
66. Sims, F.A., Applications of resins in bridge and structural engineering. Int. J. Cement
Composites Lightweight Concrete, November 1985.
67. Smyth, W.J.R. et al., Tyne and Wear Metro: Byker Viaduct. Proc. Inst. Civ. Engrs, Part
1, 1980, 68, 689–700.
68. Sims, F.A., FIP 78, UK National Report on Bridges. Concrete, 1978, May, 16–25.
69. Sims, F.A., Applications of resins in bridge and structural engineering. Int. J. Cement
Composites Lightweight Concrete, 1985.
70. Rowley, F.N., Incremental launch bridges: UK practice and some foreign comparisons.
Struct. Engr, 1993, 71.
71. Henry, D., Jerome, J.A., Modern British Bridges. CR Books: London, 1965, 86, 134
and others.
72. Anon., An elegant footbridge at Durham with an original method of construction. Con-
crete Quarterly, 1964, 60, January–March.
73. Anon., The Preston by-pass, road and bridge works. Roads & Road Constr., 1957, 35,
200–207.
74. Thomson, W.R., Motorway design and construction in Worcestershire. Roads & Road
Constr., 1959, 57, 45–50.
75. Anon., The Great N. Road in the W. Riding of Yorkshire. Roads Road Constr., 1959,
57, 55–56.
76. Anon., The Maidstone bypass road: bridges for the eastern section. Concr. Constr. Eng.,
1958, 53, 195–200.
77. Williams, O.T., The consulting engineer’s view. Some particular considerations. Proc.
ICE Conf. Motorways Br. Today Tomor., London, 1971: 48.
78. Wilson, C.B., Assessment of the reinforced concrete hinges on five M6 overbridges in
Staffordshire Proc. ICE Structs. Bldgs., 1995, 110, 4–10.
79. Anon., The London-Birmingham Motorway. Concr. Constr. Eng., 1959, 54, 337–44,
413–14.
80. Anon., Bridges on the Birmingham-Preston Motorway (Cheshire Section). Concr. Con-
str. Eng., 1962, 57, 31–32.
81. Anon., Concrete Bridges on the New Motorways. Concr. Constr. Eng., 1964, 59,
389–400, 437–43.
82. Sriskandan, K., Prestressed concrete road bridges in Great Britain: a historical, sur-
vey. Proc. ICE, 1989, 86, 274–77.
83. Taylor, H.P.J., The precast concrete bridge beam — the last 50 years. Struct. Engr, 1998,
76, 407–14.
84. Anon., Concrete piers for the Tamar Road Bridge. Concr. Constr. Eng., 1961, 56, 130–34.
85. Sims, F.A., UK national reports to FIP78 — Bridges. Concrete, 1978, 12, May, 25.
86. Anon., Rubber bearings for bridges. Concr. Constr. Eng., 1959, 54, 350.
87. Lee, D.J., The theory and practice of bearings and expansion joints for bridges, 1971.
Cement and Concrete Association, London.
88. Sims, F.A., Bridle, R.J., The design of concrete hinges. Concr. Constr. Eng., 1964, 59,
277–286.
89. Rowe, R.E., Concrete Bridge Design. CR Books: London, 1962.
90. Vlasov, V.Z., Thin walled elastic beams, Israel Program for Scientific Translations,
Jerusalem, 1961.
91. Richmond, B., Twisting of thin-walled box girders. Proc. ICE, 1966, 33, 659–75.
92. Woodward, R.J., Williams, F.W., Collapse of Ynys-y-Gwas Bridge, West Glamorgan.
Proc. ICE, Part 1, 1988, 84, 635–69.
93. Woolley, M.V., Clark, G.M., Post-tensioned concrete bridges. Struct. Engr, 1993, 71, 409–11.
94. Porter, M.G., Repair of Post-tensioned Concrete Structures. Concrete Society, Concrete
Bridges — Investigation, Maintenance and Repair, London, 1985.
95. Raiss, M., Lasting Effect, Concrete Engineering, supplement to New Civil Engineer,
1995, 46– 48.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:26:54
13.qxd 15/11/2001 15:57 Page 275

13 Reinforced and prestressed concrete


in maritime structures

Brian Sharp

Introduction In this chapter, the development of reinforced concrete and the later innovation
of prestressed concrete in maritime structures is reviewed, from its first such intro-
duction in France in 1896 until the era of North Sea Oil Platforms in the 1970s.
Most of the applications are for quay structures but related works include coastal
structures, offshore structures and lighthouses. The review is divided into broad
periods. Starting with the early applications of reinforced concrete from the turn
of the century to 1920, it traces the extensive developments of the 1920s and 1930s,
then the Second World War followed by the activity of post-war construction, which
saw the increasing size of maritime facilities. Although focused on the UK, the
review includes related overseas activity.
The aim of the chapter is threefold:
• To understand and appreciate what was achieved in these past periods and
what was done, how and why.
• Thereby to enable present-day engineers to assess, repair and renovate where
appropriate, by providing an insight into what system and details may lie
beneath the surface. For this reason, more details are given of earlier struc-
tures as it is hoped that later ones should be better recorded and be more
recognizable in structural form and detailing.
• To learn from the past and interpret the reasons for success or failure in the
performance, especially relating to structural form, detailing and durability,
to improve the design of future structures.
Engineering related to the sea is one of the oldest branches of civil engineer-
ing, the sea being a major highway for exploration, demographic expansion, fish-
ing and trade. Loading and exposure conditions for works in the sea are usually
exceptionally severe, due to the sheer mass of ships and the loads they carry, the
natural forces of winds and waves and the corrosive properties of seawater.
To combat these conditions, plain Portland cement concrete and then reinforced
concrete were readily employed as soon as the materials and systems became avail-
able. Consequently, the seawater environment has served as a proving ground and
accelerated test-bed for methods and materials.
By virtue of its strength in bulk and versatility in both in situ and precast form,
concrete has obvious advantages in maritime structures. Portland cement was mis-
trusted for use in seawater, being reported as first used in blocks for the original
Admiralty Pier at Dover in 1849.1 Previously, mass (unreinforced) concrete was
made with lime or pozzolan as the binder. In France, prior to 1914, the use of
Portland cement in port works was negligible.2
The medium of reinforced concrete by definition excludes plain unreinforced
concrete, termed ‘mass’ concrete in UK but not in the USA. However, as the intro-
duction of steel reinforcement on structural grounds causes its own problems for
durability, and as unreinforced concrete predates reinforced concrete and can still

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:27:25
13.qxd 15/11/2001 15:57 Page 276

276 Historic concrete

remain the optimum solution when structurally and economically feasible, its his-
toric and modern use is summarized as a preview. Gravity structures in the sea
using masonry blockwork and rubble concrete infills have been used from early
times and precast concrete blocks were and remain one of the most convenient
means of gravity wall construction, such as used at Folkestone between 1897 and
1905,1 as shown in Figure 13.1. Blocks of 1: 6 concrete, average mass 16 t, were
laid in bonded courses.
A modern example is the ‘Tema’ type wall3,4 as shown in Figure 13.2. This inno-
vatory form was dubbed the ‘upside-down’ wall. The heavy lines on the figure
illustrate how the classical trapezium shaped wall with the widest section at the
base is inverted to provide the widest section at the top. This inversion, initiated
in fact by Alex Leggatt when at Halcrow, provides a most efficient restoring mass,
the capping block providing an enhanced lever arm to the toe, and maximum mass
in air with minimum deduction for ‘submerged weight’.
As mentioned above, both plain concrete and even more so, and for good rea-
son, reinforced concrete was mistrusted by maritime engineers at the turn of the
century, and some of the questions related to its use, as raised in the reports
reviewed below, sound remarkably familiar. A comparable spur to the replacement
of timber in buildings by reinforced concrete because of its fire resistance was the
scourge of sea worms in timber-piled sea structures.
Maritime structures figure prominently amongst the earliest applications of
reinforced concrete from the end of the 19th century. The early days of building
construction were dominated by the patented systems, but this was not so for
maritime structures. Although the Hennebique system figures prominently in early
maritime work, both this and several of the examples in the ICE Reports of 19135
all employed ordinary straight rods.

Figure 13.1 Folkestome pier,


1897–1905.

Figure 13.2 ‘Tema’-type


blockwork wall, 1952–1965.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:27:25
13.qxd 15/11/2001 15:57 Page 277

Reinforced and prestressed concrete in maritime structures 277

Francois Hennebique is the name most associated with the development of


reinforced concrete in port works, beginning with his 8.25 m cantilever quay at
Nantes6,7 in 1896. Mouchel7 introduced the Hennebique Ferro-Concrete system
into Great Britain in 1897. As explained in Chapter 4 and illustrated in its Figure
4.7 and 4.34, the Hennebique system used normal round bars and resisted diag-
onal tension due to shear by inclined bars and open-topped stirrups made from
flat strip. The stirrups embraced single bars. Patent links were used in columns
and piles (Figure 4.2 of Chapter 4). Mouchel patented a hollow circular pile, and
also the construction of massive cylindrical piers by driving one or more square-
section piles in a cluster, which were then enclosed with a reinforced concrete cylin-
der, which was filled with concrete and steel (Figure 13.3).
The main problems for earlier port engineers were the stability of the cement
compounds in seawater and the corrosion of reinforcement. Wentworth-Shields8,9
explained that mass (Portland cement) concrete had been freely and successfully
used in railway and dock construction in the last 50 years of the 19th century.
When failures occurred, they were due to ‘unsound’ cement, that is cement
containing uncombined lime. Concrete made with such cement swelled and dis-
integrated in seawater. Port engineers were constantly on the watch for unsound

Figure 13.3 Mouchel piled


cylinder piers (from Mouchel-
Hennebique7).

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:27:25
13.qxd 15/11/2001 15:57 Page 278

278 Historic concrete

cement and required cement to be turned over by shovel at weekly intervals. Pats
of neat cement about 75 mm diameter were immersed in cold water; if the cement
contained an excess of free lime the edges would crack. Cement for Folkestone
pier1 was checked by plunging broken tensile briquettes into boiling water for 3 h,
to detect signs of swelling. Disasters had been experienced at Belfast, Aberdeen
and in France in which the concrete had been reduced to gravel and sand.
Fortunately, improved methods of cement manufacture removed these risks quite
early on.
An uninformed observer may, these days, gain the impression that corrosion of
reinforcement is new. It was observed as a major problem from the earliest
stages,2,5,6 but, mostly because of the time it can take to show in better-made struc-
tures in cold and temperate climates, was never really adequately understood.
The mechanisms of the corrosion of reinforcement in concrete are only now begin-
ning to be fully analysed, and it is not surprising that electrolysis from the effects
of stray currents from electric cranes and ‘moist’ air were strongly suspected.
One of the major problems of analysing the performance of maritime structures
is that success is as much a matter of detailing as it is of materials and is highly
sensitive to the particular ‘microclimate’, that is the location of specific members
in relation to the fluctuating sea level, in addition to the geographical ‘macro-
climate’.10,11 Many of the earlier applications of reinforced concrete in quay con-
struction involved a reinforced concrete deck spanning between timber, iron, steel
or massive piles. Such an application, often being out of the range of seawater
action and washed by rain water, was inherently durable. In other earlier exam-
ples, reinforced concrete was used in the substructure, and surmounted by con-
ventional mass concrete or masonry in the upper tidal and splash zones, again
affording ideal conditions for longevity.
This chapter concerns the details likely to be met in earlier construction as
opposed to performance, but the two are mutually related. Where possible, feed-
back is given from later chapters’ references to the same works.

The Institution of Maritime works figured prominently in the Institution of Civil Engineers (ICE)
Committee convened to review reinforced concrete practice in 1910.6 Port works
Civil Engineers in the UK, New Zealand and Australia were included in the second report of 1913,5
Committees some details of which are given below.
One of the most valuable and longest contributions to the use of concrete
in maritime work is that of the ICE Sea Action Committee, initiated in 1916 and
lasting until 1960. The programme of research into the deterioration of structures
in seawater included the use of timber, metal and concrete and observations in
some 40 ports of the UK and in the Dominions. The first report,12 published
in 1920, had identified the main problems and recognized that, as the deteriora-
tion mechanisms were not rapid, a long-term programme was required. The 15th
report13 in 1935 gave an overview of the work to that time.
The conclusions of the final 20th report on the durability of reinforced concrete
in seawater,14 published in 1960, were that the primary cause of deterioration was
corrosion of reinforcement and not disintegration of the concrete matrix itself.
In hindsight, the conclusions, as were those of programmes carried out in other
countries, were somewhat understated in comparison with the observations of
1920.12

The early years According to Wentworth-Shields,8 the first reinforced concrete maritime structure
in the UK was built at a Southampton shipyard on the River Itchin in 1899, in
(1900–20) imitation of a timber jetty. It consisted of a reinforced concrete deck 30.5 m by
12 m on reinforced concrete piles, designed and supervised by Hennebique, and

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:27:25
13.qxd 15/11/2001 15:57 Page 279

Reinforced and prestressed concrete in maritime structures 279

was reported8 to be in excellent condition in 1956. This was the Woolston jetty,7,15
(Figure 13.4), which was later incorporated into an extensive jetty complex for
the Ministry of Defence. A 1-t piece of this structure is kept in the Science Museum
Annexe at Wroughton, Wiltshire.
A coaling jetty was built at Southampton, from 1901 to 1903, for transhipping
coal from South Wales colliers to barges, and carried six large electric cranes
(Figure 13.5).7 The reinforced concrete beam and slab deck was carried on tres-
tles of reinforced concrete piles which were stiffened with longitudinal, transverse

Figure 13.4 Woolston Jetty


at Southampton, 1899 (from
Mouchel-Hennebique7).

Figure 13.5 Coal barge jetty,


Southampton, 1903 (from
Mouchel-Hennebique7).

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:27:25
13.qxd 15/11/2001 15:57 Page 280

280 Historic concrete

and diagonal bracing. The details, drawn from the ICE report5 of 1913, are given
in Table 13.1. (Principal dimensions will usually be given in metric units, but details
in the tables are given in the original units, except where otherwise stated.) Within
a few years this jetty was suffering considerably from reinforcement corrosion,
above the high water level.5 The flat stirrups had contributed to the trouble, and
cover had been reduced by bars sagging and stirrups dropping in the forms. The
reinforced concrete Town Quay at Southampton7,8 also dates from this time.
Widening of 427 m of quay by 13.7 m at Southampton docks between 1902 and
1905 is recorded in the 1913 report.5 This was also a beam and slab deck on braced
piles. The concrete was 1 : 3 and 1 : 4, with a maximum aggregate size of 22 mm
(7⁄8 in) ring with minimum cover of 38 mm (11⁄2 in). The flat stirrups were made
of steel. Extensive damage had occurred by the 1913 report, very similar to the
case above. Unlike the first reported jetty, the concrete was mixed ‘wet’. Until
the universal acceptance of the role of water-to-cement ratio, argument centred
on whether the concrete mix should be ‘wet’ or ‘dry’.
Details of the various stages of extension of the jetty up to 1915 were described
in a paper16 on repairs carried out by the Gunite process in 1936–40. A number
of practical expedients had been used, including building onto earlier cast iron
piles. Some decking incorporated Kahn bar reinforcement. The Associated
Portland Cement Manufacturers’ Swanscombe Jetty12 on the Thames, built in
1906, was also detailed with Kahn bars, and had straight bars in the beams and
‘Johnsons Lattice’ in the decks (Figure 13.6; see also Figure 4.1 and Table 4.1 of
Chapter 4). The heavy piers were built up of large diameter circular concrete blocks
with holes through which rails were threaded and grouted up.
The jetty head extension at Gladstone5,12 in Australia had a deck slab supported
by continuous longitudinal beams on braced piles and was built between 1907 and
1908 (Table 13.1). A similar braced jetty, the Clyde Wharf 5,12 built at Wellington,
New Zealand in 1908–1909 was also, in 1913, described as being in good condi-
tion (Figure 13.7 Table 13.1). The pile extensions and bracings above low water
level were all cast in situ.
The early development of reinforced concrete piles was reported by Walmisley15
in 1906–1907. Piled construction has obvious benefits for working over water
and the precast production of piles has many advantages over in-situ work. The
idea was suggested by Coignet in 1869. The various types of pile are described in
Chapter 7. The reinforcement consisted of longitudinal bars tied together by trans-
verse bars, spiral winding or expanded metal. Rolled joists and sections connected

Figure 13.6 Swanscombe


Jetty, 1906 (from Institution of
Civil Engineers12).

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:27:25
Table 13.1 Project details: 1903–1909 (imperial units unless otherwise stated)

Details Coal barge jetty, Southampton, 19035 Jetty head at Gladstone, Australia, 19085,12 Clyde Wharf, New Zealand, 19095,12

Overall dimensions 110 m long  6 m wide. Deck 10.5 m 61 m long Deck 3.7 m above ordinary low water springs
above seabed Depth alongside at low water 7.3 m Bottom edge of walings 0.6 m above OLWS
Tidal rise 4 m Tidal rise 3.7 m (Ordinary low water springs)
Deck supported on beams supported on Tidal rise 1.2 m
three-pile braced trestles
Basis of calculation Not available in 1913 Reinforcement designed by contractor Mode of calculating stresses as laid down in Marsh
working stresses Designed to carry a test load of and Dunn Reinforced concrete, London, 1906
13.qxd 15/11/2001 15:57 Page 281

7 cwt/ft2 (3.8 t/m2) Limiting (working stress) stresses column concrete:


400 psi (2.8 N/mm2) compression; beams, concrete
in compression: 500 psi (3.4 N/mm2);
steel in tension: 15,000 psi (103 N/mm2)
Piles and struts Piles: 16 in  12 in. 4 No. 15⁄8 in bars 15 in  15 in. Each tier of six piles braced Piles 18 in square, spaced 20 ft c/c in longitudinal
Struts: 12 in  12 in. 4 No. 1 in bars diagonally (original piles were hollow, 18 in direction and 9 ft transverse. Piles carrying decking
square with gravel aggregate but failed in only reinforced with 4 No. 1 in ms rods at 12 in c/c.
vessel impact). Braces and wales Piles carrying crane rails had 8 No. 11⁄2 in ms rods
14 in  19 in mortized into piles spaced around the sides of a 12 in square
Beams Under railway rails: 10 in wide  12 in deep. Longitudinal beams: 24 in  12 in. Main beams (longitudinal): spans reduced by means of
4 No. bottom bars and 2 No. top, all 13⁄16 in 10 ft apart, 20 ft span struts to the piles. Beams 131⁄2 in wide by 101⁄2 in deep
Under crane rails: 12 in wide  15 in deep. below deck slab. Outer crane beam had 3 No. 11⁄8 in rods
6 No. bottom bars and 3 No. top, all 13⁄16 in top and bottom. Other beams 3 No. 1 in top and bottom
Transverse beams: 9 in  12 in. Secondary beams (transverse): 9 in wide by 71⁄2 in below
6–8 ft apart, 10 ft span deck slab. 2 No. 11⁄8 in rods in bottom face at 13 in
below upper surface. 2 No. in top, 21⁄4 in below surface
Wales: reinforced with 4 No. 7⁄8 in rods
Stirrups to piles Round steel 9⁄16 in loops at 10 in c/c
and struts

IP: 130.194.20.173
Stirrups to beams Flat wrought iron stirrups 21⁄2 in  12 BWG Round rod shear bars in the forms of loops
(Birmingham wire gauge) on the lower Main beams: 3⁄8 in at 4 in c/c

On: Tue, 17 May 2011 06:27:25


bars. Inverted stirrups 13⁄8 in  15 BWG on Secondary: 5⁄16 in at 31⁄2 in c/c

Delivered by ICEVirtualLibrary.com to:


the upper bars
Slabs 6 in thick. Maximum span 5 ft 6 in slab 71⁄2 in thick. Panels 9 ft long by 3 ft–4 in wide between main
and secondary beams. Covered with 41⁄2 in wood
blocking. 3⁄4 in rods at 31⁄2 in c/c at 61⁄4 in depth from upper
surface, bent up over supports. Rods of adjacent spans
overlap so that steel area is double over the supports
Concrete 1 : 4 cement: natural river gravel with sand. Piles: 3 cement : 4 sand : 8 blue 1 : 2 : 3. 3⁄4 in maximum aggregate, mixed ‘wet’,
Crushed to 3⁄4 in ring metal chips almost sloppy
Rest: 3 cement : 5 sand : 8 Cement locally made, to British Standard of 1904. Average
gravel, mixed wet tensile test briquettes with neat cement, 701 lb at 28 days
Reinforcement Steel ratio: Plain round bars. Siemens Martin steel. UTS (ultimate tensile strength) 60,000 psi
Railway rail beams 13⁄4% UTS (ultimate tensile strength) (414 N/mm2), elastic limit between 50% and 60%
Crane beams 21⁄2% 28–30 t/in2 (432–463 N/mm2) slung by wires from battens)
Reinforced and prestressed concrete in maritime structures

Struts 21⁄4%
Piles 41⁄4%
Cover 1 in
281
13.qxd 15/11/2001 15:57 Page 282

282 Historic concrete

Figure 13.7 Clyde Wharf,


Wellington, New Zealand,
1909 (from Institution of
Civil Engineers5).

Figure 13.8 (a) Hennebique


piles (from Walmisley16).
(b) Mouchel Piles7.

together by flat bars were also employed. Considère employed hooped concrete
in which a thin rod was wound spirally around vertical rods. A concrete stress
of 13.8 N/mm2 could be taken in hooped concrete, as opposed to 2.8 N/mm2 for
concrete in unrestrained compression.15

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:27:25
13.qxd 15/11/2001 15:57 Page 283

Reinforced and prestressed concrete in maritime structures 283

Most piles were made according to individual patents. Coignet made hollow
piles without any steel. Hennebique piles (Figure 13.8a) were used on jetties at
Woolston near Southampton;15 Mouchel’s type of Hennebique pile had a hollow
cavity (Figure 13.8b)
Other applications of braced piled jetties of this type include Purfleet Pier,12
a coal handling structure on the Thames designed by C.S. Meik employing the
Hennebique system, built in 1904 and extended in 1911. Due to the flat stirrups
and detailing as in the structures at Southampton, cracks were soon observed in
the high water regions. Through annual inspections and regular repairs the struc-
ture remained in service, until requiring major repairs in 1950. It was converted
for use by oil tankers in 1961 and, after later repairs in 1979, was in use until
recently. The bowstring Girder Bridge illustrated in Figure 7.3 of Chapter 7 still
appears to be in place.
Extensive works for Parkeston Quay at the continental station of the Great
Eastern Railway Co., Harwich, figured in reports of 1908,17 1933,18 193419 and
1935.20 A common feature of this type of quay was that large areas of water were
decked over on piles. The new braced piled quay at Harwich,17 which used the
Hennebique system, was 329 m by 16 m (Figure 13.9). The dredged depth was
6 m at low tide and 9.8 m at high tide. The piers supporting the main beams along

Figure 13.9 Parkeston Quay


extension, Harwich, 1908
(from Twelvetrees17).

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:27:25
13.qxd 15/11/2001 15:57 Page 284

284 Historic concrete

Figure 13.10 Coventry


Ordnance Dock on the Clyde,
1908 (from Twelvetrees22).

the front were 1.40 m diameter cylinders, containing pairs of foundation piles dri-
ven into the clay beneath. All the independent piles were 355 mm square. The
deck slab was 150 mm thick. Repairs by the Gunite process were carried out in
the 1930s, 1963 and 1977.21
A very different use of reinforcement is illustrated by the dock for the Coventry
Ordnance Works on the Clyde at Scotstown in 1908.22 The tidal dock wall was
15 m high, to accommodate a tidal range of about 6 m from a lowest level below
half way up the wall. The gravity wall was economic in its use of concrete and was
achieved by a stepped back to the wall, and buttressing by wide counterforts 8 m
apart (Figure 13.10). The reinforcement did not take bending tension, but was
placed in a horizontal direction such ‘as to give ample resistance to tension wher-
ever developed, to guard against temperature cracks, and to tie the counterforts
securely to the walls where connexion of the kind is desirable’. The bars were
patent indented bars of the type which had been used throughout the sea wall at
Galveston in the USA.
A similar jetty to Parkeston was built at Dunston, Newcastle-upon-Tyne, in 1907–
1908, to serve the reinforced concrete flour mills, granary and grain cleaning tower
of the Cooperative Wholesale Society, all built using the Hennebique system, and
described in Chapter 4.68

The 1920s and Extensions of Parkeston Quay were always in the news and ‘one of the most impor-
tant contracts of its kind’ was reported in 1933–35.18–20 The existing quay was
1930s extended by 345 m by 44.5 m, with two curved approach viaducts 150 and 170 m
long. Its construction, undertaken by the Yorkshire Hennebique Contracting Co.,
was similar to that described above. At the front, the piles were driven in pairs
and sleeved with cylinders 1.8 m in diameter. Repairs in the 1930s, 1960s and 1970s
are reported by Dyton.21
The berthing arm at Clacton Pier,23 reported in 1934, was designed to resist
berthing forces by opposing rakers forming trestles of four piles (Figure 13.11).
To reduce the tension loads in the raker piles, the dead weight of the berth was
increased by constructing the deck as a stone-filled box.
By this time the rational advantages of reinforced concrete for resisting berthing
forces as an elastic structure with designed fender systems were being exploited.
The earlier reinforced concrete jetties were braced and strutted together in all
directions as low as the tide would allow, as had been done for timber jetties.8 Later,
bracings were discarded in favour of raking piles and, even later, monolithic
portal bents, which had increased strength for less cost and removed the need to
cast members in the tidal zone. Berthing forces, originally allowed for by experi-
ence, were now subject to analysis of the kinetic energy to be absorbed. Then Chief
Engineer of Christiani and Nielsen, Ove Arup, contributed interesting articles in
1934 and 193524 on this subject.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:27:25
13.qxd 15/11/2001 15:58 Page 285

Reinforced and prestressed concrete in maritime structures 285

Figure 13.11 Berthing arm


at Clacton Pier, 1934.23

Figure 13.12 New fish dock


at Grimsby: coaling appli-
ances for two ships at once,
1930 (from Comrie25).

Nevertheless, extensive works for the new impounded Fish Dock Complex
at Grimsby25 commenced in 1930 and were mostly built to the piled and braced
format (Figure 13.12 and Table 13.2).
The extension to Southend Pier26 in 1929 (Figure 13.13) had two decks,
the lower deck being at mid-tide (range 5 m) with a stiff Vierendeel-type truss
without bracings in order to obviate obstruction. The lower deck was made of
narrow precast slabs 159 mm (61⁄4 in) deep, including a 19 mm (3⁄4 in) granolithic top
finish, supported on beams built in situ. The slabs were 229 mm (9 in) thick at

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:27:25
13.qxd 15/11/2001 15:58 Page 286

286 Historic concrete

Table 13.2 Project details: 1930s (imperial units unless otherwise stated)

Details Fish Dock at Grimsby, Quay extension at Newcastle- War Department jetty at
1930 (South Quay)25 upon-Tyne, 193027 Deptford, 193428

Working Not stated Not stated Concrete in compression — 600 psi


stresses (4.1 N/mm2)
Steel in tension — 16,000 psi
(110 N/mm2)
Piles 14 in square Front piers: Groups of 3 No. 295 No. 16 in square, up to 72 ft
Piles stripped and octagonal piles encased in PC long, in raker bents at 16 ft c/c
capped by pile caps (Portland Cement) cylinders 7 ft 3 in Raked at 1 in 33⁄4
1 ft 8 in  1 ft 8 in  3 ft 0 in diameter cast in 10 ft lengths
General piles: octagonal section
at 12 ft c/c
Struts and Struts: 10 in  12 in Struts: 12 in  12 in precast Diagonal braces at 6 ft 3 in above
columns Longitudinal walings: 12 in  14 in Superstructure columns surmounting LWOST (low water ordinary spring
Columns surmounting piles: piles: 18 in  18 in tides) 18 in  12 in
14 in  14 in
Beams Main beams: 12 in  18 in Crane beams: 20 in wide  33 in deep Main beams: 3 ft deep
Secondary beams: 10 in  14 in Railway beams: 12 in wide by 20 in deep Secondary beams: 19 in
deep  12 in wide at 5 ft c/c
Deck 7 in thick, reinforced with 5⁄8 in straight 9 in thick, overlaid by 6:1 concrete Vaulted deck on main and secondary
bars at 12 in c/c and 5⁄8 in bent up laid to a fall and 2 in granolithic beams: minimum thickness 6 in and
bars at 12 in c/c, span 6 ft 3 in topping 11⁄2 in thick granolithic finish
Overlaid by 2 in granolithic not
included in design strength
Roadway at rear, 8 in thick with
21⁄2 in topping
Cement Below high water level: rapid
hardening
Elsewhere: ordinary
Concrete Piles: 204 lb cement, 33⁄4 ft3 fine, Piles and work below high water:
71⁄2 ft3 coarse (i.e. 1 : 1 . 7 : 3.3) 1 : 11⁄2 : 3
All other: 204 lb cement, 5 ft3 fine, Superstructure: 1 : 2 : 4
10 ft3 coarse (i.e. 1 : 2 . 2 : 4.4) Surface treatment: three coats of
sodium silicate applied to piles
before driving and superstructure
after completion

Figure 13.13 Extension to


Southend Pier, 1929 (from
Moller26).

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:27:25
13.qxd 15/11/2001 15:58 Page 287

Reinforced and prestressed concrete in maritime structures 287

the top and 203 mm (8 in) at the bottom, to facilitate washing away of seaweed. Dyton21
reported that no maintenance was carried out until 1976. The lower deck and piles
were ‘reasonably intact’ but the upper parts of the frames were severely decayed.
The 1930 quay extension at Newcastle upon Tyne27 was designed by Mouchel
with braced piles, sleeved piles at the front and bracing (Figure 13.14 and
Table 13.2). The jetty was designed to accommodate very heavy loading of
5 t/m2, and octagonal section piles were chosen, at 3.7 m centres. The bracing
and struts were precast and the only in-situ concrete was in the junctions of the
precast intertidal members, with a thickened section in order to maintain larger
cover.
The War Department jetty at Deptford28 (Figure 13.15 and Table 13.2) was
designed and completed in 1934, with raked piles connected by a triangular
system of low-water bracing to provide efficient strength against lateral forces and
low-level support to the fender timbers.
Meanwhile, on the continent and elsewhere, more ‘modern’ use of reinforced
concrete was being made, as reported in the Permanent International Association
of Navigation Congresses (PIANC) sessions in London in 1923.2 The general report
by Humphries29 the Chief Engineer to the London County Council, reviewed 14
reports from international sources. There was discussion of the advantages of
slag cement in sea water in Belgium. Examples were given30 of reinforced con-
crete caissons constructed at Rotterdam in 1908 with 1 : 41⁄2 concrete (340 kg/m3).
Reinforced concrete caissons had been used in the Dutch East Indies, including

Figure 13.14 Cross section of the quay extension at Newcastle upon Tyne, 1930.27

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:27:25
13.qxd 15/11/2001 15:58 Page 288

288 Historic concrete

Figure 13.15 War


Department jetty at
Deptford, 1934.28

Figure 13.16 Hamworthy


Wharf, Dorset, mid 1930s
(from Wentworth-Shields and
Gray9).

Belawan, since 1910. A number of types of ‘relieving platform’ design, with rein-
forced concrete decks on reinforced concrete sheet and bearing piles, had been
developed in Amsterdam. A UK example of this type of wall is the Hamworthy
Wharf (Figure 13.16).9 Reinforced concrete caissons were used at Yokahama31 from
before 1920 and also in Spain. Previously, ‘caissons’ had been steel or iron struc-
tures filled with concrete. The cement contents used on the continent were
contrasted with much lower figures associated with a 1 : 2 : 4 mix (referred to as
260 kg/m3) which were ‘usual in the USA’; 260 kg/m3 appears to be a very low
figure for 1 : 2 : 4 concrete.

War time A reinforced concrete ‘screwcrete’ pile had been developed for founding in silt or
sand free from boulders, and was used for a slipway at Southend32 in 1934. The
piles were formed by light metal cylinders attached to the screw shoes, which were

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:27:25
13.qxd 15/11/2001 15:58 Page 289

Reinforced and prestressed concrete in maritime structures 289

Figure 13.17 Reinforcement of Phoenix units (from Wood34).

screwed into the sea bed using a mandrel within the casing. The casing was
maintained watertight and a reinforced concrete column formed within it. This
system was used for the piles at the deepwater wharf for No. 1 Military Port built
in Faslane Bay33 in 1940. The beams and deck over the screwcrete piles were
reinforced concrete.
The floating breakwaters and piers built for the invasion of Europe incorpo-
rated many types of components, including steel ship-like units, blockships,
precast concrete boat-like units, bridge units and concrete caissons. Six different
sizes of concrete ‘Phoenix’ caisson units were produced, varying in depth from
7.6 to 18 m, but mostly of the same length, 62 m.34 Engineers are unlikely to meet
examples in service and the archive interest is therefore mostly as a record of what
was done and how it compares with what we would do now. A survey of two sur-
viving units was carried out in 1980 under the ‘Concrete in the Oceans Project,35
aimed at providing additional knowledge to improve the design, construction and
performance of concrete oil production platforms.
The total of 213 ‘Phoenix’ cellular caissons34 were built with a 1 : 11⁄2 : 3 mix with
19 mm maximum aggregate size. The working compressive stress for con crete was
5.86 N/mm2 (850 psi). In order to save time and materials, only straight bars were
used and no splays were employed at the cell corners. The floors varied from 300
to 380 mm thick, the external walls from 350 to 380 mm thick and internal walls
228 mm. Reinforcement details are shown in Figure 13.17. A leftover Phoenix
caisson was in fact used to close the largest gap at Schelphoek in Holland, after
the storm surge damage of 1953.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:27:25
13.qxd 15/11/2001 15:58 Page 290

290 Historic concrete

Feedback from the 198035 survey was not particularly illuminating as to perfor-
mance. The estimated cube strength from cores was, at 37 N/mm2, not perhaps
as high as might be expected, and the cover was between 10 and 30 mm. During
the installation of the units in 1944, problems were experienced due to lack of
anchorage of the walls to each other, caused by the lack of splays and bent bars
and guniting was used to repair miscast sections.36

Post-war until the This period was, arguably, the golden age for building reinforced concrete mar-
itime structures. Information is available in many ICE papers, but sometimes the
1970s overview gives scant information on the details of the concrete and reinforcement.
Immediately post-war, there was a vast amount of work to be done and no lack
of confidence in methods and materials. The principles of reinforced concrete had
been developed from the 1934 DSIR code into CP114 in 1948 (see Chapter 5),
there was a confident but relatively small library of concrete publications, and
prestressed methods were established. Steel was in short supply and concrete had
obvious advantages in making use of indigenous bulk materials.
The increased size of structures to accommodate the rapidly growing size of
vessels, and the transfer from the outdated enclosed and river docks to the open
sea, led to the need to build more for less and hence to re-examine the techniques
and materials previously used, mostly in other structures, such as precasting
and prestressing. Confidence was also boosted by the developments in concrete
technology, with ‘improvements’ in mix design and quality control. In hindsight,
not all of these developments were beneficial.
The dry dock built at South Shields for Brigham and Cowan,37 between 1953
and 1956, is an example of reinforced concrete replacing mass concrete for grav-
ing docks. Dry docks had previously been mostly built to accommodate passenger
liners, but the increased size of oil tankers led to the need for this dock for the
repair of vessels up to 38,000 t deadweight. The dock walls were built by placing
precast reinforced concrete cantilever buttresses at 1.5–2.3 m centres against a near
vertical clay face. The buttresses were only 300 mm wide (Figure 13.18), and were
subsequently anchored into the floor slab. Between the buttresses, concrete pan-
els 600 mm thick were cast directly against the clay face. The reinforced concrete
was generally 1 : 6.5 by weight, 19 mm aggregate and 28 N/mm2 at 28 days.
Development of the oil industry also led to the deep water jetties for the BP
Oil Terminal at Angle Bay, Milford Haven in 1958–60,38 constructed mainly of
precast units carried on Rendhex steel box piles. A paper in the contractor’s house
magazine39 gives a deeper insight into the mix design than in normal construc-
tion papers. What may not now be obvious to some engineers is that the nominal
volumetric proportions 1 : 1 : 2, 1 : 11⁄2 : 3 and 1 : 2 : 4 covered a range of mixes with
different richness and strength. The common 1 : 2 : 4 mix as a ‘normal’ grade to
CP114 had a specified works crushing strength of 20.7 N/mm2, but higher and
lower grades of this mix could be used by control of the cement content and water
to cement ratio to give stresses 25% above or below the nominal level. The mixes
for Milford Haven were specified as in Table 13.3. The development of the prac-
tical mixes from this specification is given in Table 13.4. The average strength of
the ‘A’ mix was some 51 N/mm2 and the slump only 17 mm.
Erith Jetty,40 built on the tidal Thames between 1955 and 1957, is a typical
example of the use of high quality precast and prestressed concrete. The jetty
(Figure 13.19) accommodated vessels up to 14,000 t displacement and carried three
electric cranes. The construction consisted of rigid portal frames at 7.6 m centres
for the jetty head and 15.2 m on the approach, with no other bracing. The portal
columns were cylinders consisting of prestressed shells, themselves capable
of carrying the imposed loads without assistance from the colloidal concrete
plug core.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:27:25
13.qxd 15/11/2001 15:58 Page 291

Reinforced and prestressed concrete in maritime structures 291

Figure 13.18 Dock walls for


Brigham and Cowan, 1956
(from Stott and Ramage37).

Table 13.3 Mixes specified for Angle Bay, Milford Haven

Quality Nominal Aggregate Maximum Works cf. CP 114


proportions size water-to- cube range
by volume (mm) cement ratio (N/mm2) (N/mm2)

A 1 : 11⁄2 : 3 20 0.45 38 32.3


25.9
20.7
B 1:2:4 20 0.50 31 25.9
20.7
16.5
B-1 1:2:4 20 0.55 24 25.9
20.7
16.5

The cylinder piers were made up of units 1.5 m long by 1.8 m outside diameter
and 159 mm thick walls. The units had 12 mm spiral reinforcement outside the
prestressing tendons, which consisted of sixteen 28 mm Macalloy bars, post-
tensioned to give a uniform stress of 7.24 N/mm2 after allowing for losses due
to shrinkage and creep. The minimum cover of 38 mm was monitored by frequent

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:27:25
13.qxd 15/11/2001 15:58 Page 292

292 Historic concrete

Table 13.4 Development of practical mixes* for Angle Bay, from specifications in Table 13.3

Nominal mix Trial mixes in laboratory Final works mix


by volume
Cement Sand Stone Water-to- Proportion Cement Water-to-
(0–5 mm) cement ratio by weight cement ratio
(5–10 mm) (10–19 mm)

A 1 : 11⁄2 : 3 391 590 353 901 0.4 1 : 1.5 : 3.2 365 0.46
B 1:2:4 302 630 368 747 0.5 1 : 2.1 : 4.4 302 ?
B-1 1 : 2 : 4 286 640 445 913 0.55 1 : 2.3 : 4.5 286 ?

*Mix proportions in kg/m3.

Figure 13.19 Erith Jetty,


1957: cross section on jetty
head (from Carey and
Cumming40).

checks with a covermeter, then a cumbersome article requiring a separate car


battery.
To maintain the weight of individual units to within 8 t, only the soffits of the
heavier beams were precast, the webs and, of course, the deck being completed
in situ. The beams were also post-tensioned with Macalloy bars (Figure 13.20). The
cover to ordinary reinforcement was 25 mm and that to the ‘Kopex’ tubing used
to form the prestressing bar ducts was 38 mm. The paper40 includes sample calcu-
lations with prestressing diagrams. The concrete mix and cube record details are
reproduced in Table 13.5.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:27:25
13.qxd 15/11/2001 15:58 Page 293

Reinforced and prestressed concrete in maritime structures 293

Figure 13.20 Erith Jetty,


1957: half-section of soffit
type 2 (from Carey and
Cumming40).

Table 13.5 Erith Jetty 1957: concrete records

Class and location Type of cement Mix proportions (kg/m3) 28-day strength (N/mm2)

Cement Fine Coarse Water-to- Specified Maximum Minimum Average


(19 mm) cement ratio (minimum)

PS
Precast cylinders Rapid 411 550 1302 0.36 51.7 64.3 47.6 54.7
hardening
Precast beams Rapid 481 498 1232 0.36 51.7 60.9 51.9 54.5
and soffits hardening
AA
Cylinder caps, Ordinary 427 525 1230 0.42 25.9 54.1 31.6 45.6
in-situ beams Portland
and deck

Notes: 1. Mix proportions in kg/m3 were re-calculated from original figures given in lb/112 lb cement.
2. Note that the second and third mixes have almost a ‘standard’ water content of 170–180 kg/m3 for such a mix. The first mix, for the same water-to-
cement ratio as the second, has a low water content of 147 kg/m3.

Tees Dock No. 1 Quay,41–43 built between 1959 and 1962, is similar in that the
substructure consisted of prestressed concrete cylinders in bents of three at 9.1 m
centres, carrying a prestressed beam and slab deck. The cylinders were precast
hollow sections 1.8 m long, 1.9 m in diameter, with a wall thickness of 457 mm. The
front cylinders were stressed together with eighteen 22 mm Macalloy bars and
the other cylinders with six bars. The deck consisted of twin main transverse
prestressed I beams spanning between the cylinders and supporting prestressed
deck units. The deck units were novel, being 432 mm thick prestressed concrete
dovetail interlocking units, transversely post-tensioned together with CCL cables.
In later years, by 1982, there were some random failures in the transverse post-
tensioning tendons associated with some incompletely grouted ducts, as has been
experienced in some bridges.
A landmark description of the development of reinforced concrete and steel piled
walls at Rotterdam was presented in 1966.44 Twelve years’ post-war experience
had given the Dutch engineers considerable confidence in the use of reinforced
concrete in decks and piles in conjunction with heavy section steel sheet pile walls
to form a typical relieving platform quay of the type shown in Figures 13.16 and
13.21. Examples of reinforced concrete caisson walls were given which, of course,
included those quoted earlier.30 In the discussion of this Chapter,45 the possible
corrosion problems arising from the use of steel piles and the use of higher

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:27:25
13.qxd 15/11/2001 15:58 Page 294

294 Historic concrete

Figure 13.21 Dutch relieving


platform, 1966 (from
Bokhoven44).

tensile steel reinforcement, which had previously been strictly avoided in maritime
works in order to reduce flexural cracking, were debated. At this time, the limi-
tation of crack width according to the hostility of the environment was being more
widely introduced. The crack width limit suggested was 0.1 mm. The concrete
strength was lower than might be the case in the UK, being 22.5 N/mm2, and the
cover was only 35 mm. The salinity of the water was low, of course, and, as had
been clearly stated in the 1923 papers,30 the cement used in Holland was likely
to have been a high slag blend.
It was now common for quays and jetties to consist of either beam and slab or
even thick plain reinforced concrete slabs seated on the growing range of steel
box or tubular piles, including the hexagonal Rendhex piles and composite sheet
piles. The steel piling was easier and quicker to drive and extend, and was capa-
ble of absorbing larger berthing and superstructure loads.
Port Talbot Harbour46,47 was the first tidal harbour to be built in the open sea
in the UK since the extension of Dover in 1909. It was built between 1966 and
1969 to accommodate iron ore carriers of 100,000 t deadweight and above. The
approach structure consisted of precast concrete beams carried on Rendhex steel
piles, with precast concrete slabs spanning between the beams (Figure 13.22). The
head of the berth consisted of a conventional concrete slab deck supported on
760 mm diameter steel tubular piles.
Two very large dry docks, one to accommodate tankers and bulk carriers up to
200,000 t deadweight and one to build ships up to 1 million tons deadweight, were
built in Belfast in 1965 and 1969, respectively.48,49 The more conventional repair
dock had a reinforced concrete floor, but walls of Peine steel sheet piling anchored

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:27:25
13.qxd 15/11/2001 15:58 Page 295

Reinforced and prestressed concrete in maritime structures 295

Figure 13.22 Port Talbot,


1969 (from McGarey and
Fraenkel 46).

by prestressed concrete raking piles. The shallower building dock was made of
normal reinforced concrete.
The last example of this period is the Brighton Marina,50,51 built in 1971 on an
exposed site with circular precast vertical breakwater caisson units 12.5 m diam-
eter. The characteristic strength of the concrete was 31.5 N/mm2 and reinforce-
ment was generally with high yield deformed bars. The walls were designed to
CP114, with 65 mm cover to reinforcement and surface crack widths were limited
to 5% exceedance of 0.26 mm.

The offshore scene began with light-houses. There is firstly the curious Nab Tower,
Offshore and built in the First World War with a reinforced concrete base. Kish Bank Lighthouse52
North Sea oil was built for the Commissioners of Irish Lights, 8 miles out of Dublin Bay, from 1963
to 1965, using a patented Swedish method of telescopic caisson construction. The
solid structure was founded directly on the seabed, and was designed for a wave
height of 13.7 m. The telescopic form of construction improved the stability dur-
ing towing and sinking. The outer caisson consisted of three concentric cylinders
standing on a 19 m thick base slab and interlocked by 12 radial walls (Figure 13.23).
The design required a 28-day cube strength of 31 N/mm2, and 4% entrained air.
The mix was 1 : 4.43 by weight with a water to cement ratio of 0.425. Ordinary Port-
land cement was used except for the outside wall extending from below mean sea
level to the top, where sulphate-resisting cement was used. The cover to the base
slab and external wall face was 50 and 38 mm to the inside faces of the outer wall.
A condition survey commissioned after some 25 years, in 1990, showed the struc-
ture to be in very good condition. Although difficult conditions restricted access,
the average cover to reinforcement recorded by covermeter was very close to the
specified figure. Measured equivalent cube strengths from cores at the first extended
landing were consistently high, being between 80 and 90 N/mm2.
The Royal Sovereign Light Tower53,54 was built a few years later, 11 km offshore
from Eastbourne, using the telescopic principle in a different way. The tower
was prestressed. Ordinary reinforced concrete was specified as 41 N/mm2 and

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:27:25
13.qxd 15/11/2001 15:58 Page 296

296 Historic concrete

Figure 13.23 Kish Bank


Lighthouse, 1965 (from
Hansen52).

prestressed as 55 N/mm2. Cover to reinforcement was 50 mm in ordinary reinforced


areas and 75 mm in prestressed areas. Sulphate-resisting Portland cement was used
in the wave zone. A limited condition survey55 was also conducted for this struc-
ture some 10 years later in 1980 under the ‘Concrete in the Oceans Project’.
Oil and gas in the North Sea lie some 3000 m below the waves, in depths of water
up to 150 m with wave heights of 30 m or more. The mass of a concrete structure
in these conditions is likely to be of the order of 250,000 t. A new generation of
concrete sea structures began in 1971 with the Ekofisk artificial island, 270 km from
the North Sea coast.56 The structure consisted of a nine-cell crude oil reservoir,
similar to a grain silo, measuring 50 m by 50 m in plan and 90 m high. This was
surrounded by a ‘Jarlan’ perforated breakwater screen with a mean diameter of
95 m (Figure 13.24). The prestressed structure was constructed by slipforming and
precasting, the whole construction upwards from the cellular foundation being
carried out afloat in deep but protected water. The concrete was specified at
50 N/mm2 cube strength, using plain Portland cement. As a result of an anticipated
subsidence of about 6 m caused by oil extraction, an additional protective barrier,
which forms a complete ring around the earlier structure, was completed in 1990.57
The Condeep structures for the Beryl A and Brent B production platforms were
ordered in 1973 and completed in 1975. They consisted of a base of 19 cylindrical
cells providing oil storage capacity and three tapering concrete shafts supporting
a steel deck.58 The plan dimension across five interconnected cells was 100 m
and the total height to the top of the shafts was 147 and 173 m for Beryl A and
Brent B, respectively, producing a truly impressive and beautiful structure
(Figure 13.25). Like Ekofisk, they were built partly in dry dock and partly afloat

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:27:25
13.qxd 15/11/2001 15:58 Page 297

Reinforced and prestressed concrete in maritime structures 297

Figure 13.24 Ekofisk artificial


island, 1973 (from Marion
and Mahfouz56).

Figure 13.25 Condeep


platform under construction,
1975.58

in a deep water fiord. The concrete technology for the prestressed structure was
described by Moksnes59 in 1975. Platforms of this nature were ordered and con-
structed in a similar way to ships, that is under certification and insurance provi-
sions. A British example is Cormorant ‘A’, 300,000 t, at Ardyne Point. (Figure 13.26).
In the period 1970–80 a considerable number of papers were written about
the requirements for concrete in what were seen as the onerous conditions of the

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:27:25
13.qxd 15/11/2001 15:58 Page 298

298 Historic concrete

Figure 13.26 Cormorant ‘A’


at Ardyne Point, 1974–1978
(Sir Robert McAlpine).

North Sea. A number of these papers60,61 by Browne and others reported the
current state of the art and established a confidence. In reality, however, the expo-
sure conditions of the cold North Sea as regards the performance of concrete and
reinforcement corrosion are not quite as aggressive as one might initially think,
and the nature of construction in prestressed concrete led almost automatically
to high grade concrete and high quality construction which withstood the condi-
tions for the required commercial service life, which is likely to be between 25 and
50 years.

Overseas and the British engineers were very active in the design and construction of port works
overseas in the post-war reconstruction era. Starting with the emerging African
Middle East states and then Libya, Aden and the new oil states of Kuwait and then the rest of
the Gulf states, Saudi Arabia and Iran, there was a strong accent on maritime facil-
ities. The speed of construction was usually of the essence, often with a shortage
of inland transport and hence locally available cement and aggregates. Although
mass concrete was preferred in a number of cases for a number of reasons, with
the demand for speed and economy and the increasing size of many of the facil-
ities, it is not surprising that steel-piled and reinforced concrete structures tended
to predominate.
There were, naturally, fears expressed concerning the use of structural steel
in the sea, which, in time, have largely been seen to be over-cautious. Renewed
confidence in the application of reinforced concrete, however, hit a number of snags
which were not so obvious or avoidable as is sometimes assumed, and which
are only now being more fully understood. The environment, in fact, provided
accelerated exposure conditions and served to broaden understanding, owing to
experience of conditions not met anywhere else worldwide, except perhaps in
bridge decks and tunnels. A number of examples are quoted by Gerwick.62

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:27:25
13.qxd 15/11/2001 15:58 Page 299

Reinforced and prestressed concrete in maritime structures 299

Some earlier rapid reinforced concrete failures in the Middle East had been dis-
missed as failures of materials and workmanship and rapidly demolished. What
had really been demonstrated was the limitation of standard design and details
to European and American practice which worked well enough in very different
exposure conditions elsewhere in the world. The comparison of environmental
conditions is quantified by Fookes.10
A graphic illustration of the significance of the water-to-cement ratio to the
discontinuity of capillary pores, drawing from the classic work of Powers and
Brownyards,63 amongst others, was given by Torben Hansen64 in 1989, who
described the deterioration of the concrete superstructure of a pier at Brega Indus-
trial Port in Libya after less than 10 years’ service. The high absorbency of sur-
face concrete of good quality under these conditions is explained in other papers,11
and it is in this context that the differential performance of reinforced concrete
structures in various parts of the world needs to be analysed.
Other major examples of reinforced concrete include the harbours at Port
Rashid, Dubai and Mina Zaed, Abu Dhabi, built from 1968 and the supertanker
dry docks at Bahrain (1973/74) and Dubai65,66 (1973/80). Feedback on these
projects is available in papers by John11,67 and others.

Conclusion In the maritime field, design and construction details are often more significant
than the choice of materials themselves. Reinforced concrete in the sea has a
relatively long and successful history, but many of the questions raised and dis-
cussed almost a century ago are sometimes revived as though they were novel.
Our predecessors, naturally, linked air and moisture with corrosion, but drew close
parallels with the performance of bare metal, masonry and timber. It is only
recently that the very different deterioration mechanism of reinforced concrete
in some specific maritime conditions has been better understood, and can explain
the very early observations12 that, depending on the magnitude of cover, it is the
infrequently wetted areas above the intertidal zone which can be most at risk,
because drying of the surface zone enables the chloride content to concentrate and
permits the flow of oxygen. Wetness itself, and regular ‘wetting and drying’, can
be an advantage. Concrete does not dry out below the surface layer, and the depth
of this layer is critical in arid conditions.67,68
Design for durability of reinforced concrete subjected to severe marine expo-
sure or deicing salts has not matched advances in structural computation.69,70 Unlike
land-based concrete, both the practice and expectations of concrete in the sea
remain controversial. The next generation of European (CEN) codes will include
a much improved classification of exposure conditions specific to the various dete-
rioration mechanisms, which is explained with reference to sea structures by Leem-
ing,71 Slater and Sharp,72 and is reflected in the revision of BS 6349, Pt. 1: 2000.73
Detailing practice and the specification of achievable tolerances for fixing rein-
forcement are likely to improve. Materials, particularly cement, continually change,
and so one cannot simply compare specifi-cations and achievements of different
dates, and codes of practice can take too long to adapt and recognize some basic
principles in relation to durability,
The above review of structures has been mainly drawn from the literature and
personal experience. Wherever possible, subsequent reports of earlier works have
been traced. As it is essential that future designs are based on the performance
of structures in the field and not on academic experiment, a scientific follow-up of
structures such as those that have been identified, with the interest and support of
their current owners, would provide an ideal sequel and benchmark of performance.
The author would welcome any feedback on the structures identified, or others.
Finally, an illustration of changing properties is provided by a recent case when
a contractor was surprised to meet ‘high’ concrete strengths when demolishing a

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:27:25
13.qxd 15/11/2001 15:58 Page 300

300 Historic concrete

1950s jetty on the Thames. He had expected strengths akin to the standard
20–25 N/mm2 specified during the 1960s. However, in those days, the mean
strength to meet 25 N/mm2 was likely to be 40 N/mm2. The concrete most likely
contained some 400 kg/m3 cement and had a water to cement ratio of about 0.42.
The mean 28-day strength, according to Road Note 4,74 would be some 40 N/mm2,
increasing to 65 N/mm2 after 1 year. Unlike modern cements, strength gain was likely
to continue, to be in the region of 80 N/mm2 today. Hopefully, the information
given in this book may reduce the risk of such a misunderstanding.
The author acknowledges the encouragement given by the associated authors
in this book, and James Sutherland (his first section leader in the design office)
and Mike Chrimes in particular. Also to his senior mentors in port engineering,
particularly Peter Scott,3 Harry Ridehalgh, Bob Daniels and Eric Loewy4 who, in
fact, recommended and reviewed the original paper.

References 1. Ker, H.J., The extension, widening and strengthening of Folkestone Pier. Discussion
comment by Sir William Mathews KCMG (President). Min. Proc. Instn Civ. Engrs, 1908,
171, 106.
2. Permanent International Association of Navigation Congresses. Proc. 13th International
Congress on Navigation, London, 1923, Section 11, Ocean navigation communications.
3. Scott, P.A., Port of Tema. Proc. Instn Civ. Engrs, 1965, 32, 211–53.
4. Loewy, E.I., Pannet, R.J., Collinge, H.S., Blockwork quay walls. Design construction
and utilisation in developing countries. PIANC 27th Int. Navigation Congress, Osaka,
May 1990, Section II — Subject 3, 141– 46.
5. Institution of Civil Engineers. Second Report of the Committee on Reinforced
Concrete. Institution of Civil Engineers, William Clowes & Son, 1913.
6. Institution of Civil Engineers. Preliminary and Interim Report of the Committee on
Reinforced Concrete. Institution of Civil Engineers, 1910.
7. Mouchel-Hennebique, Ferro-Concrete, 4th edn. L.G. Mouchel and Partners Ltd.:
Westminster, 1921.
8. Wentworth-Shields, F.E., Early marine structures. Concr. Constr. Eng. 1956, January,
25–29.
9. Wentworth-Shields, F.E., Gray, W.S., Reinforced Concrete Piling. Concrete Publications
Limited: London, 1938.
10. Fookes, P.G., A simple guide to risk assessment for concrete in hot dry salty environ-
ments. Proc. 4th International Conference on Deterioration and Repair of Reinforced
Concrete in the Gulf Region. Bahrain Society of Civil Engineers, 1993.
11. John, D.G., Corrosion deterioration of reinforced concrete structures in the Middle
East. Conference on Structural Improvement Through Corrosion Protection. The Insti-
tute of Corrosion: London, 1992.
12. Institution of Civil Engineers. First Report of the Committee Appointed to Investigate
the Deterioration of Structures of Timber, Metal and Concrete Exposed to the Action
of Sea Water. Institution of Civil Engineers, London, 1920.
13. Department of Scientific and Industrial Research. Deterioration of Structures of
Timber, Metal and Concrete Exposed to the Action of Sea Water. Fifteenth Report of
the Committee of the Institution of Civil Engineers, London, 1935.
14. National Building Studies. Research Paper No. 30. The Durability of Reinforced
Concrete in Sea Water. Twentieth report of the sea action committee of the Institution
of Civil Engineers. HMSO: London, 1960.
15. Walmsiley, A.T., Reinforced concrete piles in tidal waters. Concr. Constr. Eng.
1906–1907, 1, 88–100.
16. Pannel, J.P.M., Cement-gun repairs to maritime reinforced concrete structures with
special reference to the Town Quay, Southampton. Institution of Civil Engineers,
Maritime and Waterways Engineering Division Meeting, Maritime paper No. 2,
December 1945, 3–37.
17. Twelvetrees, W.N., Parkeston quay extension, Harwich. Concr. Constr. Eng.
1907–1908, 2, 166–94.
18. Parkeston Quay extension: report. Concr. Constr. Eng. 1933, 28, 63.
19. Parkeston Quay, Harwich: report. Concr. Constr. Eng. 1934, 29, 64.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:27:25
13.qxd 15/11/2001 15:58 Page 301

Reinforced and prestressed concrete in maritime structures 301

20. Extension to Parkeston Quay, Harwich: report. Concr. Constr. Eng. 1935, 30, 56–57.
21. Dyton, F.J., Case histories of repairs of maritime structures. Maintenance of Maritime
Structures. Institution of Civil Engineers, 1978, 89 –103.
22. Twelvetrees, W.N., Reinforced concrete wharves and quays. Article III. Docks and quays
on the Clyde. Concr. Constr. Eng. 1907–1908, 2, 273–78.
23. Berthing arm at Clacton Pier: report. Concr. Constr. Eng. 1935, 30, 58–59.
24. Ove Arup., Design of piled jetties and piers. Concr. Constr. Eng. 1934, 29, 37–41; 1935,
30, 41–48.
25. Comrie, J., The new fish dock, Grimsby. Concr. Constr. Eng. 1934, 29, 545–58.
26. Moller, A., Extension to Southend Pier. Concr. Constr. Eng. 1930, 25, 57–61.
27. Quay extension at Newcastle-upon-Tyne: report. Concr. Constr. Eng. 1934, 29, 65–69.
28. War Department jetty at Deptford: report. Concr. Constr. Eng. 1934, 29, 86–87;
677– 80.
29. Humphries, G.W., General Report. Concrete and reinforced concrete. Their applica-
tions to hydraulic works; means to insure their preservation and their watertightness.
Proc. 13th International Congress on Navigation, London, 1923, Section 11, Ocean
navigation communications. PIANC, Brussels.
30. Ringers, J.A., Tellegen, G., Report (Dutch experience), Holland. Proc. 13th Interna-
tional Congress on Navigation, London, 1923, Section 11, Ocean navigation commu-
nications. PIANC, Brussels.
31. Aki, K., Okabe, S., Reinforced concrete caisson for the Yokohama Harbour new
extension works. Proc. 13th International Congress on Navigation, London, 1923,
Section 11, Ocean navigation communications. PIANC, Brussels.
32. Reinforced concrete screw piles, slipway at Southend-on-Sea: report. Concr. Constr.
Eng. 1929, 29, 273–77.
33. Knight, C.W., Stork, S.G., Nos 1 & 2 Military Ports. The Engineer in War, Vol. 2. Docks
and Harbours. Institution of Civil Engineers, 1948, 3–35.
34. Wood, C.R.J., Phoenix (Mulberry units). The Engineer in War, Vol. 2. Docks and Har-
bours. Institution of Civil Engineers, 1948, 336–68.
35. Marine Durability of the Mulberry Harbour Units at Portland. Concrete in the Oceans
Project, Phase 2, Project 6. Taylor Woodrow Construction Ltd., Research Laborato-
ries, 1981.
36. Pannel, J.P.M., Cement-gun repairs to maritime reinforced concrete structures with
special reference to the Town Quay, Southampton. Institution of Civil Engineers,
Maritime and Waterways Engineering Division Meeting, Maritime paper No. 2,
Observations by Mr. G.M. Trehorne Rees, December 1945, 25.
37. Stott, F.P., Ramage, L.M., The design and construction of a dry dock at South Shields
for Messrs Brigham and Cowan Ltd. Proc. Instn Civ. Engrs, 1957, 8, October, 161–92.
38. Grove, G.C., Siting, design and construction of two terminals for large oil tankers. Proc.
Instn Civ. Engrs, 1964, 27, January, 99 –152.
39. Michelson, T., Concrete design and control — some information from the execution of
the deep water jetties for BP Oil Terminal, Angle Bay, Milford Haven. CN (Christiani
and Nielsen) Post, 1961, February, 24–27.
40. Carey, R., Cumming, C.G., The design and construction of Erith jetty. Proc. Instn Civ.
Engrs, 1961, 18, January, 15– 42.
41. A prestressed concrete quay at Tees dock: the substructure. Concr. Constr. Eng. 1961,
December, 413–20.
42. A prestressed concrete quay at Tees dock: the superstructure. Concr. Constr. Eng. 1962,
February, 71–77.
43. Tees Dock No. 1 Quay. Cement and Concrete Association: London, 1961.
44. Bokhoven, W., Recent quay wall construction at Rotterdam Harbour. Proc. Instn Civ.
Engrs, 1966, 35, December, 593–613.
45. Bokhoven, W., [Discussion on above reference.] Proc. Instn Civ. Engrs, 1967, 37,
685–700.
46. McGarey, D.G., Fraenkel, P.M., Port Talbot Harbour: planning and design. Proc. Instn
Civ. Engrs, 1970, 45, April, 561–92.
47. Ridgeway, R.J., Kier, M., Hill, L.P., Low, D.W., Port Talbot Harbour: construction. Proc.
Instn Civ. Engrs, 1970, 45, April, 593–626.
48. Geddes, W.G.N., Sturrock, K.R., Kinder, G., New shipbuilding dock at Belfast for
Harland and Wolff Ltd. Proc. Instn Civ. Engrs, 1972, 51, January, 17–47.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:27:25
13.qxd 15/11/2001 15:58 Page 302

302 Historic concrete

49. Ross, K., Rennie, W.J.H., Cox, P.A., The new dry dock at Belfast. Proc. Instn Civ. Engrs,
1972, 51, February, 269–94.
50. Terrett, F.L., Ganly, P., Stubbs, S.B., Harbour works at Brighton Marina: investigations
and design. Proc. Instn Civ. Engrs, 1979, 66, May, 191–208.
51. Llewellyn, T.J., Murray, W.T., Harbour works at Brighton Marina: construction. Proc.
Instn Civ. Engrs, 1979, 66, May, 209–26.
52. Hansen, F., Design and construction of Kish Bank Lighthouse. Trans. Instn Civ. Engrs
Ireland, 1965–66; 92, 248–99.
53. Antonakis, C.J., A problem of designing and building for a structure at sea (Royal Sov-
ereign Lighthouse). Proc. Instn Civ. Engrs, 1972, December, 95 –126.
54. Thorskov, S., The Royal Sovereign Light Tower. CN (Christiani and Neilsen) Post, 1962,
February, 12–22.
55. P66 final report, Concrete in the Oceans, Phase 11. Marine durability study of the Royal
Sovereign Lighthouse. Taylor Woodrow Research Laboratories, December 1981.
56. Marion, H., Mahfouz, G., Design and construction of the Ekofisk artificial island. Proc.
Instn Civ. Engrs, 1974, 56, November, 497–511.
57. Broughton, P., Waagaand, K., The Ekofisk protective barrier. Proc. Instn Civ. Engrs
Wat. Marit. Energy, 1992, 96, June, 103–19.
58. Norwegian Contractors. Promotional literature. Oslo, 1975.
59. Moksnes, J., Condeep platforms for the North Sea — some aspects of concrete tech-
nology. Offshore Technology Conference, Houston, Texas, May 1975, 339–50.
60. Browne, R.D., Domone, P.L., Concrete for surface and underwater structures. Sym-
posium on Materials for Underwater Technology, Admiralty Materials Laboratory,
Poole, 1973, Society for Underwater Technology.
61. Browne, R.D., Domone, P.L., Geoghegan, M.P., Deterioration of concrete structures
under marine conditions — their inspection and repair. Institution of Civil Engineers
Conference on Maintenance of Marine Structures, London, October 1977.
62. Gerwick Jr., B.C., International experience in the performance of marine concrete.
Concrete International, American Concrete Institute, May 1990, 47–53.
63. Powers, T.C., Brownyands, T.C., Studies of the physical properties of hardened Port-
land cement paste. Research Laboratories of the Portland Cement Association, Chicago,
Bulletin No. 22, March 1948.
64. Hansen, C.T., Marine concrete in hot climates — designed to fail. Mater. Structs,
RILEM, 1989, 22, 344–46.
65. Daniels, R.J., Sharp, B.N., Dubai Dry Dock: planning, direction and design consider-
ations. Proc. Instn Civ. Engrs, 1979, Part 1, 66, 75–92.
66. Cochrane, G.H., Chetwin, D.J.L., Hogbin, W., Dubai Dry Dock: design and construc-
tion. Proc. Instn Civ. Engrs, 1979, Part 1, 66, 93–114.
67. John, D.G., Leppard, N., Wyatt, B.S., Cathodic protection repair applied to reinforced
concrete deck support beams for Mina Zayed Port, Abu Dhabi. Proc. 4th International
Conference on Deterioration and Repair of Reinforced Concrete in the Gulf Region,
Bahrain, 1993.
68. Bijen, J.M., Durability aspects of the King Fahd Causeway. RILEM Conference on
Concrete in Hot Climates, Torquay, September 1992.
69. Aitcin, P.C., Durable Concrete. In: S. Nagataki, T. Nireki, F. Tomasawa (eds), Dura-
bility of Buildings and Components, Vol. 6, E & F.N. Spon, 1993.
70. Rostam, S., Schiessl, P., Next generation design concepts for durability and performance
of concrete structures. In: S. Nagataki, T. Nireki, F. Tomasawa, (eds), Durability of Build-
ings and Components, Vol. 6. E & F.N. Spon, 1993.
71. Leeming, M.B., Durability of concrete in and near the sea. In: R.T.L. Allen (ed.), Con-
crete in Coastal Structures. Thomas Telford, 1998, Chapter 3.
72. Slater, D., Sharp, B.N., Design guides, specifications and the design of coastal struc-
tures. In: R.T.L. Allen (ed.), Concrete in Coastal Structures. Thomas Telford, 1998,
Chapter 4–6.
73. British Standards Institution. Maritime Structures, Part 1, Code of Practice for gen-
eral criteria. BS 6349: Pt. 1: 2000.
74. Road Research Road Note No. 4. Design of concrete mixes. Department of
Scientific Research, Road Research Laboratory, HMSO, 1950.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:27:25
14.qxd 15/11/2001 15:58 Page 303

14 The Concrete Institute 1908 –23,


precursor of the Institution of
Structural Engineers

Anita Witten
Synopsis The reasons for the foundation of the Concrete Institute are briefly outlined. Its
organization and scope are described, and its achievements, particularly its role
in drafting the London County Council Reinforced Concrete Regulations and in
establishing a standard notation for reinforced concrete, are discussed. A short
analysis of its membership is given.

Introduction There are subtle variations of emphasis in the accounts of the reasons for the found-
ing of the Concrete Institute (from 1923 the Institution of Structural Engineers),
but essentially there was concern amongst building control officers, architects, engi-
neers and others that concrete and designs using it should be better understood,
not least by those ultimately responsible if failure occurred. The consensus of the
Institution of Civil Engineers (ICE), whose members1 had long experience of
concrete for foundations and were certainly using it for buildings, was that regu-
lations for reinforced concrete construction would unduly fetter its members while
there was still so much about the material which was imperfectly understood.2
Others saw regulations as a necessary step forward.3 The early history of
the adoption of reinforced concrete as a building material in the UK is described
in papers by Hurst57 and Bussell58 elsewhere in this issue and in works such as
Hamilton.5
It seems likely that the number of people across several professions involved in
construction using concrete and the number wishing to know more about it and
its uses, none of whose needs were being adequately met by the existing institu-
tions, was sufficient to encourage the formation of another body. Concrete was the
first new construction material with such great potential for all sorts of uses for
several generations, and it would be surprising if it did not cause a ferment in the
industry as people began to become familiar with it.
The driving force behind the Concrete Institute was Edwin O. Sachs (Figure
14.1), a dynamic architect with much experience of theatre design and the effects
of fire upon buildings. In 1897 he established the British Fire Prevention
Committee and as its Chairman initiated the first independent fire testing station
Figure 14.1 Edwin O. Sachs. in the world,4 and in 1906 he founded the excellent journal Concrete and
Constructional Engineering (although his involvement with this was probably not
public knowledge).

Events preceding At this distance we must rely on published information and one or two published
histories such as that by Hamilton.5 Table 14.1 lists the important publications
the foundation which preceded the founding of the Institute. Sachs was ‘induced’ to look for
a suitable basis for forming a scientific society that appreciated and welcomed
the cooperation and advice of all those commercially concerned in concrete

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:28:19
14.qxd 15/11/2001 15:58 Page 304

304 Historic concrete

Table 14.1 Important publications preceding the founding of the Concrete Institute

Date Content

8 April 19056 Letter from William Dunn summarizing the current position and urging
the RIBA, with the ICE, to appoint a committee to prepare a standard
specification for reinforced concrete work (including calculations) to
help architects to assess designs submitted by the specialists
March 1906 First issue of the journal Concrete and Constructional Engineering.
Initially bi-monthly, recent works in reinforced concrete at home and
abroad were described and illustrated, extensive reviews of the
position of reinforced concrete in other countries given, and
developments in theory and practice noted. Edited and partially
funded by Edwin O. Sachs7
28 April 19068 Report on the first meeting of the Joint Reinforced Concrete Committee,
chaired by Sir Henry Tanner and comprising representatives of RIBA,
the District Surveyors Association, the Institute of Builders, the
Incorporated Association of Municipal and County Engineers, the War
Office, the Admiralty and ‘distinguished scientists’
December 19069 Announcement of the formation by the British Fire Prevention
Committee of a Special Commission on Concrete Aggregates, chaired
by Sir William Preece, and comprising engineers, architects, surveyors,
firemen, cement manufacturers and representatives of the Admiralty
and War Office for the purpose of reporting on and defining ‘the
aggregates suitable for concrete floors intended to be fire-resisting,
having due regard to the question of strength, expansion and the
chemical constituents and changes of the aggregates’
27 May 190710 First report of the RIBA Joint Reinforced Concrete Committee adopted
at RIBA General Meeting
Autumn 1907 Circular issued by a member of one of the reinforced concrete firms,
proposing the formation of a society made up of those firms
First half of 190811 Reports in the industry press of the formation of the Concrete Institute

whilst having the standing of a technical institution of the first order and the active
guiance of the leaders of the technical professions affected.12

The new Institute The first meeting of the Council of the new Institute was held on the 21st of July,
1908 at the Ritz Hotel, with Sachs in the chair. The Earl of Plymouth, First
Commissioner of Works 1902–1905, was appointed as President.13 One hundred
founder members had signed up14 at a subscription of 1 guinea (£1.05). Tables 14.2
and 14.3 show the employment of Council members and the overlaps in
membership between the influential bodies, in collaboration not competition,
and Figure 14.2 shows the growth in membership of the new body. The Council
members must all have been known personally by Sachs; this shows the breadth
of his interests and contacts in the industry and his ability to inspire hard
work and enthusiasm in others. The objects of the Institute were agreed by the
meeting to be:

(a) to advance the knowledge of concrete and reinforced concrete, and direct
attention to the uses to which these materials can be best applied;
(b) to afford the means of communication between persons engaged in the design,
supervision and execution of works in which concrete and reinforced
concrete are employed (excluding all questions connected with wages and
trade regulation);
(c) to arrange periodical meetings for the purpose of discussing practical and
scientific subjects bearing upon the application of concrete and reinforced

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:28:19
14.qxd 15/11/2001 15:58 Page 305

The Concrete Institute 1908–1923 305

Table 14.2 Council members in 1908 also sitting on other committees

Council member ICE Council RIBA 1907 BFPC Special RIBA 1911 ICE 1908 RC ICE 1911 RC
and employment member at Joint Committee Commission Joint Committee Committee Committee
some time on RC on Concrete on RC
Aggregates

Earl of Plymouth
Sir Douglas Fox, x
President ICE 1899–1900
Sir William Mather
Sir William Preece, x x
President ICE 1989–99,
Chief Engineer, GPO
Sir Henry Tanner, x x x
HM Office of Works
Edwin O. Sachs x
H.H.D. Anderson,
manufacturer of
Portland cement
B. Blount x x
A.E. Collins, x x x
City Engineer,
Norwich
C.H. Colson, Admiralty x x
William Dunn, Architect x x
B. Hannen,
Cubitt & Co.
W.T. Hatch, x
Chief Engineer,
Metropolitan
Asylums Board
W.H. Hunter, x
Chief Engineer,
Manchester
Ship Canal
W.H. Johnson,
Chairman, Johnson,
Clapham & Morris
C.F. Marsh, x x
Metropolitan
Water Board
F. May, Chairman, x x
Trussed Concrete
Steel Co.
J. Munro, Director,
Stuart’s Granolithic Co
F. Purton, Manager,
New Expanded
Metal Co.
A. Ross, President x x
ICE 1915–16, Chief
Engineer, Great
Northern Railway
L. Seraillier, Manager,
Patent Indented
Steel Bar Co.
J.S. de Vesian,
Mouchel & Partners
J. Winn, School of x
Military Engineering
G.C. Workman, Coignet
E.P. Wells

BFPC: British Fire Prevention Committee; RC: Reinforced Concrete.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:28:19
14.qxd 15/11/2001 15:58 Page 306

306 Historic concrete

Table 14.3 Composition of the ‘concrete bodies’

Concrete RIBA Joint BFPC Special RIBA Joint


Institute Committee on Commission Committee on
Council 1908 Reinforcced on Concrete Reinforced
(25 in total) Concrete 1907 Aggregates Concrete 1911
(15 in total) (23 in total) (18 in total)

Architects 2 5 4 6
Engineers 9 2 14 8
Contractors or 8 1 1 0
producers of
proprietary
systems
A public service 6 4 8 5
Unascribed 6 5 5 3

Excluding those unascribed, some people fall into two categories.

Figure 14.2 Number of


members, derived from
reports in the Concrete
Institute Transactions and
annual reports (excludes
‘special’ subscribers and
honorary members).

concrete, and to conduct such investigations and to issue such publications


as may be deemed desirable.
The qualifications for membership were:
(a) persons professionally or practically engaged in the application of concrete
or reinforced concrete and the production of their constituents;
(b) persons of scientific, technical or literary attainments specially connected with
the application of concrete, reinforced concrete and their constituents.15
All members of the Council were to be British subjects.
It seems likely that many of the founder members were nearing the end of their
careers and enrolled to encourage the new body; there are quite a number of
deaths recorded in the first volumes of Transactions and others doubtless retired.
By April 1910, when the first list of members was published, only 89 founder
members remained; this was reduced to 78 by February 1912.
Four committees were established immediately (science, parliamentary, rein-
forced concrete practice and tests), and their terms of reference approved in
March 1909.16

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:28:19
14.qxd 15/11/2001 15:58 Page 307

The Concrete Institute 1908–1923 307

Administration Initially the administrative work was done by an Executive of six members, chaired
by Sachs, which included the Honorary Secretary, A.E. Collins, who was city engin-
and organization eer of Norwich (and resigned at the end of 1909 due to pressure of work) and the
Honorary Treasurer, E.P. Wells. Sachs did a great deal of the day-to-day work him-
self — the Institute’s offices were at 1 Waterloo Place, Pall Mall, London, SW1, while
his own address at the time as 7 Waterloo Place — and it proved difficult to get
the Executive together as often as necessary.17 The Institute got off to a good start,
but there was a hiatus when Sachs was taken ill in November 1909 and was still
convalescing at the first Annual General Meeting (AGM) on the 17th of February,
1910.18 Only 10 days’ notice of the date of the AGM was given (contrary to the
Articles of Association, which stipulated 30) and there were one or two discussions at
the meeting which show clearly the lack of anyone present with a sufficient work-
ing knowledge of the Articles. Sachs would undoubtedly have had this knowledge.
As a result of this obvious over-reliance on one person, the Articles were changed
so that Council had the responsibility and the management of the Institute’s affairs.
Sachs seconded the resolution and received an enthusiastic vote of thanks at the
end of the meeting which approved the changes.19 A permanent secretary was
appointed to assist Council.20
Possibly because Sachs was out of action for quite a long period,21 the Institute
floundered slightly in its direction forward. This may have been a natural occur-
rence because the initial momentum had worn off and much time needed to be
spent on the London County Council (LCC) regulations, and because Sachs’
undoubted vision and energy were missing from Council meetings. However, around
the end of 1910 the Earl of Plymouth retired from the office of President and
Sir Henry Tanner was appointed in his place. He was Chief Architect to HM Office
of Works from 1898 to 1913 and promoted the use of reinforced concrete in the UK,
overseeing the first official use in the UK of reinforced concrete for a complete
building in the King Edward Building, Newgate Street, London, for the Post Office,
and he capably and carefully steered the Institute through the next 15 months,22
and was succeeded by E.P. Wells. Sachs was appointed as fifth Vice-President.
During 1911 the way forward was considered, and this resulted in approval
of a proposal before Council for extending the scope of the Institute. A sub-
committee of ten Council members23 considered how to put it into practice, and
its report was adopted at the third AGM on the 9th of May, 1912. The revised
objects were:
(a) to advance the knowledge of concrete and reinforced concrete, and other
materials employed in structural engineering, and to direct attention to the
uses to which these materials can be best applied;
(b) to afford the means of communication between persons engaged in the design,
supervision and execution of structural engineering works (excluding all ques-
tions connected with wages and trade regulation);
(c) to arrange periodical meetings for the purpose of discussing practical and
scientific questions bearing upon the application and use of concrete and
reinforced concrete and other materials employed in structural engineering
for any purpose whatsoever.24
Structural engineering was defined as ‘that branch of Engineering which deals
with the scientific design, the construction, and the erection of structures of all kinds
in any materials’, and structures as ‘those constructions which are subject princi-
pally to the laws of Statics, as opposed to those constructions which are subject prin-
cipally to the laws of Dynamics and Kinematics, such as engines and machines’.25
The full title was changed to ‘The Concrete Institute: an institution for structural
engineers, architects, etc.’, the breadth of the Institute’s membership being
considered one of its great strengths, allowing all points of view on a subject under
investigation to be taken into account.26

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:28:19
14.qxd 15/11/2001 15:58 Page 308

308 Historic concrete

An annual course of technical lectures on ‘some branch of structural engineering’


was to be started, and examinations in structural engineering were to be held annu-
ally, ‘to test the scientific or technical attainments of applications for Studentship’.24
Other small changes to the Transactions and meeting arrangements were made, and
a Bronze Medal was introduced for the best paper presented each session (Table 14.4).
This new scope had a beneficial effect on the number of members,27 which
continued when additional classes of membership were added28 and the first
Honorary Members elected (Table 14.5).29
During the 1913–14 session, informal meetings of junior members were insti-
tuted, and an Examination Board was established in December 1913, with Professor
Henry Adams as Chief Examiner, to run entrance examinations for the Graduate
and the Associate Member (now Member) grades of membership. A syllabus was
compiled30 and it was proposed to hold the first examination in 1915.
The First World War, consequent increases in the cost of paper and printing,
and some controversy over policies31 brought things to a relative standstill for a
few years. The publication of the Transactions was postponed and the content
reduced32 (volume 11 was never published at all) and from 1920 onwards
announcements, very brief meeting reports, etc., were published in Concrete and
Constructional Engineering.33 Work and discussion on the various reports in progress
continued through the war, but publication was greatly delayed.
The first examinations were deferred, and were eventually held on the 13th
and 14th of May, 1920. By 1920 educational lectures had resumed, and visits of

Table 14.4 Bronze medals awarded for the best paper presented in each session

Session Title and author

1910–11 The aesthetic treatment of concrete, by Beresford Pite


1911–12 Fireproofing, by Richard L. Humphrey
1912–13 Steel-frame buildings in London, by S. Bylander
1913–14 Calculations and details for steel-frame buildings from the draughtsman’s
standpoint, by W.C. Cocking
1914–15 The design of quay walls, by F.E. Wentworth-Shields
1915–16 Shearing resistance of reinforced concrete beams, by Oscar Faber
1916–17 Southampton Docks: re-modelling of an old dry-dock, by Robert N. Sinclair
1917–18 No records available
1918–19 The geology of aggregates and sands, by P.G.H. Boswell
1919–20 The attrition of concrete surfaces exposed to sea action, by J.S. Owens
1920–21 No records available
1921–22 What is the use of the modular ratio? by H. Kempton Dyson

Table 14.5 Honorary Members elected during the 1912–13 session

C. Bach Director of the Materialprüfanstalt, Stuttgart


N. Belelubsky Director of the Technical Laboratories, St Petersburg
H. Le Chatelier Inspector General of Mines, Paris
Paul Christophe Principal Engineer, Office of Works, Brussels
Edmond Coignet Director of Edmond Coignet Ltd.
Armand Considère Inspector General of Public Works, Paris
Fritz von Emperger Editor of Beton und Eisen, Vienna
R. Feret Chief of the Technical Laboratories, Boulogne
F. Hennebique Director of the Hennebique Ferro-Concrete Company
A. Martens Principal of the Prussian Testing Laboratories, Berlin
A. Mesnager Director of the Technical Laboratories, Paris
Emil Morsch Chief Engineer, Wayss & Freitag AG
C. Rabut Engineer to the Office of Works, France
F. Schule Director of Materialprüfungsanstalt, Zurich
Arthur N. Talbot Professor of Engineering, University of Illinois, USA
W.C. Unwin President ICE 1911–12

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:28:19
14.qxd 15/11/2001 15:58 Page 309

The Concrete Institute 1908–1923 309

inspection (including one to the works for the new Selfridge store) were being
arranged again.
The number of members, the finances and activities continued to be healthy after
the war, and it seems that during 1921 the idea of changing the title to
The Institution of Structural Engineers was considered.34 Very little background
information is available on this as the Council papers are not available at the time
of writing and published sources say nothing, but as no changes were made to the
Objects of the Institute established in 1912, it seems likely that the new name
generally reflected the facts. Some concern at the new title was reflected at the ICE
Council meeting on the 26th of September, 1922, but it was decided to take no action.
The title change was approved on the 28th of September, 1922 and The Struc-
tural Engineer issued monthly from January 1923. Branches, first mooted some
years earlier, were now permitted, and the Lancashire and Cheshire Branch was
inaugurated in November 1922, followed by the Western Counties Branch in
January 1923. The first Gold Medal of the Institution of Structural Engineers was
awarded to Henry Adams in recognition of his services to the Institution and his
long and distinguished service as Chairman of the Board of Examiners.35
All the signs are of a professional association flinging itself confidently, with
energy and purpose, towards the future.

Membership The Institute attracted members from all professions with an interest in reinforced
concrete. Figures 14.3 and 14.4 show the remarkably consistent professional and
geographical spread over time. Most of those abroad were working in the British
Empire or on railway construction. The subscription was initially 1 guinea (£1.05),
raised to 2 guineas (£2.10, or £105.50 at 1993 prices) in 1913 when the membership
reached 1000.

Achievements Meetings

Papers on a wide range of subjects were presented and discussed at meetings of


the Institute (Table 14.6). There are useful remarks about various aspects of the
industry in the verbatim discussions published in the Transactions (e.g. Seraillier36).

Building regulations

In December 1908 the Institute was asked by the LCC to consider proposed amend-
ments to the London Building Acts included in the LCC (General Powers) Bill to
be presented to Parliament in 1909, and to respond almost immediately. In Janu-
ary, the LCC requested delegates from ‘the leading societies connected with build-
ing works’37 to consider the proposals in the Bill ‘regulating the erection of buildings
of iron or steel skeleton construction’.38 William Dunn (Chairman of the Science
Standing Committee and member of the RIBA Joint Committee on Reinforced Con-
crete) and E.P. Wells (who had much experience of testing work and research) rep-
resented the Concrete Institute. In addition, the Parliamentary Standing
Committee put in considerable effort petitioning against the Bill.39 As a result, when
the Act came into force, under Section 23 the Concrete Institute, the Institution
of Civil Engineers, the RIBA and the Surveyors Institution (now the RICS) were
to consider proposed regulations to do with the construction of buildings wholly
or partly of reinforced concrete. During 1910, the Concrete Institute committees
spent 19 h discussing these draft regulations and recommending amendments in
addition to those put forward by the four societies in conference.40 The annual
reports summarize the progress of the consultation;41 an examination of the orig-
inal LCC proposals and of the amendments recommended by all the various

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:28:19
14.qxd 15/11/2001 15:58 Page 310

310 Historic concrete

Architects or
Surveyors
Other 7%
professionals
6%
Other
Architects/other professionals
professionals 14%
16%
Architects
Engineers
or Surveyors
32% Others
9%
41%
Others
49%
Engineers
Engineers 38%
Others 36%
52%

(a) (b) (c)

Architects or Other professionals Other professionals


Surveyors Other professionals
5% 5%
Architects or 5% Architects or
Other 7% Architects
Surveyorsor Surveyors
professionals
Surveyors
9% 9%
6%
9%

Others
Figure 14.3 (a) Founder Others 45%
Others Others
45%
members; (b) members 49% Engineers 45% Engineers
Engineers
elected Engineers
in 1908; (c) members 41%
41%
41%
38%
elected in 1909; (d) member-
ship in 1910; and (e) mem-
bership in 1919 (derived from
designatory letters given in
the 1910 and 1919 List of
members). (d) (e)
(c) (d)

Abroad Abroad
30% 36%

UK UK
70% 64%

Figure 14.4 (a) Location of


members in 1910; and
(b) location of members in
1919 (derived from the 1910
and 1919 List of members). (a) (b)

bodies would be interesting in charting the development of reinforced concrete,


but unfortunately space does not permit it here. The draft of the first set of
Regulations occupied 12 h of Institute committee time in 1911–12 and that of the
second set 47 h in the 1913–14 session.42 The RIBA and Surveyors Institution
supported most of these recommendations, and the ICE did not make any detailed

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:28:19
14.qxd 15/11/2001 15:58 Page 311

The Concrete Institute 1908–1923 311

Table 14.6 Papers presented to the Concrete Institute

1908 The composition and uses of plain and reinforced concrete, by Charles F. Marsh
The examination of designs for reinforced concrete work, by William Dunn
1909 The commercial aspect of reinforced concrete, by Lucien Seraillier
Concrete in arched bridge construction, by E.P. Wells
Some notes relating to the setting of Portland cement with description of
method adopted for regulating the same, by H.K.G. Bamber
Some points relating to reinforced concrete as applied in the United States, by
R.L. Humphrey
1910 Reinforced concrete chimney construction, by Ernest R. Matthews
Notes on the Le Chatelier boiling test of Portland cement, by D.B. Butler
The effects of sewage and sewage and sewage gases on Portland cement
concrete, by Sidney H. Chambers
Reinforced concrete bins, by H. Kempton Dyson
The British Aluminium Company’s works at Kinlochleven, by A. Alban H. Scott
The manufacture of Portland cement, by A.C. Davis
General concrete practice, by Thomas Potter
1911 The dissociation of competitive designs and tenders, by R.W. Vawdrey
Swanscombe reinforced concrete pier, by C. Percy Taylor
The aesthetic treatment of concrete, by Beresford Pite
The YMCA building, Manchester, by Alfred E. Corbett
Some recent works in reinforced concrete, by G.C. Workman

amendments.43 The Concrete Institute’s suggestions were largely embodied in


revised draft regulations44 which came into force on the 1st of January, 1916,45 fol-
lowed in 1916 by the first explanatory handbook.46 Debate in the industry, and the
Institute’s involvement continued. The development of design theory and practice
during the 1920s and 1930s is described in Bussell’s58 paper elsewhere in this issue.

International standard algebraical notation

In response to a letter to members asking for suggestions for matters which


the new Concrete Institute should consider, J. Sherwood Todd suggested the
consideration of the standardization of methods and symbols in calculations for
reinforced concrete work, and E. Fiander Etchells (President 1920–23) suggested
the publication of a standard notation for reinforced concrete formulae. Both sug-
gestions were discussed by the Science Standing Committee, together with a draft
international notation proposed by the International Commission on Reinforced
Concrete (an offshoot of the International Association for Testing Materials) which
had been forwarded to Sachs who was the UK representative on the Commission.
The principles adopted were:
(a) the use of initial letters;
(b) the use of significant subscripts;
(c) discrimination between the use of small and capital letters;
(d) sparing use of the Greek alphabet.47
A draft report was issued on the 28th of July, 1909, but the principle of using
initial letters was found to be a barrier to its adoption in Europe48 (different
languages preferring different initial letters) and the Continental use of Greek
letters for intensity of stresses would not be readily adopted in the UK and USA.
The draft was despatched to the American Joint Committee on Concrete and
Reinforced Concrete for consideration. It was adopted for use in textbooks,
for the second RIBA report49,50 and by the LCC for its proposed reinforced
concrete regulations. Revisions were made periodically and a version with explana-
tory notes by E. Fiander Etchells was published by E. and F.N. Spon in 1918.51

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:28:19
14.qxd 15/11/2001 15:58 Page 312

312 Historic concrete

Other activities

A library was established as soon as space was available, and members were asked
to donate books. The first list of additions appeared in the Concrete Institute Trans-
actions, Vol. 3, page xiii.
The Institute was invited to appoint representatives to various committees and
to participate in joint efforts, including:
(a) in 1908, the Special Commission on Concrete Aggregates (set up by the British
Fire Prevention Committee);
(b) in 1910, the Engineering Standards Association (now the British Standards
Institution) Sectional Committee on Portland Cement and the Sectional
Committee on Bridges and Building Construction;
(c) in 1910, the RIBA Joint Committee on Reinforced Concrete;
(d) in 1913–14, the International Association for Testing Materials, for reporting
of accidents to reinforced concrete buildings;
(e) in 1916–17, the conference convened by the District Surveyors Association
to consider the interpretation of the LCC (General Powers) Act 1909 with
reference to steel-frame buildings.52 The report was complete and awaiting
publication in May 1918.53
The Institute also initiated various investigations by its committees, although
too much was attempted for the time and finance available (the LCC regulations
took up a great deal of time) and nothing of great impact resulted, although pub-
lication in the Transactions (and discussions) of several interim reports must have
been helpful in disseminating current thinking and are indicative of the state of
knowledge in the UK. Some of the subjects investigated were:
(a) a standard specification for reinforced concrete work;
(b) adhesion of and friction between concrete and steel;
(c) reinforced concrete piles;
(d) the effects of oils and fats on concrete;
(e) standardization of methods of taking off quantities;
(f) advice to clerks of works and others on methods of properly executing con-
crete work;
(g) standard concrete mixtures;
(h) the use of cinder, ash and breeze in concrete;
(i) methods of making concrete watertight.

Several cooperative efforts came to fruition. In 1910 the Joint Committee for
Loads on Highway Bridges was set up, with representatives from the Institution
of Municipal and County Engineers and the Institution of Municipal Engineers.
Their report was published in 1918. In 1913–14, jointly with the Quantity Surveyors
Association (now part of the RICS), drafting began on a Standard Method of Mea-
surement for Reinforced Concrete. It was also considered by the National Fed-
eration of Building Trades Employers of Great Britain and Ireland, and by the
Institute of Builders. The final report, Measurement of reinforced concrete in building
works, was approved during the 1914–15 session54 and published by the Quantity
Surveyors Association. The Concrete Institute wished to extend it with a second
part on engineering works and intended to publish both parts at a later date, but
this seems never to have happened.
In 1915–16 an advisory council appointed by the Committee of the Privy Coun-
cil for Scientific and Industrial Research requested information on specific prob-
lems of concrete industry which required ‘scientific investigation’.55 The Institute
offered further help and later applied for a research grant, intending to coordi-
nate research in various different universities and laboratories. In the following

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:28:19
14.qxd 15/11/2001 15:58 Page 313

The Concrete Institute 1908–1923 313

Table 14.7 Publications

1918 Mnemonic notation for engineering formulae


Report of Joint Committee on Loads on Highway Bridges
Recommendations to clerks of works and foreman concerning the execution
of reinforced concrete works
1920 A standard specification for reinforced concrete work
1921 Reports to the Department of Scientific and Industrial Research on the
research work on concrete carried out under the direction of the Concrete
Institute during the years 1917–19
1924 International Cement Congress 1924: report on papers and discussions
1925 Report of Joint Committee on Loads on Highway Bridges (2nd edn)
1926 Scale of charges for consulting structural engineers
Report on aluminous, rapid hardening, Portland and other cements of
a special character
1927 Report on steelwork for buildings. Part I: loads and stresses
Report on steelwork for buildings. Part II: details of design and construction
Report on loads and stresses for gantry girders
1928 Report on reinforced concrete for buildings and structures. Part III: materials
and workmanship

session (1916–17) a programme was drawn up for testing the properties of


various sands, aggregates and their concretes, and work started at various
universities.56
After the war, some publications started to appear and once the Institution of
Structural Engineers got going several more were published (Table 14.7) within
a few years of its name change which were particularly important; the valuable
experience of the early years had not been wasted.
It would be interesting to compare the initial aims and the influence and effec-
tiveness of the Concrete Institute with those of other bodies formed later: the Rein-
forced Concrete Association (founded c. 1930), the Cement and Concrete
Association (founded 1935) and the Concrete Society (founded 1966). Unfortu-
nately, that must remain a subject for future study.

References 1. Harrison, C.A., Reinforced concrete for railway structures. 1907 Engineering Con-
ference. Institution of Civil Engineers: London, 1907: 24–25.
2. Read, R.D.G., Discussion on reinforced concrete structures. 1907 Engineering Con-
ference. Institution of Civil Engineers: London, 1907: 56–57. (Unpublished letter from
the Institution of Civil Engineers to London County Council, 21 November 1910.)
3. Anon., The Concrete Institute. Builders J. Arch. Eng., 1908, February, 197.
4. Welch, C., London at the Opening of the 20th Century. W.T. Pike & Co: London, 1905;
Anon., Our Fiftieth Anniversary. Concr. Constr. Eng., 1956, 51, 1–3; Anon., Obituary
Notice. Concr. Constr. Eng., 1919, 14, 556.
5. Hamilton, S.B., A note on the history of reinforced concrete in buildings. National
Building Studies Special Report 24. HMSO: London, 1956.
6. Dunn, W., Reinforced concrete floors. RIBA J., 1905, 8 April, 373.
7. Anon., Our Fiftieth Anniversary. Concr. Constr. Eng., 1956, 51, 1–3.
8. Anon., Chronicle. RIBA J., 1906, 28 April.
9. Concr. Constr. Eng., 1, 408.
10. RIBA J., 1907, 1 June, 515–41.
11. Anon., A Concrete Institute. Concr. Constr. Eng., 1908, 2, 423; Anon., A Concrete Insti-
tute. Builders J. Arch. Eng., 1908, 29 January, 95; Anon., A Concrete Institute. 26 Feb-
ruary 1908, 197; Anon., The Concrete Institute. RIBA J., 1908, 9 May, 412.
12. Sachs, E.O., Memorandum. Concr. Inst. Trans., 1, v.
13. Anon., The Righ Hon. the Earl of Plymouth, GBE, CB etc. Struct. Engr, 1923, 1, 76–77.
14. Anon., The Concrete Institute. Builders J. Arch. Eng., 26 February 1908, 197.
15. Anon., Notes: constitution. Concr. Inst. Trans., 1, vii.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:28:19
14.qxd 15/11/2001 15:58 Page 314

314 Historic concrete

16. Anon., References to Standing Committees. Concr. Inst. Trans., 1, xxiii.


17. Concr. Inst. Trans., 3, 12.
18. Anon., Discussion on annual report and accounts. Concr. Inst. Trans., 2, 75.
19. Anon., Extraordinary general meeting. Concr. Inst. Trans., 3, 1–3.
20. Second annual report, Concr. Inst. Trans., 3, 224.
21. Anon., Our Fiftieth Anniversary. Concr. Constr. Eng., 1956, 51, 3.
22. Concr. Inst. Trans., 4, 275, 285.
23. Concr. Inst. Trans., 4, 163– 64.
24. Anon., Objects of the Institute. Concr. Inst. Trans., 4, iv.
25. Concr. Inst. Trans., 4, 272.
26. Concr. Inst. Trans., 4, 278–79.
27. Anon., Report of Council for 1912–13 session. Concr. Inst. Trans., 5, 407; Concr. Inst.
Trans., 6, 2.
28. Concr. Inst. Trans., 5, 414.
29. Concr. Inst. Trans., 5, 411.
30. Anon., Syllabus of the examination. Concr. Inst. Trans., 6, 21–24.
31. Anon., Award of the gold medal to Professor Henry Adams. Struct. Engr, 1923, 1, 33.
32. Concr. Inst. Trans., 9, Diii.
33. Anon., The Concrete Institute. Concr. Constr. Eng., 1920, 15, 444.
34. Anon., Annual general meeting and annual report. Concr. Constr. Eng., 1922, 17, 407.
35. Anon., Award of the gold medal to Professor Henry Adams. Struct. Engr, 1923, 1, 33.
36. Serraillier, L., Discussion at Fourteenth Meeting of the Concrete Institute. Concr. Inst.
Trans., 3, 115–16.
37. Concr. Inst. Trans., 1, xxi.
38. Anon., Amendments to the London Building Acts (London County Council General
Powers Bill, session 1909). Concr. Inst. Trans., 1, xii.
39. Concr. Inst. Trans., 1, xxv–xxvii.
40. Concr. Inst. Trans., 3, 220–21.
41. Concr. Inst. Trans., 5, 409; Concr. Inst. Trans., 6, 9; Concr. Inst. Trans., 6, 156.
42. Concr. Inst. Trans., 6, 10.
43. Concr. Inst. Trans., 6, 9–10.
44. Concr. Inst. Trans., 6, 156.
45. Regulations made under the provision of section 23 of the London County Council
(General Powers) Act 1909 with respect to the construction of buildings wholly or partly
of reinforced concrete. LCC: London, 1915.
46. Andrews, E.S., The Reinforced Concrete Regulations of the London County Council,
July 1915 (under the London County Council General Powers Act 1909 and now in
force: a handy guide containing the full text, with explanatory notes, diagrams and
worked examples). Batsford: London, 1916.
47. Anon., The report of the Concrete Institute on standard notation. Concr. Constr. Eng.,
1910, 5, 2.
48. Anon., Standard notation for reinforced concrete. Concr. Inst. Trans., 2, xii.
49. RIBA Joint Committee on Reinforced Concrete: Second Report. RIBA: London, 1911.
50. Concr. Inst. Trans., 4, 269.
51. Concrete Institute. Mnemonic notation for engineering formulae: Report of the Sci-
ence Committee of the Concrete Institute with explanatory notes by E.F. Etchells.
Etchells Concrete Institute: London: Spon, 1918.
52. Concr. Inst. Trans., 9, Dvii.
53. Concr. Inst. Trans., 9, Miv.
54. Concr. Inst. Trans., 6, 163.
55. Concr. Inst. Trans., 7, 162.
56. Concr. Inst. Trans., 9, Dv.
57. Hurst, B.L., Concrete and the structural use of cements in England before 1890. Proc.
Instn Civ. Engrs Structs & Bldgs, 1996, 116, 283–95.
58. Bussell, M.N., The development of reinforced concrete: design theory and practice.
Proc. Instn Civ. Engrs Structs & Bldgs, 1996, 116, 317–34.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:28:19
15.qxd 15/11/2001 15:59 Page 315

15 Concrete in tunnels

Alan Muir Wood

Introduction Concrete in tunnels serves usually in a simple functional capacity, as an arch or con-
tinuous ring providing support to the ground, acting in simple compression. So
far as is practicable, reinforcement is avoided, since it complicates the process of
emplacement of concrete in situ and may need special protection against corrosion,
especially where the concrete is exposed to aggressive water under pressure.
Generally, aesthetic attractions of concrete in a tunnel would be otiose where
the lining is hidden from view or only passed in semi-darkness at high speed. The
exception to this rule is predominantly for underground railway (predominantly
metro) stations or caverns for recreational or similar use. The severe geomet-
rical artistry of features of sewer tunnels have a special fascination from their
hermetic nature, exploited for example by Carol Reed’s film of The Third Man.
The adoption of concrete for a tunnel lining needs to be considered essentially
as part of a system, intimately dependent on the overall logistics of the other
processes of the system: excavation, spoil removal, primary support, preparation,
transport and placing the concrete into position. A particularly direct example of
this feature of the system explains the timing of concrete as successor to masonry
(in this context including brickwork as well as natural stone) — or, particularly in
North America, the use of timber — as a ‘permanent’ lining. In consequence, this
brief account relates the developments in the use of concrete in tunnels to devel-
opments of other aspects of the total tunnelling process. Much of the development
is seen, on this account, to have occurred incrementally from simple beginnings,
as special plant was developed in response to the specific economic demands for
placing concrete in a tunnel.

Applications of Concrete may be used structurally in tunnels in three different forms:


concrete (i) as concrete placed in situ behind formwork (in the section Concrete placed
in situ), of which a sub-set entails the grouting of previously placed aggregate;
(ii) as linings composed of precast units, normally assembled in rings, forming
a complete, usually circular, ring or a partial arch (in the section Precast
segmental linings);
(iii) as sprayed concrete (shotcrete) which requires no formwork (in the section
Shotcrete).
The functional objectives of a concrete lining may satisfy one or more of these
requirements:
(a) to support the ground or, more precisely, to provide an adequate degree of
support to allow the ground to support itself;
(b) to exclude water or other fluid or gas in the ground;
(c) to contain water or other fluid or gas under pressure;
(d) to provide a smooth intrados to the tunnel, especially to reduce resistance to
the flow of water or gas.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:29:01
15.qxd 15/11/2001 15:59 Page 316

316 Historic concrete

The lining may serve secondary purposes, for example for the attachment of
an internal facing or for support of service pipes, cables or equipment.

Concrete placed The traditional form of primary support for tunnels entailed heavy timbering
in situ developed on different principles in different geographical areas. The forms were
appropriate to the conditions for tunnelling and installed in accordance with
traditional rules of the pattern of the system. All such systems required the
partial or full removal of the timbering to be undertaken in short lengths as the
secondary support (i.e. the permanent lining) was advanced, either in headings
occupying a limited fraction of the cross-sectional area of the tunnel or for the
full-area section, the choice depending on the system of timbering and upon the
nature and special problems presented by the ground. A masonry (in this
text including brickwork) lining was suitable for such a system; concrete placed
in situ was not, since the work had to be undertaken in short and possibly
variable lengths and sections. Success often depended on early use of the lining
for support.
In consequence, the earliest uses of concrete placed behind formwork occurred
either where the ground required virtually no support (e.g. where protection
against rock falls in service remained a consideration) or where steel supports,
initially in the form of arches packed off the rock surface, had replaced timbering
as primary support. These examples are therefore to be found where competent
relatively unfractured rock coexists with areas of population growth or early indus-
trial development.
Early examples of concrete linings included tunnels of the early 1880s on the
now disused Glenfarg line on the North British Railway between Kinross and Perth,
for which W.R. Galbraith was responsible,1,2 constructed contemporaneously with
the Forth (rail) Bridge. Concrete on this occasion was so novel a material to the
Government (railway) Inspectorate that holes needed to be made for the purposes
of their examination, who pronounced the work to be ‘infinitely better than brick-
work’.3 Another example in Scotland, demonstrating that absence of timbered
tunnel support hastened progress, was in the provision of concrete lining to the
previously unlined sections of the 30-km length of the water supply rock tunnels
to Glasgow from Loch Katrine in 1886–88, for the purpose of increasing flow (by
35% according to Simms3). The concrete for tunnels at this time was hand-mixed
using as aggregate stone broken on site and local sand.
For the Perkasie and Muscanetong Tunnels of the Philadelphia and Reading
Railroad company in 1886–88, concrete sections of arch lining were cast in sec-
tions 20–75-ft long on centring placed and removed by the railroad company4,5
(Figure 15.1).
There are a number of accounts from North America of the use of in-situ con-
crete for water and rail tunnels in the period 1890–1900. One of these was the
Cascade Tunnel for the Great Northern Railway in Columbia (1891–92) for which
the arch was cast in 12-ft sections after side-walls had been built in concrete shovel-
led from a platform, at the level of the wall-plate, built in 500-ft lengths (Stevens25),
above the level of the muck trains.
For the Tremont Street Tunnel for the Boston Subway (1900), the side-walls
were constructed in rectangular drifts, with the arch constructed behind a roof
shield travelling on track along the 2-ft 9-in wide side-walls, the completed tun-
nel being 20-ft 5-in high  23-ft 3-in wide. This is probably the first instance of
an in-situ concrete lining being placed behind a shield. The concrete was placed
manually into the crown along curved cast-iron troughs and contained by timber
stop-ends. Vertical pipes were cast in for the purpose of filling the crown with grout
after concreting. The 16 thrust rams of the shield each shoved against a 3.25-in
dia. iron ‘push bar’ set end to end against previous push-bars cast into the

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:29:01
15.qxd 15/11/2001 15:59 Page 317

Concrete in tunnels 317

Figure 15.1 Concrete


placed in situ c.1890.

concrete lining (see also Bonnin23). A similar system had been attempted for the
Siphon de l’Oise in Paris but had to be abandoned.6
Lee7 describes concrete placed for the double-track Peakshill Tunnel, using tim-
ber side forms and arch forms in 12-ft lengths, with the final key in 4-ft lengths,
the concrete placed by shovels and tampers, with external waterproofing in tar-
paper treated with hot tar (elsewhere at this time bitumen felt was being used for
this purpose).
As steel arch supports came into more general use at this period, concrete arch
formwork initially comprised timber laggings on steel arches themselves slung from
the tunnel supports. An early use of travelling formwork is described for the Gal-
litzin Tunnel, Harrisburg, Penn. The roof was timbered with the arch lined behind
a 20-ft long travelling shutter of timber laggings on steel frames, the shutter trav-
elling on rails and lowered 9-in when being advanced.8
During the next 25 years or so, collapsible steel formwork, hinged to each side
of the crown, was being developed by the Blaw Steel Centering Co. and others,
initially requiring diagonal tendons to hold the formwork to shape during con-
creting. During the same period developments in all aspects of batching, mixing,
transporting and placing concrete achieved great economies and accompanied a
general acceptance of the superiority of concrete as a tunnel lining material.9
Concrete was being conveyed down vertical pipes from the ground surface with
provisions to prevent segregation, and satisfactory pneumatic placers were devel-
oped after years of trials. Subsequently, collapsible forms allowed the placing
of continuous lengths of arch lining (Figure 15.2), while slip-forming was later
exploited for the tunnel invert.10
Subsequent developments of note included the use of plasticizers and water-
proofing agents, developed particularly by Kaspar-Winkler, later Sika, from around
1920,11 with the earliest uses of mortars to control leakages of the Alpine Tunnels.12
The most celebrated use of a tunnel lined by prepacked aggregate subsequently
grouted is that of the 11-ft dia. Kemano penstock tunnel for Alcan’s Kitimat

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:29:01
15.qxd 15/11/2001 15:59 Page 318

318 Historic concrete

Figure 15.2 The develop-


ment of collapsible formwork
between 1929 and the
present day (courtesy of CIFA,
Milan).

hydropower project, in British Columbia, to ensure that a high proportion of inter-


nal pressure (up to about 2400 ft head) would be transmitted from a 5-in thick steel
lining to the surrounding rock (the proportion measured as around 75%). The
aggregate was placed by ‘rock-blower’, the grout pumped over horizontal distances
up to 3000 ft and up to 1500 ft vertically. The project was completed in 1954.
Jaeger13 describes several experiences in prestressing pressure tunnels, by exter-
nal grouting and by the use of circumferential prestressing cables, of the concrete
surrounding a steel lining, together with accounts of several failed attempts.
One particular development of in-situ concrete concerns the placing of concrete
immediately behind a tunnel boring machine (TBM), of which the Boston Sub-
way described above, was an early forerunner. The Press Beton TBM was based
on successful experiences in constructing metro tunnels for the Moscow metro in
1972 (Roisin — personal account). The Press Beton TBM would excavate the
ground by means of a rotating cutter-head (Figure 15.3) with the thrust devel-
oped against two sets of hydraulic rams, one set bearing directly against special
shoes which compacted the concrete lining, the other set against (say) 300-mm
wide rings of flanged steel segments providing the internal shutter for the newly
placed concrete. As the lining advanced, so was the ‘trailing’ ring — of (say) six

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:29:01
15.qxd 15/11/2001 15:59 Page 319

Concrete in tunnels 319

Figure 15.3 Concrete in situ


behind shield (from Brevet
d’Invention [patent applica-
tion] of Bade, CFE, Ed.
François & Fils, Frankignoul
29 Jan 1976).

rings of formwork — progressively dismantled and re-erected at the leading posi-


tion. The system was developed and adapted by Hochtief in Hamburg in 1978
using steel-fibre reinforcement for the concrete, and later for a metro tunnel in
Lyon.

Precast segmental One of the earliest systems of segmental lining of a complete ring was that intro-
linings duced by McAlpine14 in 1903, first used for the City of Glasgow Main Drainage.
The rings of concrete segments had tongue-and-groove circumferential joints, in
each of which was embedded a steel bar hoop as reinforcement. Such linings were
then used in different sizes for many years.
A widely used form of precast concrete lining, resulting from the general
progress in the manufacture of precast concrete, comprised a series of bolted
flanged segments, following the general form of cast-iron segments, the traditional
lining for shield-driven tunnels since Barlow’s Tower Hill Subway of 1869.15
Reinforced concrete segments were used for a New York subway in 1930 but
their wider adoption followed from the 2.75 mile extension of the Central Line
of London’s Underground in 1936.16 Linings of this type were most commonly
erected within the protection of the skirt of a shield with the annular void filled
with cement grout after each advance of the shield.
Speed, economy and control of ground settlement were much enhanced by the
development of linings in concrete segments directly expanded against the ground.
These linings have been developed for shield-driven tunnels in stiff clay, particu-
larly the London clay. The Don-Seg lining (Figure 15.4) was first used for the Lee
valley — Hampton water tunnels, following a successful use in an experimental
tunnel in 1951.17 This lining, of 21-in rings of 10 identical tapered segments, was
expanded by thrusting alternate segments into place by the shove rams of the
shield. Subsequent linings of similar type but using fewer tapered segments
are less readily adaptable than the Don-Seg to slight variations in the cut profile
of the ground (on account of corrections for line, for curves or for increased
‘coming-on’ of the ground prior to building the lining).15 For tunnels of 8–10 m
dia, the preference has been, on account of ratio of diameter to ring width, to
expand the ring by means of jacks inserted in recesses between selected segments,

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:29:01
15.qxd 15/11/2001 15:59 Page 320

320 Historic concrete

Figure 15.4 Don-Seg.

especially high economy being achieved for the Cargo Tunnel at Heathrow Air-
port in 196822 illustrated by Figure 15.5. To ensure no significant imposed bend-
ing stresses in the lining, longitudinal joints between segments were shaped convex
— convex, requiring high precision (0.2 mm) in casting and in overall dimen-
sional controls.
There have been many subsequent developments in concrete linings, expanded
and bolted, with many variants of attachment. The current trend is towards flush-
faced segments and for larger segments than for the earlier tunnels to take advan-
tage of mechanization. Segments therefore require reinforcement, with possible
need for protection against corrosion. Steel-fibre reinforcement has also been used
for precast segments. Sealing of the joints between segments has prompted major
developments of sealants. The most widely used form of seal has been the shaped
neoprene gasket, originating from the Phoenix gasket in Germany in 1974, which
girdles each segment and is sealed by pressure between segments, applied by
bolting or by TBM rams. An alternative form uses a neoprene with hydrophilic
additive which causes the seal to expand on contact with water.

Shotcrete The earliest known major use of mortar linings for tunnelling was for the treat-
ment of the arch of Alpine rail tunnels, preventing dripping of water onto the track,

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:29:01
15.qxd 15/11/2001 15:59 Page 321

Concrete in tunnels 321

Figure 15.5 Heathrow Cargo


Tunnel.

at the time of electrification, from 1917 onwards, the thickness, 75–120 mm, depend-
ing on the structural soundness of the original masonry lining, with additives pro-
moting low permeability and high resistance against sulphates and other salts.12
The early history of shotcrete is described by Bussell in this volume.
Rabcewicz18 may well claim precedence in setting out in 1944 the virtues of a thin
deformable concrete lining in achieving an economic stable rock/lining compos-
ite structure. This could only be achieved however with the advent of shotcrete
which was used for the first time for the Lodano-Losogno tunnel in Switzerland
for the Maggia hydro-power scheme (1951–55). Since that time there have been
many developments in Informal Support of this nature, in many types of ground,
including Sprayed Concrete Lined (SCL) tunnels in clay.15 The term ‘Informal
Support’ denotes the freedom to modify the elements of the system in relation to
need and to geometrical variation of the lined cavity. The techniques include the
New Austrian Tunnelling Method (NATM) but this term is avoided on account of
the extraordinarily conflicting definitions of this form.
The principal objective of a shotcrete lining is to create a rapid form of support
adequate to contribute to a stable ground/lining composite structure but sufficiently
flexible to tolerate a calculated degree of deformation as ground movements,
inevitable with the advance of the tunnel, occur. For the Snowy Mountains hydro-
electric project, pneumatically applied mortar (with sand up to 4.7 mm size) with
steel reinforcement was used from 1961.19 The more general use of shotcrete as
a primary form of ground support was developing in several European countries
in the 1960s, usually with mesh and rock-bolts prior to an in-situ concrete lining.
The concrete also served to provide a relatively smooth base to which to attach
protective ‘fleece’ and a waterproof membrane outside the secondary concrete lin-
ing, as a protection against inflow particularly where icing might otherwise occur.
The development of mixing and placing equipment has kept pace with the
market demand. For many years the debate has proceeded between the relative
merits of the ‘dry’ and the ‘wet’ processes. In the latter, the concrete is mixed and
pumped to the nozzle. Dry mix is impelled pneumatically in an earth-dry state to
the nozzle, where the water is added. The general consensus20 is now that
wet-mix is preferred for reasons of:

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:29:01
15.qxd 15/11/2001 15:59 Page 322

322 Historic concrete

• economy — capacity and rate of application;


• working environment — control of dust and reduction of ‘rebound’ (i.e. concrete
which fails to adhere);
• quality — while average strength may be less than dry-mix, quality is less variable.

Following recent developments, shotcrete suitable for tunnelling normally contains


accelerators and microsilica, having qualities of good adhesion, high early strength
and ductility. It often contains steel-fibre reinforcement applied with the mix.
Application over wide areas is undertaken by remote control or robotics to improve
working conditions for operatives.

References 1. Popplewell, L., A Gazetteer of the Railway Contractors and Engineers of Scotland
(1871–1914), Vol. 2, 1989. Published privately ISBN 0 906637 14 7.
2. Rickard, P., Tunnels on the Dore and Chinley Railway. Min. Proc. Instn Civ. Engrs,
1894, 116, 115–75.
3. Simms, F.W., In: D. Kinnear Clark ed., Practical Tunnelling, 4th edn. Crosby Lock-
wood and Son, 1896.
4. Ford, P.D., Discussion on Fitzgerald, 1894.
5. Fitzgerald, D., Lining a waterworks tunnel with concrete. Trans. ASCE, 1894, 31,
294–328.
6. Hewitt, B.N.M., Johannesson, S., Shield and Compressed Air Tunnelling. McGraw-
Hill, 1922.
7. Lee, G.W., The cost of concrete tunnel lining and of tunnel excavation. Eng. News,
1903, 50, 531–32.
8. Anon., Construction work on the Pennsylvania Railroad between Harrisburg and
Gallitzin. Eng. News, 1903, 50, 273–76.
9. Fitzgerald, J.H., Tunnelling Equipment — VI. Evolution of the concrete lining plant,
Eng. News. Rec., 1931, 9, April, 616–19.
10. Kidd, B.C., The Orange-Fish tunnel in South Africa. Proc. Sth Wales Instn Engrs, 1977,
93, 27–33.
11. Hegnauer, H., Notes from the archives of Kaspar Winkler & Co from 1920–1930. Sika
Promotion, 1984, 1 Zurich, 64–71.
12. Streuli, R., Tunneldichtung mit Sika an der Gotthardlinie, Sika-Nachrichten, 1943, 12,
1–8.
13. Jaeger, C., Present trends in the design of large pressure tunnels and shafts for under-
ground hydro-electric power stations. Proc. Instn Civ. Engrs, 1955, 4(1), 116–200.
14. Anon., The McAlpine system of reinforced concrete tunnel lining. McAlpine, 1935.
15. Muir Wood, A.M., Tunnelling: Management by Design. E. & F.N. Spon, London, 2000.
16. Groves, G.L., Tunnel linings with special reference to a new form of reinforced con-
crete lining. J. Instn Civ. Engrs, 1943, 20, 29–41.
17. Scott, P.A., A 75-inch diameter water main in London: a new method of tunnelling in
London clay. Proc. Instn Civ. Engrs, 1952, 1(1), 302–317.
18. Rabcewicz, L.v., The New Austrian Tunnelling Method. Water Power, 1964, 16, 453–57,
511–514; 17, 19–24.
19. Moye, D.G., Unstable rock and its treatment in underground works in the Snowy Moun-
tains scheme. Proc. 8th Commonwealth Min. Metall. Congress, 1965, 6, 429–44.
20. Anon., Shotcrete in tunnelling. Status Report. International Tunnelling Association:
Lyon, 1991.
21. Anon., Methods of work on the East Boston Tunnel extension of the Boston Subway.
Eng. News, 1902, 47(4), 74–76.
22. Muir Wood, A.M. and Cribb, F.R. Design and construction of the cargo tunnell at
Heathrow Airport. ICE Proc., 1971, 48, 11–34; 50, 187–201.
23. Bonnin, R., The new roof shield of the Metropolitan Railway Tunnel of Paris. Eng.
News, 1905, 54, 324–25.
24. See Chapter 5.
25. Stevens, J.F., The Cascade Tunnel, Great Northern Railway. Eng. News, 1901, 45,
23–26.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:29:01
16.qxd 15/11/2001 16:00 Page 323

16 Water-retaining structures in Britain


before 1920

Michael Gould
General The construction of water-retaining structures, sewers, reservoirs, sewage tanks,
and swimming baths, as well as bases for gasholders, followed a definite sequence:
development from brick, usually1 with a clay puddle backing, through brick with mass concrete
backing, then all mass concrete to reinforced concrete. However, development was
not sequential, with some engineers still building brick structures when others were
using concrete.
Experience revealed that brick, generally of local manufacture or sometimes
produced on site, was frequently porous, and that ordinary lime mortar was both
porous and susceptible to dissolution in water. From the 1850s, concrete began
to be employed as backing on economic grounds, particularly with the spectacu-
lar strength being shown in tests of Portland cement concrete.2
The use of ‘composite’ walls, as they were described, became more general in
the late 1860s, led, e.g. by designers such as Vivian Wyatt for the Chartered Gas
Co. of London, but concern about the porosity of concrete meant that a puddle
cradle continued to be employed. Puddle was also felt to have advantages in areas
of poor ground to accommodate movement. Asphalt lining was used in porous
new red sandstone at Liverpool gas works (1874) and as a seal between two lay-
ers of concrete ‘wall’, in Sheffield (c. 1899), to deal with movement from mining
settlement.
As the puddle was an integral part of the construction, these tanks could not
be tested for watertightness without the backfill, and the walls were designed for
retaining only the soil.
Concrete was also employed for the base of tanks, although in areas of imper-
meable strata such as London clay, a base was sometimes dispensed with entirely.
Some brick gasholder tanks were strengthened by concentric rings of hoop iron,
and this was also used on some concrete tanks. This practice was advocated by
V. Wyatt who:

‘considered that hoop iron was most essential in concrete walls. He found
it had a tremendous affinity for cement; it did not rust; it never swelled,
but took hold with the greatest avidity, and it was difficult to separate
the two. They formed as it were one homogeneous mass from top to
bottom’, 3

although he also puddled the floor and walls.


It was recognized that large lengths of mass concrete would crack and it was rec-
ommended that walls, e.g. should be subdivided by vertical divisions. However, the
only joint treatment available initially was to fill the gap formed with strong cement.4
Problems with joints continued for some years. Lansdown water tower, Bath, built
by Coignet in 1926, holds 100,000 gallons at 65 ft. Although still in use, this gave
trouble from rusting at the joints and the tank was Gunite coated in 1932–34.5
To offset such problems some tanks were lined, the low 40,000 gallon tower at
Ham Green Hospital, Bristol, also by Coignet, was, e.g. rendered with Pudlo.6

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:29:49
16.qxd 15/11/2001 16:00 Page 324

324 Historic concrete

It had been suggested that watertight concrete required cement sufficient to fill
the interstices, although a stronger surface render was also advocated.7 Later, other
approaches were used.8,9 One was the use of concrete additives, such as soft
soap,10,11 oil,12 clay13 or commercial preparations such as Ceresit.14 An alternative
was a waterproof surface layer, e.g. asphalt; or coating, e.g. sodium silicate.15 Later,
it was realized that many additives had no effect and that dense concrete would
be watertight16 if properly made with correctly graded aggregate.17
It was not until 1938 that the Institution of Civil Engineers, in consultation with
the other Institutions, produced a code of practice for water-retaining structures.18
This advocated the limitation of direct tensile stress and bay sizes, as well as
the use of alternate bay casting, to prevent cracking. A minimum percentage of
steel was proposed, although it was then permissible to put this all into the ten-
sion face.
The attitude of the Local Government Board (LGB), which sanctioned loans for
public works, delayed the change from mass to reinforced concrete in the years
after 1900. The LGB considered that reinforced concrete was less durable and loans
were given on a 15-year repayment period, against 30 years for mass concrete.19
For example, the high level reservoir at Exmouth had concrete walls lined with
blue brick and a roof 6 in thick, in 9 ft 9 in bays between steel joists, reinforced by
expanded metal.20 This, the LGB suggested, was not reinforced concrete and a
30-year repayment period was allowed. In 1916, the reservoir at Wath upon
Dearne was built with mass concrete/brick walls, with a 12 in clay cradle. Here,
the loan on the reinforced roof was restricted to 10 years.21 In a time of low
interest rates, this represented a considerable increase in the rate burden.

Sewers Apart from cement mortars used for brick sewers, concrete was first used in
sewerage as a backing. Oxford sewerage in 1876, used brick sewers with a square
concrete surround, flanged cast iron pipes (with concrete and iron filings joint filler)
or stoneware with Stanford patent joints, the barrel being built up to a sliding fit.22
Similarly, Croydon used earthenware, some with 3 in concrete surround; one ring
of brick with a 1 in cement waterproofing collar, and surround; or two rings of brick
with a cement collar between.23 It was suggested that Portland cement was better
than lime for resisting chemical action of sewage, but that an internal ring of
brick was needed to protect the concrete backing.24 Given the ‘two-ring’ brick
construction used at Croydon with no backing, it is now unclear whether it was
the one brick ring or the backing of the composite sewer which was considered
as providing structural strength.
Concrete pipes had been used in USA since 1845, and some experiments were
carried out in England by W. Buckwell in 1849. It was not until 187525 that Sharp,
Jones & Co. started making ‘rock concrete tubes’, using broken stoneware aggre-
gate. These were used at Bournemouth (where some cracked longitudinally ‘in
course of ordinary engineering precautions having been neglected’), Bromley and
Frome.26
Elsewhere, in 1881, Peterborough Corporation laid 2 miles of ‘concrete carrier’27
to take settled sewage to the sewage farm.28 Newhaven used fine cement concrete
for cast in-situ sewers in 1887, in either 6 ft by 6 ft semi-circular sections with a 1 ft
6 in cunette, or 4 ft diameter pipes with a brick invert.29
In 1891, the Monier system for pipes reinforced by mesh was reported.30
However, it was considered only suitable for large sewers due to difficulty in form-
ing the bore.31 This was partially overcome by the Bonna system, promoted by
the Columbian Fireproofing Co. of Acton. A steel core-barrel was reinforced exter-
nally with longitudinal and a helix of plus-section bars before concreting. These
two elements provided the necessary strength. Reinforced concrete was cast inside
the barrel, but only as a protection. Collar joints were used (Figure 16.1).

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:29:49
16.qxd 15/11/2001 16:00 Page 325

Water-retaining structures in Britain before 1920 325

Collar

Outer layer
reinforced concrete Grout

Core
barrel
Figure 16.1 Bonna pipe
Bitumen
joint. (Inner layer r.c. not shown)

Figure 16.2 Pipes at


Swansea (C&CE Photo).

Bonna pipes, cast vertically (Figure 16.2) were used for water mains in Swansea,32
for a sewer in Acton33 (Figure 16.3) and for a sewage pumping main (117 ft head)
in Norwich.34 Thereafter, reinforced concrete was increasingly used for pipes;
Professor Matthews reviewed progress in 1919.35 Later, small diameter pipes
became a reality with centrifugal spinning.

Culverts Reinforced concrete proved suitable for culverts of various cross sections
(Figure 16.4). Culverts at Birmingham and West Hartlepool were designed by the
Expanded Metal Co.,36 that at Birmingham was above ground and was surrounded
by stonework. Kilton was a Trussed Steel Concrete Co. design,37 while Ouseburn38
and Belfast39 were in Hennebique Ferro-concrete. The outfall culvert at Belfast
(Figure 16.5) was on timber piles. Work was suspended by the Ministry of Muni-
tions in 1917 and not completed until the 1930s.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:29:49
16.qxd 15/11/2001 16:00 Page 326

326 Historic concrete

Figure 16.3 Pipes at Acton.

Patent ‘Aqueduct’ blocks incorporating steel-ribbed granite-concrete pipes were


used at Great Crosby.40

Reservoirs and Many early water supplies consisted of a well or river intake and a steam driven
pump. This worked best with a constant flow against a constant head and a reser-
tanks voir was often constructed as a pump-header; from these, the modern service
reservoir developed. Such headers were open with sloping sides finishing at ground
level. However, in 1852, the (London) Metropolis Water Act required that all
filtered water reservoirs within 5 mi of St Paul’s Cathedral had to be covered, and
this provision gradually spread throughout the country. The slowness of
this process is shown by the fact that some early Mouchel designs related to the
covering of sloping sided reservoirs, e.g. Cowes in 1909,41 while Camborne, built
by the Expanded Metal Co. in 1910, was uncovered.42
Early reservoirs were covered by brick (usually barrel) vaulting or, by brick jack-
arching. Spans necessitated brick supporting pillars, although some jack-arching
was carried on cast iron columns. Sloping sides were a hindrance when covering

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:29:49
16.qxd 15/11/2001 16:00 Page 327

Water-retaining structures in Britain before 1920 327

3' dia. steel ribbed pipe

Granite
concrete
block

3'11"
6'11"

(a) Great Crosby (b) West Hartlepool


8"
18"

20'
14'

14"
13' 30'
2'

"
4'9
42'
Figure 16.4 Reinforced
concrete culvert sections. (c) Kilton (d) Ouseburn

5
8 @ 6"
3
8 @ 6"

1
/2" @ 6" 1 in
8 in

1
/2 in

Strap
4 in

3
8 " @ 6"

Timber pile

Figure 16.5 Belfast outfall


culvert corner detail.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:29:49
16.qxd 15/11/2001 16:00 Page 328

328 Historic concrete

reservoirs and vertical sides were introduced, although a tank in Wolverhampton


(1851) had an odd horse-shoe cross-section.43 Vertical sides were of brick (occa-
sionally stone) backed by clay puddle, brick backed by concrete, or, later, mass
concrete, sometimes also backed by puddle.
Reinforced concrete was used from 1900; that much built was not publicised
is a problem when discussing the history of water supply. In some reports the
specialist designer for the reinforcement may not be credited, others partially
so (Skegness reservoir, c. 1910, used ‘Columbian’ bars and is credited to Percy
Griffith, perhaps employed by Industrial Constructions Ltd).44 Two review papers
on reservoirs were published in 188345 and in 1915,46 and much of what follows
is taken from them.
As mass concrete became acceptable, whether brick or concrete was used
depended largely on material availability. Thus Sudbury, c. 1871, 55 ft × 55 ft, was
of brick backed with concrete while Kettering, c. 1873, of similar size, was built of
stone with clay puddle.47 Kettering had a stepped floor (Figure 16.6).
A problem with barrel arching was the horizontal thrust at the wall; with jack-
arching this is resisted by tie bars between the girders. In 1874 at New Cross, the
wall of the 1.75 million gallon reservoir (brick backed by concrete with brick arches
on brick piers) was seen to be rotating and an arch invert was added. Other tanks
were built with floors arched between the lines of columns. Elsewhere, the out-
ermost of the roof arches was carried round to meet the wall lower down, the thrust
now being vertical. Woolwich Common, c. 1870, had the brick/concrete
end-wall built as a series of vertical axis arches to brick piers at the apexes, although
it is not recorded why this arrangement was used
Although most early reservoirs were rectilinear, circular examples are recorded.
Greenwich Park, built as an open reservoir in 1845 with sloping sides and base
diameter 154 ft, was lined with 12 in lime cement on 9 in puddle. This did not stand
well and a brick lining was added. In 1871, it was covered by concentric barrel
arches, the outer being carried to a point half way down the side slope.
As experience with concrete grew, its use became more general. Deptford reser-
voir, built in 1872 by covering a filter bed, used the bed materials to form
concrete roof arches, tapering from 20 to 10 in in thickness, carried on longitu-
dinal brick arches. At Oldham in 1910, a complete reservoir in this arrangement
was built in reinforced concrete.48 Addington (Croydon) was mass concrete, includ-
ing the roof,49 but the discussion suggests that this was not the first to be made
thus. Fifteen vertical cracks in the walls had to be cut out and refilled with
concrete. The construction in 188650 of the filters for Carlisle water-works in mass
concrete passed with almost no comment (although as these tapered from 4 to 2.5 ft
in a height of about 5 ft the design was very safe).
Some hybrid constructions were undertaken. Upleadon (Gloucester, c. 1900)51
had roughly dressed masonry walls rendered internally (7 ft tapering to 2 ft 6 in
over 18 ft), a floor of 1 ft 3 in of concrete and a concrete arch roof carried on steel
girders, the arches tapering from 3 ft to 1 ft 6 in over the 12 ft 4 in span; the rise
was 2.5 ft. The girders had brick piers with three courses of footing on the floor.
In Kent, a number of late 19th century circular concrete tanks were reinforced
with three or four 1.5 in wide hoop-iron bands every foot (Figure 16.6). Initially
these were horizontal (Knockholt, c. 1889, West Wickham and Eltham) but at
Sundridge Park they were placed vertically. Southfleet (built after 1904) was
similar but was the first of the design to omit a brick wall lining.
These tanks were not without problems. Eltham cracked at the wall/floor joint,
the horizontal floor steel not being tied into the walls. Subsequently, a bitumen
sheet was incorporated at this joint. Southfleet’s concrete walls cracked, apparently
due to the expansion of the steel roof joints52 which were built rigid into the wall.
The construction of tanks for sewage paralleled those built for water supply,
although the use of all brick construction seems to have continued for longer.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:29:49
16.qxd 15/11/2001 16:00 Page 329

Water-retaining structures in Britain before 1920 329

Tie

Concrete

(a) Plumstead 1854 (b) Sudbury c. 1871

Puddle

Brick

Brick
Concrete
Puddle Concrete

(c) Kettering c. 1873 (d) Farnborough (Kent) 1879

I beam on C.I.
columns
Bow plate Cross-tied
girder by I beams

(e) Burton on Trent c. 1880 (f) Nottingham 1885

Hoop
iron

Figure 16.6 Reservoir cross


sections. (g) St Helens 1893 (h) Eltham

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:29:49
16.qxd 15/11/2001 16:00 Page 330

330 Historic concrete

In 1884,53 e.g. a 3.25 acre holding tank54 was constructed at Portsmouth, built in
brick with cement backing and ‘a straight thick water-tight joint between’ (i.e. of
stronger cement).

Water towers In 1900, Mouchel55,56 commenced design of what is accepted as the first water
tower in reinforced concrete in Britain. Completed c. 1903, this small structure
was for green watering at Meyrick Park golf course, Bournemouth.57 Thereafter
the number of concrete water towers built rose steadily.
Early design followed one of three patterns. First were tanks on solid shafts, of
several cross-sectional arrangements; this mirrored brick construction. Second were
plain tanks, some on frail looking legs. Some, such as the Mouchel tank at Red
Barn, Fareham (probably the oldest now surviving) (Figure 16.7), had no leg brac-
ing but most had one or more ring bracing or, more commonly, wheel bracing.
Bracing, normally horizontal, was occasionally vertical. The third design was of a
tank on a central shaft and an outer ring of columns, as Mouchels 300,000 gallon
tower at Newton-le-Willows (1905).
The tanks on these towers sometimes leaked, and many designers lined the tank
as a precaution. Newton tank cracked horizontally between the hoop steel when
filled. It was suggested that this was due to the sun’s action on one side58 but with
Marsh,59 a standard text of the time, suggesting that the wall spanned vertically
between the hoop steel, it seems likely that the possibility of vertical bending in
the wall was not catered for.
On small towers, the floor and roof simply spanned to the sides, although some
towers had a domed floor. Mouchel often used a grillage of beams under the floor,
clearly seen on the still standing tower at Trim, Ireland c. 1909.60
Expanded metal perhaps lent itself best to solid sided construction. Two orna-
mented circular examples stand in County Down (at Donaghadee and the Down
Asylum, dating from 1912 and c. 1910).61 This company also built square towers
as at Drumcondra, Dublin and Holsworth, Devon (Figure 16.8).62 Greenbank Gas
Works, Blackburn, had a shallow tank on six asymmetrical spaced legs to accom-
modate a loading conveyor for the coke hoppers.63

Figure 16.7 Red Barn water


tower (B. Otter Photo).

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:29:49
16.qxd 15/11/2001 16:00 Page 331

Water-retaining structures in Britain before 1920 331

Figure 16.8 Expanded metal


water towers. (a) Down
asylum c. 1910;
(b) Holsworthy;
(c) Drumcondra, Dublin. (a) (b) (c)

The patent Kahn bar was intended for beams and would appear to have little
benefit for towers. However, a number were described as ‘Kahn’. The earliest
noted, at Severus Hill in York (1914,64 Figure 16.9), built around a stand-pipe,
was described as the largest built to that date (300,000 with a maximum height of
103 ft; the height of Newton was 83 ft).
The Indented Bar Co. also built towers. Cleethorpe (1907–10) had a steel tank
set on a metal ring at a height of 110 ft to allow for expansion movements;
the Ferbeck Chimney Construction Co. built the solid sided shaft.65 It still
stands, although the 250,000 gallon tank was beyond repair by 1962 and was
replaced by a shorter concrete tank.66 A smaller solid sided tower, also for the
Great Grimsby Waterworks Co., was at Immingham. The company also built a
number of multi-legged towers with a central core in the 1920s, culminating in
Goole, being the largest in Europe in 1927 (750,000 gallons with an overall height
of 145 ft).67
BRC designed that for the West Cheshire Water Co. at Whitby, c. 1915 (Figure
16.10). It was hexagonal with a diagonal length of 57 ft.68 It was supported on thin
square 42 ft long legs with one wheel bracing mid-height. The floor extended
beyond the legs and was supported by down-stand beams. The pattern of attached
beams and columns to the walls suggests that they were designed as slabs span-
ning horizontally and perhaps also vertically.
Some railway companies built water towers. The North Eastern Railway built
at least two octagonal by Truscon69 and three circular by Mouchel. Mouchel
designed a rectangular tower for the North British Railway in 1908.70 The Lon-
don Brighton and South Coast Railway’s Kahn tank at Lancing was on an arched
circular beam over six 36 in by 9 in columns and curtain walls. The 6 in thick base
was domed over the 25 ft span (with a rise of 4 ft), while the roof was only 4 in
thick.71
Hospital towers were constructed at Warrington Union by BRC and at Aberdeen
Poorhouse by the Indented Bar Co., both built by 1916.72 The latter was square
and held 110,000 gallons in two compartments — Mouchel & Partners also
designed some square towers.
Some individuals commissioned private towers. The earliest designed by Mouchel,
were at Roydon, Essex and Gopsall Hall, Leicester (both c. 1909).73 A circular
solid-sided example stands at Ridgmont, Beds., a private supply provided by
the Duke of Bedford.74 Milton Hill, near Oxford was built to the Kahn system in
191675 (Figure 16.11, although there is some doubt as to how much, other than

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:29:49
16.qxd 15/11/2001 16:00 Page 332

332 Historic concrete

Sketch drawing prepared from old photograph.


Height 103 ft max., capacity 300,000 gallons. 1914, Kahn.

Ring brace

Ring brace
Ten outer
inclined legs

Wheel brace
Four
inner vertical legs

Ring brace

Ring brace

Figure 16.9 Severus Hill


water tower, York. (Note: rear legs not shown)

the two water tanks, one above the other, was in concrete). A small drum on tall
legs was provided by BRC at Brookmans Park76 while a square tower over a well
was built by the Indented Bar Co. at the Grange, Beaconsfield,77 both before 1916.
In 1914, Coignet built an elegant tower for Sir Oswold Mosley at Rolleston on
Dove; it stands but is unused. It was filled in 14 days and a ‘slight dampness’ at
the floor/wall joint disappeared after ‘a few days’.78

Swimming baths In 1846, the construction of municipal baths and wash houses was permitted,79
with set charges for bathing.80
By 1865,81 the Act had been adopted by only eight metropolitan parishes
(earliest St Marylebone, 7.11.1846) and 24 boroughs (earliest Birmingham,
7.10.1846). Birmingham provided three ‘swimming baths’ as the permitted ‘open
bathing places’.82
It was not until 187883 that the legislation was extended to ‘covered swimming
baths’. However, as this Act simply doubled charges set for classes 1, 2 and 3 and,
as no definition of these classes was included, it appears that it only regularized
the then current practice. The bath built by the Bury Improvement Commissioners
in 1864, e.g. included first and second class swimming baths, each 52 ft by 20 ft.84
At Burnley, two baths (72 ft by 37 ft, and 62 ft by 30 ft), were called ‘plunge baths’.85
Both sites appear to have been covered.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:29:49
16.qxd 15/11/2001 16:00 Page 333

Water-retaining structures in Britain before 1920 333

Figure 16.10 Whitby water


tower (BRC Brochure).

In their next Report, the LGB noted that they, not the Treasury, now approved
loans for baths. Surprisingly, they seem not to have issued any circular letter indi-
cating what was considered acceptable. Even so, in 1879, loans were given to Aston
Manor, Croydon and Nottingham Boroughs.
From the 1880s, although the construction of swimming baths became wide-
spread, details were rarely reported, more account being given of the facilities.
York baths, opened in 1880 and sized 70 ft by 25 ft, are simply reported86 as having
walls of white glazed brick and white glazed tiles on the floor — features common
to most baths. However, construction techniques for swimming pools probably
followed those for reservoirs. As late as 1907, the use of brickwork walls was being
advocated in preference to concrete, although ‘a good thickness of cement con-
crete’ in the floor was considered acceptable.87
Elsewhere, all concrete construction was used. Carlisle baths,88 1883–84, had
cement concrete walls 2 ft thick, with a floor 1 ft thick, and tiling. As discussed above,
the watertightness of such construction was considered suspect. At Ealing,89 hot
bitumen was run between the glazing brick and the wall, while the open air bath
at Tottenham90 was rendered with Limmer asphalt. An asphalt lining (presumably

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:29:49
16.qxd 15/11/2001 16:00 Page 334

334 Historic concrete

Figure 16.11 Milton Hill


water tower. (Kahn, Incorporated electricity generator, pump house and two water tanks in tower)

Baths
(men) Baths (women)

Yard

Wash Waiting
house room
Boiler room
In
(women)
Tickets
Open-air
swimmimg pool In
(men)

Wc
Wc

Covered pool

Figure 16.12 Sketch plan


South Norwood baths,
Croydon, c. 1888. (Changing boxes arranged around each pool)

between the concrete and glazing) was used for Battersea,91 said to be the largest
covered bath when built (150 ft by 50 ft). Croydon baths,92 1889 (Figure 16.12)
were constructed in concrete with ‘no clay puddle’, and tiles placed in neat cement.
As with tanks, the change to reinforced concrete was gradual. In 1907, two baths
(100 ft by 30 ft, and 96 ft by 27 ft) were built for Chelsea by the Expanded Metal
Co. The floor was supported on 24 in  18 in ground beams at 10 ft centres, the
wall and floor being 12 in thick, with an asphalt layer between the concrete
and glazing. A 100,000 gallon storage tank was incorporated below the public foot-
path (Figure 16.13).93 Mesh sheets were used with rolled steel joists (RSJs) encased
for beams and columns.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:29:49
16.qxd 15/11/2001 16:00 Page 335

Water-retaining structures in Britain before 1920 335

e
f
Reinforced by expanded metal
sheets and encased steel

7'3" 2'3" Bath (a) Two angles (3"◊ 3")


(× 16')
5'10" 30ft (b) Four angles (3"◊ 3")
a
(× c 6') (× 100ft) (c) 12"◊5" ◊ 35lb RSJ
d (d) 8"◊6"◊31lb RSJ
d a f (e) 7"◊ 3Ω"◊ 15 lb RSJ
10"
Tank (f) 6" ◊5"◊ 24 lb RSJ
holding

7"
18"
c
c 100,000g

d
b

Figure 16.13 Cross section


of Chelsea bath. d

Thinner sections were used by Mouchel and Partners for Dundee baths.94 Here,
the floor rested on 8 in  14 in or 6 in  12 in beams spanning to piles. The floor
was only 5 in thick and the walls 4–5 in. How watertight this pool was is not
recorded. Mouchel’s first ferro-concrete baths, dated to 1906,95 were for Roedean
school and for Goole UDC. Not all baths built were covered, Dundee, Northam
and Pollockshaws, e.g. being open. Northam, Southampton,96 replaced a mass con-
crete structure on the sea-shore. This had had insufficient weight and had cracked
due to tidal variation on uplift pressures.

Gasholder bases Early gasholders were of limited capacity and comprised of cast iron segments
bolted together. They were quick to erect and some outlay could be recovered by
selling them on to other, smaller concerns, or for scrap, when increasing demand
necessitated their replacement by larger containers. The use of cast iron tanks was
soon found to be uneconomic and excavated tanks lined with brick or stone walls
with stone flag, brick or tile base became the norm. The tank was filled with water
as a seal, and the gas stored in a cylindrical sheet iron vessel or bell, which floated
in the water, rising and falling as the demand for gas fluctuated. The movement
of the bell was controlled by side guides.
Although the excavation could be taken across the whole base, sometimes a
trench was excavated around the perimeter to accommodate the wall, and the exca-
vation completed with the wall in place. The centre could have been blasted out
in rock, or excavated, leaving a rise or dome in the centre. This had the advan-
tage of helping to support the bell when lowered but could, if carried too high,
affect the storage and could be difficult to seal. This form of construction was
by 1868 ‘not of a kind to be now recommended’97 but it continued to be used,
e.g. at Becton in 1890–91, see Figure 16.14.98 Brick facing was generally used,
increasingly on aesthetic grounds.
At Imperial Gas Works Bromley,99 the whole of the bottom of the tank was
covered by ‘well prepared’ blue lias lime concrete 12 in thick, of a 1 : 5 mix, although
here the foundation was of blue clay. The Phoenix Gas Works, Kennington,
used wall footings of Portland cement concrete, a 1 : 7 mix, placed in two courses

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:29:49
16.qxd 15/11/2001 16:00 Page 336

336 Historic concrete

(a) Birmingham (b) Redheugh, Newcastle

200 ft dia. 36 ft deep 152ft dia. 30ft deep


Pier
Stone

Support Brick
for 6 in
bell Concrete

30 in Puddle Puddle
12in

24 in

27 in 180 ft dia. 45 ft 6 in deep 86 ft


dia. Arch
Plan piers
Dry 194ft dia. 37ft deep
Pier 9in
12 in 1:25
2
30 in 1:
24in Puddle
Figure 16.14 Sketches of 60 in
gasholder tanks wall/floor
sections. (c) South Metropolitan gas works (d) Beckton

each 2 ft thick, but relied entirely on the blue clay as a floor seal. Concrete piers,
apparently unreinforced, ranging from 23 ft 6 in to 16 in and founded on the clay,
carried the bell. Similarly, at Salford, c. 1880, concrete 12 in thick was used only
in wall footings, the floor consisting of 2 ft of puddle. York stone was used for the
coping, the rest stones (to support the bell when lowered) and as inserts in the
brick to carry the guide rails. Birmingham, by contrast, had the brick wall founded
on the puddle but concrete 6 in thick (2 ft under the piers to support the bell) over
the floor puddle. The walls were reinforced with both hoop iron and bands of five
courses of brick laid in Portland cement. At Bow a double brick skin was employed
enclosing a concrete heart, the brick courses being carried across at intervals.
Finally, in 1871–72 George J. Livesey built a Portland concrete tank, 153 ft dia-
meter, 35 ft deep, with a brick lining 9 in thick, bonded into the concrete, the wall
and floor being puddled. The cost was stated as two-thirds that of an all brick
tank.100 This successful application encouraged other designers to make Portland
cement concrete tanks. At Redheugh, Newcastle, puddle concrete walls with
a single brick face and a concrete floor was used (although, oddly, the whole
wall had a brick foundation). The whole wall thickness was reinforced with hoop
iron. J. Douglas, engineer at Portsea, had information101 from tests at Portsmouth
Dockyard which indicated a 7-day strength of Portland cement concrete of
299.2 lb/in2, 30-day strength of 417.3 lb/in2, and 2-year strength of 675.8 lb/in2
(earlier Medina cements had given 150 lb/in2). Portsea tank was concequently built
up using small moulds and made with West Medina Mills cement supplied by
Charles Francis and Company. A 9 in brick facing, separate from the concrete, was
employed purely for aesthetic reasons.
Livesey, in 1875–76, made a new tank for the South Metropolitan Company
entirely of Portland cement concrete (1 : 7 mix) covered in a Portland cement
render 1⁄2 in thick and with no puddle or brick. It was 180 ft in diameter, 45 ft 6 in
deep, with the iron guides embedded in the concrete. It was founded in chalk
at a depth of 42 ft, with water bearing sand above. The walls were 5 ft thick taper-
ing to 2 ft 3 in. The tank was completed in 1875 and partially filled, showing no

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:29:49
16.qxd 15/11/2001 16:00 Page 337

Water-retaining structures in Britain before 1920 337

signs of leakage, but when, in December it was refilled with freezing canal water
vertical cracks appeared, causing concerns about the viability of all concrete tanks,
and their vulnerability to temperature shocks, although at the time it was realized
the ground conditions may have contributed. Livesey himself felt he should have
used hoop iron, although other engineers were sceptical of its value.
Hooped concrete walls were used at Waterford in Ireland where a tank had to be
founded on a peat bog of 12–25 ft overlying silty sand. A 9 ft wide perimeter trench
was excavated, a layer of planks was then laid. Onto this, a layer of 1 : 6 Portland
cement mortar incorporating larger pieces of schist was spread. The first layer was
3 ft deep with an angled skewback at the centre to receive the dome base. Three hoops
of 4 in by 1⁄2 in wrought iron were embedded in this layer to help resist thrust of the
dome. The concrete wall was 4 ft thick at the base and 20 in at the top, with hoops
at the guide supports. Over the winter, the walls were loaded with stone and there
was differential settlement where the peat was deepest. The tank was then modified.
The wall for the gasholder at East Greenwich (c. 1890) stood 21 ft above ground
and 13 ft into it. It was 3 ft 6 in thick at the top tapering to 4 ft 6 in just below ground,
it then stepped to 5 ft continued to the base. The wall above ground was embanked,
was not considered to be sufficiently strong in concrete and was reinforced with
hoop iron bands, 5 in  5⁄8 in riveted together at 2 ft centres. These were ‘strutted
outwards while filling the concrete around them’.102 If this was not released until
the concrete had set then it should have induced a modest degree of prestress.
The tank built by Wyatt at Beckton, concrete with a brick inner face, was unusual
in that the taper was taken inwards as providing better soil support. Many tanks
had outside piers giving added support, but at Beckton these were also inside,
and they were finished with flat brick arches to carry the coping.
Unsurprisingly reinforced concrete was employed in the 1890s by British gas
engineers for tanks, although the earliest example appears to have been the
Monier system tank at Fredericksberg Works, Copenhagen. Some engineers were
sceptical of the progressive nature of the design, however.103

Other structures The ship test tank at the National Physical Laboratory, Teddington104 (now British
Maritime Technology Ltd.), see Figure 16.15, was completed in 1911. Towing Tank

Figure 16.15 NPL Tank.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:29:49
16.qxd 15/11/2001 16:00 Page 338

338 Historic concrete

No. 1105 is 52 m long by 9.1 m wide, maximum water depth is 3.7 m. This is a large
U-shaped channel incorporating a circulating water pipe on one side, the concrete
varying in thickness from 600 mm to 1.2 m. Reference 104 implies that this was
built in mass in lengths of 40 ft (12.2 m), the floor first, followed by the walls shortly
after.
Concrete was also widely used in other structures subject to fluid pressure such
as basements. By 1917, reinforced concrete gas mains were being tested.106 Pipes
of 14 and 24 in diameter were cast using the Coignet system and were impreg-
nated with tar before installation.

Conclusion By 1920, and in spite of the conservatism of the LGB, reinforced concrete con-
struction was well established for use in water-retaining structures.

Appendix Below is given the number of contracts for different classes of work undertaken
by Mouchel and Partners up to 1919 (from Ref. 55):
• reservoirs 34 (of which 29 were public supplies):
— coverings, 15,
— lining, 1.
(the reservoirs included the freshwater ‘pipe reservoir’, 500 ft long by 4.5 ft
diameter, built thus due to difficulties over land acquisition);
• water towers 50 (of which only 17 were public supplies);
• sewage tanks 15;
• swimming baths 12 (6 public, 6 private);
• other tanks (a large miscellaneous group, probably mostly small units), 176,
include tanks for water and rain water, chemicals, oil and tar, for filtering and
settlement, and roof tanks.

References For simplicity, the later title, the Proceedings of the Institution of Municipal Engineers, has
been used throughout.
1. King, A., Treatise of coal gas. Reports gasholder bases in north west England built
entirely of brick, 1870, 2, 62.
2. Douglas, J., Gasholder tanks in concrete. Br. Assn Gas Eng., 1874, p. 34.
3. King, A., Treatise of coal gas. King expresses some doubt about the wisdom of using
hoop iron, 1870, 2, 32.
4. Newman, J., Notes on concrete. SPON, 1887, p. 76.
5. In RC in 1926 C&CE, 1927, 22. Repairs ibid., 1932, 27, 655 and 1934, 29, 453. Also
W&WE, 1926, 30, 15.
6. C&CE, 1916, 11, 47– 49.
7. Newman, J., Notes on concrete, SPON, 1887. Pages 46 and 86 resp. The first
probably had the same effect as the second, i.e., it produced a stronger concrete.
Chatham Dockyard used a 4 : 1 (although later figures suggest this should have read
6 : 1) render on a 12 : 1 mass.
8. Gadd, W.L., Waterproofing concrete. C&CE, 1908, 3, 154 –57.
9. Plumb, R.A., Waterproofing concrete. C&CE, 1910, 5, 497–501.
10. Memorandum. C&CE, 1911, 6, 320–1.
11. Grittner, A., Waterproof concrete. C&CE, 1912, 7, 810 –13.
12. Page, LW., Waterproofing qualities of oil mixed concrete. C&CE, 1913, 8, 619 – 25.
13. Waterproofing concrete. C&CE, 1907, 2, 490.
14. Industrial notes. C&CE, 1910, 5, 451.
15. Memorandum. C&CE, 1910, 5, 933.
16. Memorandum. C&CE, 1912, 7, 865 & 932.
17. Waterproofing concrete. C&CE, 1911, 6, 7–9. A wide range of additives were tested.
18. Code of practice for the Design and Construction of Reinforced-Concrete Structures
for the Storage of Liquids. Institution of Civil Engineers: London, 1938.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:29:49
16.qxd 15/11/2001 16:00 Page 339

Water-retaining structures in Britain before 1920 339

19. This subject appears throughout. C&CE, 1907–1908, 3– 4.


20. Proc. Instn Munic. Engrs, 1909–10, 36, 136–51.
21. Proc. Instn Munic. Engrs, 1916–17, 43, 304.
22. Oxford main drainage. Proc. Instn Munic. Engrs, 1876–77, 3, 146–55.
23. Crump, W.S., Sewerage works for Croydon Rural Sanitary Authority. Proc. Instn
Munic. Engrs, 1879–80, 6, 94–108.
24. Newman, J., Notes on concrete. SPON, 1887, p. 98.
25. C&CE, 1911, 6, 399.
26. Woods, C.F., Rock-concrete tubes. Proc. Instn Munic. Engrs, 1879–80, 6, 58–65.
27. It is possible that this was a concrete channel.
28. The Engineer, 1881, 51, 3.
29. Newhaven drainage. The Engineer, 1887, 64 & 92–96.
30. Given as an Appendix to Angell L. Mesh to cast iron beams. Proc. Instn Munic. Engrs,
1891–92, 18, 422.
31. Proc. Instn Munic. Engrs, 1904 –1905, 31, 382.
32. Wyrill, R.H., Reinforced concrete watermains. C&CE, 1906, 1, 185–92.
33. Sewer at Acton. C&CE, 1907, 2, 248–53.
34. Proc. Instn Munic. Engrs, 1907–1908, 34, 185–87; C&CE, 1907, 2, 319.
35. Mathews, E.R., Use of reinforced concrete in sewer construction. C&CE, 1919, 14,
357–64.
36. Mathews, E.R., Reinforced concrete in aqueduct and culvert construction. C&CE,
1908, 3, 129–33.
37. RC culvert at Kilton. C&CE, 1910, 5, 207–11.
38. Proc. Instn Munic. Engrs, 1912–13, 39, 323–24.
39. RC outfall sewer construction for Belfast. C&CE, 1917, 12, 5–12.
40. RC sewers. C&CE, 1907, 2, 407– 409; 489 (which illustrates a section supporting a
10 ton load).
41. Illustrated in reference 56, as is another in Portsmouth. Other reservoir covers are
illustrated in various issues of C&CE.
42. Lamborne new waterworks. C&CE, 1910, 5, 516–20.
43. Given in discussion in reference 45.
44. Reservoir for Skegness waterworks. C&CE, 1912, 7, 436–38.
45. Morris, W., Covered service-reservoirs. Proc. ICE, 1882–83, Pt.III, 73, 1–62.
46. Morris, W., Circular covered service-reservoirs. Proc. ICE, 1914 –15, Pt. 1, 199, 409–21.
47. Illustrated in discussion in reference 45.
48. Reservoir for Oldham Corporation. C&CE, 1912, 7, 432–35.
49. Walker, T., Some of the public works in Croydon. Proc. Instn Munic. Engrs, 1880–81,
17, 12–19.
50. Carlisle waterworks. The Engineer, 1886, 64, 515.
51. James, H.O., Newent waterworks of the Gloucester Corporation. Proc. ICE, 1901–
1902, Pt. II, 148, 342–43.
52. The roof was apparently on the Barrett System with the wet cement supported on
timber sheets between the bottom flanges of metal I beams.
53. Boulnois, H.P., Drainage of Portsmouth. Proc. Instn Munic. Engrs, 1884, 12, 128–31.
54. For discharge over one hour at high tide.
55. Mouchel-Hennebique. Ferro-concrete, list of works, 1897–1919. Produced by the
company. The date given is sometimes the completion date and sometimes the enquiry
date (for water towers the split is about 50 : 50).
56. Hennebique Ferro-Concrete Handbook. Produced by the company in c. 1921.
57. Referred to in C&CE, 1907, 1, 133 as new works, illustrated in Reinforced Concrete
by Winn, 1907; C&CE, 1956, 50, 248 as first in reinforced concrete and in Concrete,
September 1974, 47 as Historic Concrete No. 1. Although statutorily listed, it was
demolished c. 1997.
58. Preliminary report on the committee on reinforced concrete. ICE, 1910, 176–78. Was
first lined with bitumen protected by concrete in 1910. Although statutorily listed was
demolished in 1979.
59. Marsh, C.F., Dunn, W., Reinforced Concrete, 3rd edn. Constable: London, 1906,
403– 407 ‘for longitudinal steel assume wall acts as a slab spanning between the hoops
and built in at the ends’.
60. Hennebique Ferro-Concrete, 1909, 1, 346.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:29:49
16.qxd 15/11/2001 16:00 Page 340

340 Historic concrete

61. Cox, R.C., Gould, M.H., Civil Engineering Heritage of Ireland. Thomas Telford:
London, 1998; 144 – 45.
62. Expanded Metal Company Brochure No. 5, 1923.
63. Reinforced concrete coke hopper and water tank at Blackburn gasworks. C&CE, 1910,
3, 85. The coke hoppers braced the legs of the water tank.
64. In The York Waterworks Co. 1677–1968 published by the company. For repairs see
Water Services December 1985, p. 519.
65. Water tower at Cleethorps. C&CE, 1908, 3, 77– 80.
66. Historic concrete No. 48. Concrete, November 1987, 23.
67. Silcock, E.J., Goole water tower. W&WE, 1937, 41, 573–75.
68. Illustrated in C&CE, 1916, 11, 477.
69. Reinforced concrete water towers. The Engineer, 1911, 111, 86.
70. Reinforced concrete water reservoir at Edinburgh, NB. C&CE, 1908, 3, 86.
71. C&CE, 1916, 11, 545.
72. C&CE, 1916, 11, 477 & 482.
73. This date is probably that of the initial enquiry.
74. Mouchel Archives.
75. C&CE, 1916, 11, 544.
76. C&CE, 1916, 11, 478.
77. C&CE, 1916, 11, 482.
78. Rolleston on Dove water tower. C&CE, 1914, 9, 132–33. Also Historic concrete
No. 71, Concrete, October 1980, 29.
79. An Act to Encourage the Establishment of Public Baths and Wash-houses 9 &10 Vict
c. 74, 1846.
80. 1847, 10&11 Vict c. 61. Charges set were one person over 8 years (including towel)
cold 1d per bath warm 2d, several children (up to 4) (towel each) cold 2d per bath
warm 4d, open bathing place where several persons bath in the same water — half-
penny each. 9&10 Vict c. 87 extends the Act to Ireland but as this legislation was not
altered as was the English Act, it seems probably that it was little used.
81. Return of Bathhouses. Parliamentary Papers, 1865, xlvii, 279.
82. Most likely for labouring men, and for men and women, the 1846 Act requiring sep-
aration of the sexes.
83. 1878, 41&42 Vict c. 14.
84. Proc. Instn Munic. Engrs, 1883–84, 10, 17. These were presumably used by each sex
at different times.
85. Ibid., 1889–90, 16, 114.
86. Ibid., 1881–82, 8, 49.
87. Ibid., 1907–1908, 34, 79.
88. Hepworth, J., The Carlisle public baths. Ibid., 1887–88, 14, 274–78.
89. Ibid., 1889–90, 16, 51.
90. Proc. ICE, 1905–1906, xxxii, 180.
91. Ibid., 1905–1906, xxxii, 246.
92. Walker, T., Some of the public works of Croydon. Proc. Instn Munic. Engrs,
1890–1891, 17, 7.
93. Reinforced concrete swimming baths and storage tank at Chelsea. C&CE, 1907, 2,
235–40. In preparing the figure it has been assumed that the top of the tank is at
the same level as the duct.
94. Swimming bath, Dundee. C&CE, 1910, 5, 595–96. This says walls were reinforced
by columns 6 ft  5 in; it is unclear if this means 6 ft counterforts or 6 in attached
columns.
95. Ferro-concrete, list of works 1897–1919. This is probably the date of the initial enquiry.
96. Crowther, J.A., Northam open-air bath, constructed in ferro-concrete. Proc. ICE,
1912–13, xxxix, 534 ff.
97. Clegg, S., Practical Treatise on the Manufacture and Distribution of Coal Gas, 5th
edn. Trubner & Co.: London, 1868: 231, footnote.
98. Trewby, G.C., Description of a four-lift gasholder at Beckton. Incor. Inst. of Gas Eng.,
Trans, 1892, 167–172 and plates. The ‘four-lift’ refers to the telescopic bell.
99. King, A., Treatise of coal gas, 1870, 2, 47–109.
100. Douglas, J., Gasholder tanks in concrete. Br. Assn Gas Eng., 1874, 28.
101. King, p. 31.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:29:49
16.qxd 15/11/2001 16:00 Page 341

Water-retaining structures in Britain before 1920 341

102. Livesey, F., A description of a six-lift gasholder and tank now being constructed at
East Greenwich. Incor. Inst. Gas Eng. Trans., 1892, 156 –166 and plates.
103. Marshall, F.D., The Monier System of cement construction as applied to gas works.
Incor. Inst. Gas Eng. Trans., 1895, 145–65. The drawings are a little unclear as to
how the tank, for tar and ammonia, was built, but the walls appear to have been flat
arches to vertical I beams strutted off the floor.
104. Experimental tanks at the National Physical Laboratory, Teddington. C&CE, 1911,
6, 787–88.
105. Catalogue of facilities produced by the International Towing Conference.
106. Reinforced concrete gas mains. C&CE, 1917, 12, 612–16.
This illustrates women casting the pipes.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:29:49
17.qxd 15/11/2001 16:01 Page 343

17 Historic concrete in dams

D.A. Bruggemann, K.J. Hollock,


G.P. Sims
Introduction This chapter deals with the historic use of concrete as applied to dams from the
early days of masonry with thin mortar joints to modern roller compacted con-
crete and thin concrete arch dams.
The purpose of this chapter is to provide an outline of the development of
concrete dams taking into account the progress made over the last century in the
manufacture of cementitious materials, advances in concrete technology and
the refinement of analytical techniques for the design of arch dams. The chapter
is intended to provide a source of information for engineers who may be involved
in the rehabilitation of and modification to old dams.
The chapter deals with the concrete in the body of dams rather than in the
appurtenant structures such as draw-off works and spillways. These structures are
covered in other chapters on structural use of concrete. Information on the con-
crete mix proportions is provided, and covers the mortar in early masonry dams
up to modern roller compacted concrete. The references are intended to provide
a basis for further, more detailed literature searches for specific issues related to
rehabilitation or modification.
The development of cementitious materials, in relation to dam engineering, lead-
ing up to the introduction of pozzolans is described. This is followed by a discus-
sion on the transition from masonry to concrete dams. The evolution of the various
types of concrete dam (gravity, buttress and arch) is outlined, taking into account
the driving factors, such as cost, increasing size of dams and the mechanisation of
construction methods which led to roller compacted concrete (RCC). The chapter
concludes with data on mix proportions from early masonry dams, gravity
dams constructed in the 1950s and the 1970s, and arch dams constructed in the
1980s and 1990s.

Development of Rankine1 has distinguished between fat lime, hydraulic lime and cement as shown
in Table 17.1.
cementitious The distinction between these materials is useful in understanding the likely
material behaviour of an old dam, even though the boundaries between materials are some-
times hazy. The period of rapid growth in the use of cement began in the last years
of the 19th century as a result of intense research and promotion. Thus dams in
the 20th century have usually been constructed using cement, either in a masonry
mortar, or as the basis of concrete.

Table 17.1 Cementitious materials1

Description Silicate content Hardens in air Hardens under water

Fat lime Little or none Slowly No


Hydraulic lime 10 –30% Yes Slowly
Cement 40–60% Yes Quickly

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:30:31
17.qxd 15/11/2001 16:01 Page 344

344 Historic concrete

Before that lime was the most common building material and some of the largest
cathedrals in the world have been built with masonry cemented with it. Lime is
made by burning limestone, in the process driving off carbon dioxide (CO2),
leaving a clinker of calcium oxide, quicklime. When the quicklime is slaked with
water, it disintegrates into a fine grained powder or paste, depending on the
volume of water added. The pure slaked lime formed in this way is also known
as fat lime. It can be used for the construction of masonry but it hardens slowly
in air and not at all when isolated from atmospheric CO2. Masonry buildings made
with fat lime have been demolished and the lime remote from the air surface has
been found to be little harder than when it was placed.2
Roman masonry dams, of which there are several in Spain, may have used the
stronger and more durable roman cement comprising a mixture of lime and a
natural volcanic ash known as pozzuolana, from which the modern term pozzolan
is derived. The roman cement, like some naturally made cements contain more
silicate, often in the form of volcanic ash, or as clay minerals. When limestone
and clay containing silicates are burned together, they form a material capable
of setting in the absence of atmospheric CO2. The practical benefit of this is
that mortar made with the material would set hard even under water. Canal
building at the end of the 18th century required structures that would last well
under water and this led to the development of what came to be known as hydraulic
limes.
The production of hydraulic lime by the addition of silicates has been practised
in India for many centuries until fairly recently. This process comprised the add-
ition of ‘surkhi’, a clay rich in silicates, to fat lime.3 The setting process is speeded
up in the air and continues, albeit slowly, in water. A great deal of time and effort
was expended in producing the lime and ‘surkhi’ on dam sites.4
Hydraulic lime can be made into a satisfactory mortar that achieves a strength
similar to that obtained by cement mortars, albeit after about a year; concrete
made with it is less successful, presumably because of the length of time it takes
for the concrete to achieve useful strength. Tests conducted on large samples of
masonry show that the strength achieved by mortar made from hydraulic lime is
not much different from that achieved by cement mortar or indeed of cement
concrete.5,6
Portland cement was invented in 1825.7 It was strongly promoted and its advan-
tages over lime were gradually accepted. It was more expensive than hydraulic
lime, needing grinding and quality control to give a reliable product. It can be
made reliably into concrete. As the availability of skilled masons declined, and their
cost rose, the possibility of replacing masonry with concrete became an economic
necessity. In Europe, the use of lime in the construction of dams declined sharply
in the second half of the 19th century and cement was the normal material from
the beginning of the 20th century.
The first use of Portland cement for dam construction is attributed to Boyd’s
Corner Dam a 24 m high masonry dam, constructed in 1872, in Putnam County,
New York and at Pérolles in Switzerland almost simultaneously.8
The early cements were produced in vertical kilns and were coarse ground by
today’s standards. The cement was slow setting and construction rates were low,
thus the evolution of heat of hydration was low and little cracking occurred. From
about 1905 to 1920 the coarse-ground Portland cement was giving way to fine-
ground cement produced in rotary kilns. In the decade 1920–1930 there was a
surge in the use of mass concrete, at a time when the technology of concrete was
inadequately understood. The deterioration, after a short period of service, of
many concrete dams in Europe and in the USA was causing concern. The 1920s
were followed by rapid advances in concrete technology. Between the two World
Wars, the techniques of concrete production developed from a rough-and-ready

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:30:31
17.qxd 15/11/2001 16:01 Page 345

Historic concrete in dams 345

art producing a product of uncertain and often unsatisfactory quality, to the highly
developed and rigidly-controlled processes used today to produce good quality
concrete.9
In developing countries, however, cement remained too expensive and adequate
manpower was available at an acceptable cost until the middle of the 20th
century. Thus, the persistent use of locally produced hydraulic limes (lime-‘surkhi’)
in India. Being locally made, the quality is often variable and a stronger more
durable mortar is made with the addition of cement. Lime-surkhi mortar was
specified for the Chikkahole Dam which was completed in masonry in 1966.6
During the 1930s, when large volumes of concrete started to be placed at fast
rates, ‘low heat’ cements were developed in the USA to reduce the heat gener-
ated during hydration of the cement. This effect was achieved by limiting the
combined proportion of the tricalcium silicate (C3S) and tricalcium aluminate (C3A)
to less than 58%. This property is incorporated in cements classified as Type II
and Type IV in US practice.10

Pozzolanic As the 20th century progressed, efforts in dam engineering were made to reduce
the amount of cement in concrete for dams. These efforts were made in the inter-
materials ests of economy and to reduce the temperature rise in concrete and led to the
inclusion of pozzolans in concrete.
Pozzolans are siliceous substances, sometimes containing aluminium, that react
with calcium hydroxide to form cementitious compounds. Pozzolans may be
naturally occuring minerals such as clay or volcanic ash, or man-made materials
such as fly ash or ground blast furnace slag. Pozzolans have several advantages
over cement. They reduce the heat of hydration of concrete, enhance plasticity
and workability, and can improve concrete quality and resistance to Alkali Aggre-
gate Reaction (AAR).
As early as 1916, the United States Bureau of Reclamation (USBR) were
concerned about the build-up of heat of hydration of cement in massive concrete
dams. It was recognized that since the temperature rise in the interior of the mass
was nearly adiabatic, a decrease in the quantity of cement would decrease the
quantity of heat proportionately. On this basis, the USBR reduced the quantity of
cement by half for Arrowrock and Elephant Butte dams. This reduction was
done by the replacement of half of the cement with ground sand, which was a
forerunner of the pozzolan replacements of the 1950s. The resulting concrete
mass was as workable as all-cement concrete, but had lessened resistance to
weathering.11
The USBR continued to use pozzolan in the concrete for their dams. Friant
dam completed in 1942 included 20% pumicite pozzolan in the concrete. All the
dams constructed by the USBR between 1952 and 1963 included pozzolan in
the concrete. The pozzolanic materials included fly ash, calcinated clay, calcinated
shale and pumicite.10
An example of the use of man-made pozzolans (flyash) in the UK is reported
at the Lednoch Dam in Scotland, completed in 1958.12 Two mixes were used in
the construction of the dam and these are described later in the chapter.
Other advantages that have led to the increasing use of pozzolans to replace
cement in the second half of the 20th century include a lower alkali content of
the cementitious mass, hence a better resistance to Alkali Silica Reaction (ASR).
The properties of pozzolans vary widely and it has been found necessary to under-
take extensive tests on the pozzolanic mixes. It has been found necessary to include
air-entraining agents. Pozzolanic concretes tend to gain strength more slowly than
cement concrete. Some pozzolans cause large shrinkage and others reduce the
durability of the concrete.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:30:31
17.qxd 15/11/2001 16:01 Page 346

346 Historic concrete

Concrete dams developed from masonry dams, which had been in operation for
Transition from a substantial period. Schnitter8 notes the existence of four masonry dams built in
masonry to south-western Romania in the 1730s and 1740s and there are earlier examples
concrete dating back to the medieval period.
Masonry was popular for three reasons;4 first, masonry was a building skill that
has been readily available since the medieval period as demonstrated by churches
constructed in that period. Second, masonry is a cheap and practical building
technique when cement is scarce. Third, the spillway can be incorporated in the
masonry structure and so there is no need for a separate structure. In addition
to the above, the method has politically significant gains in terms of employment
potential because construction is labour intensive. It is for this reason that masonry
dams have been constructed in India until recent times.4
Transition took place from:
• masonry, where shaped blocks of stone were bedded on mortar and the joints
packed with stone and mortar through;
• cyclopean masonry, where random pieces of stone were placed on and
surrounded by concrete to;
• concrete construction, initially including large stone ‘plums’ or displacers placed
into the concrete (cyclopean concrete).
A fine example of a masonry dam is Vyrnwy Dam, which was completed in 1891
and is shown during construction in Figure 17.1. Note the large cranes required
to lift the blocks into place. The use of cyclopean masonry persisted in the
UK into the 20th century as indicated by the construction of Angram Dam
(completed 1912). The dam is illustrated in Figures 17.2 and 17.3, which show
the dam during construction and after completion respectively. The first dam to
be constructed in the UK from concrete with displacers was Blackbrook constructed
in 1906.
Many masonry dams, such as Boyd’s Corner dam, were constructed in the USA.
The end of the masonry dam era in the USA was marked by the construction of

Figure 17.1 Vyrmwy dam


under construction (Binnie
Archive, ICE).

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:30:31
17.qxd 15/11/2001 16:01 Page 347

Historic concrete in dams 347

Figure 17.2 Angram Dam


under construction (courtesy
of Yorkshire Water plc).

Figure 17.3 Angram Dam


after completion (courtesy of
Yorkshire Water plc).

New Croton Dam (91 m high) completed in 1905, in New York state and Theodore
Roosevelt Dam (85 m high) completed in 1911 and located on the Salt River in
Arizona. It is interesting to note that about half of the cement required for these
dams was Portland cement imported from England.11

The development The development of concrete dams was driven by the following factors:
of concrete dams • the understanding of the influence of the water–cement ratio on the strength
and durability of concrete;
• the need to control the temperature rise in the concrete as the size of dams
increased and as the rates of construction increased;
• economy;
• improvements in design techniques.
These factors led to change in the types of dam broadly from massive gravity
structures through buttress dams and cylindrical arch dams to double curvature

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:30:31
17.qxd 15/11/2001 16:01 Page 348

348 Historic concrete

arch dams and finally to gravity RCC dams. The evolution was gradual and not
in strict sequence as much development was carried out in parallel.
There were some engineers who seemed to have understood the importance
of the water–cement ratio before others. The engineer responsible for the Lower
Crystal Springs dam, constructed in 1890 about 30 km south of San Francisco, paid
special attention to the control of the water content in the concrete (some 250 kg/m3
referred to bone dry aggregates).8 Some 2 years after the completion of this dam,
L. René Féret of the French ‘Corps of Bridge and Highway Engineers’ published
results of tests carried out in 1892, showing the pre-eminent influence of the
water–cement ratio on the quality of concrete.13 The water–cement-ratio ‘law’ was
rediscovered for the USA and the English speaking world in 1918 by Duff
Abrams,14 and after this date the need to control the water–cement ratio was
generally accepted by the dam engineering fraternity.
The Hoover Dam shown in Figure 17.4 demonstrates the dramatic increase in
the quantity of concrete in the dam and the rapid rates of placing concrete which
were experienced in the early part of the 20th century. The dam was constructed
on the Colorado River in the south-western USA between 1931 and 1936. It was
designed by the USBR as a 221 m high massive curved gravity dam and contained
about 2,500,000 m3 of concrete.10 At that time the dam was 60% higher and 2.5
times larger than any existing dam. Concreting at Hoover dam took barely two
years with up to 8000 m3 of concrete placed per day.8,15 Figure 17.5 shows the fully
automated batching plant used at Hoover dam and gives an indication of the
magnitude of the concrete operation and would not look out of place on a modern
construction site for a large concrete dam. The dramatic increase in the speed
of construction was emphasized further at Grand Coulee Dam where concrete
placing rates of twice that at Hoover were required. Grand Coulee Dam is a straight
gravity dam containing about 7,400,000 m3 of concrete and was built between 1933
and 1942 by the USBR.8,15
Several techniques were used at Hoover Dam to control the temperature
of the concrete during the setting process. These techniques included the use of
low heat cement and post cooling by circulating cold water through pipes embed-
ded in the concrete. The post cooling technique was used for the first time

Figure 17.4 Hoover Dam


completed 1936.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:30:31
17.qxd 15/11/2001 16:01 Page 349

Historic concrete in dams 349

Figure 17.5 Fully automated


concrete batching plant,
Hoover Dam.

at Hoover, but low heat cement had been used for the first time in the USA
at Morris Dam built between 1932 and 1934, near Pasadena, California and at
Hoover dam.15
The practice of precooling concrete materials prior to mixing to reduce the max-
imum temperature of mass concrete began in the early 1940s and was extensively
employed in the construction of large dams during the late 1940s and 1950s.15
Gravity dams reached their prime in the 1960s8 but they were becoming too
expensive in comparison with earth and rockfill dams, which had benefited from
the development of earth moving technology during the Second World War and
the subsequent construction of the interstate highway network in the USA.16 In
response to rising costs, the concept of Roller Compacted Concrete (RCC) was
developed in 1970 in the USA by J.M. Raphael.17
The difference between RCC and conventional concrete dams is that earth
moving equipment is used to construct RCC dams. Concrete is placed in layers of
several hundred metres in length, and 300 mm thick, and consolidated by vibrat-
ing rollers. The construction period is much shorter, and consequently the labour
and plant costs are reduced. The RCC construction process is much safer than con-
ventional concrete dams because the layers are long and wide, and the height of
each lift is minimized. There is little or no formwork needed. Machinery is read-
ily available and so is effective and flexible. The first large RCC dam, Shimajigawa
Dam, was constructed in Japan between 1978 and 1981.18 By the end of 1998, 209
large (15 m high) RCC dams had been constructed and were under construction.19
In order to reduce the quantity of concrete in gravity dams and thus reduce
costs the buttress dam was developed from the massive gravity dam by introducing
intermediate spaces. The buttress dam has several advantageous features arising
from the introduction of the spaces such as reduced quantity of concrete, reduced
uplift and heat dissipation. These features are discussed below.
The spaces allow water seepage through foundation and dam to discharge not
only downstream, but also side- and upward into them, thus reducing the uplift
pressures. For the classical triangular profile with a vertical upstream face, the total
elimination of uplift pressures would theoretically have permitted a reduction
of the dams mass by 40% without reducing its stability. However, the actual

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:30:31
17.qxd 15/11/2001 16:01 Page 350

350 Historic concrete

reduction proved considerably smaller, because the intermediate spaces had to


be closed upstream with slabs, arches or a thickening of the buttress heads to
make them contiguous. The saving in costs was even more modest due to the
fact that buttress dams, per cubic metre of concrete, required more formwork of
a more complicated shape and, often, a higher cement content in the less easily
placed concrete. On the other hand, the intermediate spaces facilitated the dissi-
pation of the heat produced during the hardening of the concrete, so that
elaborate cooling measures were seldom needed during the construction of a
buttress dam.8
The buttress dam proved popular in the early part of the 20th century in areas
such as Australia and the western USA. These areas were undergoing similar
pioneering developments far from manufacturing centres and thus favoured
engineering solutions which minimized imports and transport costs.20a
The buttress dam type is designed in several forms:
• multiple arch,
• flat slab or Ambursen Dam,
• contiguous buttress.
A multiple-arch dam, Mir Alam, was constructed in 1804 in India by the Madras
Engineering Corps for the water supply of Hyderabad. It was almost 100 years
before the next multiple-arch dam was constructed. This dam was a 19 m high
dam constructed across the Belubula River near Sydney between 1896 and 1897.
At the same time, a similar design was being studied for a 31 m high structure near
Ogden, north of Salt Lake City, Utah.20a
An interesting example of a multiple-arch dam because it was the first applica-
tion of a double-curved dome-structure in dam engineering7 is shown in Figure
17.6 which shows the 76-m high Coolidge Dam under construction in 1928, 140 km
east of Phoenix, USA. The maximum arch span is 54.9 m and the thickness from

Figure 17.6 Coolidge Dam


under construction, 1928.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:30:31
17.qxd 15/11/2001 16:01 Page 351

Historic concrete in dams 351

1.2 to 6.1 m. In spite of its sophisticated formwork, the dome concept proved about
2% more economical than a conventional multiple-arch, 11% more economical than
a conventional arch, and 27% cheaper than a gravity structure.10
At the beginning of the 20th century reinforced concrete was relatively new and
spreading fast and a competitor to the multiple arch was the flat slab buttress dam
developed and patented by Nils F. Ambursen.8 This type of dam became quite
popular and by the end of the 1920s, more than 200 around the world had been
constructed, thus outnumbering multiple-arch dams by far. Among the flat slab
dams, was the 41-m high La Prele (or Douglas) irrigation dam, built from 1908
to 1909 in eastern Wyoming.8 The dam had to be provided with a new upstream
slab in 1977 since the original one had disintegrated by about 20% of its thickness
in the rough mountain climate.8,21 Outside the USA, the flat slab buttress dam
design was especially adopted after the Second World War for about 50 dams
exceeding 15 m in height in Norway.8,22
Despite the temporary popularity of reinforced concrete dams, especially the
Ambursen flat slab type, the rise in labour costs in relation to the prices of mater-
ials and the ensuing mechanisation, worked strongly against them. Moreover,
several of them had deteriorated rapidly in severe climatic conditions to which
they were sometimes exposed, or they otherwise developed cracks and leaks. By
the late 1920s engineers were increasingly questioning the wisdom of using thin
concrete members in dams at all.8
The problems with the flat slab reinforced concrete dams led to the develop-
ment of the contiguous buttress dam. Examples of these are V. Carranza irriga-
tion dam in north-eastern Mexico, the Burgomillodo dam, completed in 1929,
north of Madrid and the Dixence dam in Switzerland, completed in 1935.8
An example of the contiguous buttress dam in the UK is Haweswater Dam com-
pleted in the 1940s and shown in Figure 17.7. A diamond headed configuration
was adopted to concentrate the water load into each buttress and the buttresses
were thickened at the downstream end to give a continuous downstream face for
greater lateral stability and resistance against buckling under load.23

Figure 17.7 Haweswater


Dam, completed 1941 (Binnie
Archive, ICE).

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:30:31
17.qxd 15/11/2001 16:01 Page 352

352 Historic concrete

The move towards minimum cost dams (always perceived as minimum-concrete


dams) culminated in the design of arch dams. Arch dams may contain as little as
20% of the concrete of the equivalent gravity dam.23 Arch dams can be defined
as concrete (or masonry) structures, the base of which measured less than half their
height and which thus had to rely on their curvature in plan for the transmission of
part of the water load laterally into the valley flanks.8
Arch dams are designed, both in the single or double-curvature versions.
The strength of the rock in the abutments is therefore extremely important and
weaknesses have led to failures.23 The early arch dams tended to be the single curva-
ture type, e.g. Zola dam, 41-m high, completed in 1854. Zola dam was the first
dam to be designed on the basis of a stress analysis. The stresses were calculated
from the ‘cylinder formula’. From the 1830s through to the 1930s the method of
analysis of arch dams was refined from a series of independent arches to the ‘trial
load method’24 which determined the load distribution in the horizontal and ver-
tical directions by a trial and error method. Much of the research into methods
of analysis of arch dams was carried out by the USBR as part of the investigations
for the Hoover dam. After adaptation to electronic data processing and other
refinements, the method remained the most frequently used. It has now tended
to be replaced by the finite element method.8
G. Albert Pelletreau introduced the concept of the double curvature arch
dam in a study in 1879.8 He found that in order to minimize the volume of an
arch dam, its radius of curvature should decrease from crest to base. The ‘cylin-
der formula’ also yields smaller arch thicknesses for a given allowable stress.25
The idea was revived in the early 1900s in the USA: the variable radius design
was applied on the 51-m high Salmon Creek dam in south-western Alaska,
completed in 1914.8
The main break through for arch dams came in 1905 to 1925, mainly in the
fast developing west of the United States. Important contributions also came from
Switzerland, Italy and France, which took the lead in the field upon the decline
of arch dam construction in the United States in the 1930s.20b By the end of that
decade all-important elements had been assembled for the spectacular qualitative
as well as quantitative development of arch dams after the Second World War and
their proliferation around the globe. At present the age of the arch dam appears
to be drawing to an end except for very high structures.20b
The move towards the minimum-cost dams culminated in the design of the thin
arch dams in countries where suitable sites exist for such designs; in the Alps,
Portugal, Spain, etc.23 The design of theses structures has reached the extremes
of sophistication and now involves complex computer analysis as a complement to
(or substitute for) model tests in the laboratory. The degree of sophistication is such
that many of these structures are expensive to monitor, operate and maintain. There
are very few such structures in the UK; only Monar Dam is of any size and is shown
under construction in Figure 17.8 and after completion in Figure 7.11.
A recent example of a double curvature arch dam is shown in Figure 17.9. This
is the 122-m high Victoria dam constructed on the Mahaweli Ganga in Sri Lanka
between 1980 and 1984. In order to control heat generation, low heat cement was
used, mixing water was chilled to 5°C, some of the water was replaced with flaked
ice and a post cooling system was installed. Concrete placement rates of up to
40,000 m3 per month were achieved.26 It is interesting to note that the rate of place-
ment achieved is comparable to rates of production which were achieved at Hoover
dam some 50 years earlier and is a tribute to the organisational skills of the early
dam construction contractors in the USA.
The application of post tensioned pre-stressed concrete has not had a wide appli-
cation to the body of new dams. It has been used in special locations, e.g. the spill-
way, gate house, abutments and for remedial strengthening of old dams. An
exception to this is Allt na Lairige dam, constructed between 1953 and 1956, which

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:30:31
17.qxd 15/11/2001 16:01 Page 353

Historic concrete in dams 353

Figure 17.8 Monar Dam


under contruction.

Figure 17.9 Victoria Dam,


Sri Lanka, completed 1984.

was the first dam in western Europe to be constructed through out in prestressed
concrete. The dam is shown under construction in Figure 17.10 and note the very
thin section. An interesting fact about this dam is that the section of this dam is
used as the logo for the International Commission on Large Dams (ICOLD).

Concrete mixes In the early part of the century, concrete mixes were proportioned by volume.
The mix often varied over the height of the dam and at the masonry dams
Howden and Derwent (1912, 1916) the cement aggregate ratio varied as follows:23
• base section 1 : 6;
• middle section 1 : 51⁄2;
• top section 1 : 5.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:30:31
17.qxd 15/11/2001 16:01 Page 354

354 Historic concrete

Figure 17.10 Allt na Lairige


Dam, a post-tensioned,
pre-stressed concrete dam
completed 1956.

Figure 17.11 Monar Dam


completed 1963.

The proportion of stone to concrete in these dams formed about 44% of the whole
mass. The blocks were quarried Millstone grit, ranging from 0.5 to 6 t in mass,
and rubble stone was used for the concrete aggregate. The strength of the
concrete ranged from 9 to 24 MPa, with an average of 18 MPa at 1 year old, while
the crushing strength of the stone was 43–53 MPa.23
Engineers realized that for gravity dams, only low-strength high-density
concrete was required for the majority of the dam body. This realization in con-
junction with considerations of cost of cement and its thermal properties, led to
the specification of low cement-content concrete for the heart of the dam with ‘nor-
mal’ concrete on the exposed faces. The terms ‘hearting’ concrete and ‘facing’

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:30:31
17.qxd 15/11/2001 16:01 Page 355

Historic concrete in dams 355

Table 17.2 Concrete mixes in two UK gravity dams

Lednoch, 195812 Cow Green, 197023

Facing Hearting Facing Hearting

Cement and fly ash (kg/m3) 235 173 350 180


Fly ash proportion (%) 15 20 20 20
Aggregate (kg/m3) 76 mm 500 529 – 640
38 mm 830 426
19 mm 500 529 560 426
Coarse sand 406 575
Fine sand 472 498 135 64
Water–(cement  fly ash) 0.5 0.6 0.5 0.84
ratio
Water–cement ratio 0.6 0.75 0.63 1.05
Density (kg/m3) – – – 2520

concrete were coined, with an intermediate mix ‘infill’ concrete being used to infill
pockets in the rock foundation.23 The term ‘infill’ concrete was also used to infill
between buttresses (in certain designs) or to infill around shafts and galleries when
box outs had been left for their construction.23
Typical mixes for facing and hearting concrete are shown in Table 17.2.12,23
It is interesting to note that the water–cement ratios are similar and that the
proportions of cement and pozzolan are similar for the hearting concrete. The
cementitious material content of the facing concrete is 350 kg/m3 in 1970 compared
with 235 kg/m3 in 1958. The increase in cementitious material suggests that
increased durability was sought at Cow Green.
Currently most gravity dams are constructed from roller compacted concrete
(RCC). Dunstan 199919 notes that broadly there are five forms of RCC construc-
tion as shown below:
• High paste RCC in which the cementitious content (cement and pozzolan) is
greater than 150 kg/m3. An example of this type of dam is Upper Stillwater
Dam in the USA, completed in 1987.27
• Medium paste RCC in which the cementitious content is between 100
and 149 kg/m3. An example of this type of dam is Salto Caxias Dam in Brazil,
completed in 1998.28
• Roller Compacted Dam (RCD) which is a technique unique to Japan. The
cementitious content is generally 120–130 kg/m3 and produces a relatively
expensive, but high quality, structure suited to the high dynamic loading
conditions in Japan.19 An example of this type of dam is Shimajigawa Dam
completed in 1980.18
• Lean RCC in which the cementitious content is less than 99 kg/m3. An
example of this type of dam is Willow Creek constructed in the USA in 1982.27
• Hardfill dams, which are essentially cement stabilized rockfill and are particu-
larly suited to areas in which there is high seismic loading and relatively weak
foundations.18 The first large hardfill dam was Marathia Dam completed in
Greece in 1993.18
It is interesting to note that the Portland cement content of RCC in the main coun-
tries of application (China, Japan, USA and Spain) is between 75 and 80 kg/m3.
The difference in cementitious content is created by the difference in pozzolan
content.19 These figures show that commonly a large proportion of the Portland
cement is replaced with pozzolan generally with attendant cost savings and lower
heat of hydration.
Table 17.3 summarizes the RCC types and gives the proportion of each type
either under construction or completed. For purposes of comparison with

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:30:31
17.qxd 15/11/2001 16:01 Page 356

356 Historic concrete

Table 17.3 Roller compacted concrete dam types19

Description Proportion (%)

High paste RCC cementitious content 150 kg/m3 46.5


Medium paste cementitious content 100–149 kg/m3 21.5
RCD (Japan) cementitious content 120–130 kg/m3 18
Lean RCC cementitious content 99 kg/m3 12.5
Hardfill 1.5

Table 17.4 Concrete mix for two modern double curvature arch dams

Victoria Dam, Sri Lanka, 198426 Katse Dam, Lesotho, 199729

Volume of concrete (m3) 670,000 2,360,000


Cement and fly ash (kg/m3) 230 220
Fly ash (%) 0 30
Aggregate maximum size (mm) 150 150
90-day Characteristic strength – 24 MPa
28-day 23 MPa (specified), –
30 MPa (achieved)

gravity dams, the concrete mixes for two modern double curvature arch dams are
shown in Table 17.4. The total cementitious material content is similar in both cases
except that Katse Dam in Lesotho incorporated fly ash. Arthur et al.30 note that
fly ash was readily available from nearby South Africa and it was a viable option
to replace some of the cement with this pozzolan. Cole and Neal 199126 do not
mention if cement replacement by fly ash was considered for Victoria Dam. The
table also shows that the cementitious content is higher than for RCC dams and
similar to the older gravity dams.

Conclusion The chapter has examined the development of cementitious materials their appli-
cation to dams including the use of pozzolans. The transition from masonry to
mass concrete dams and the development of concrete dam types has been explored.
Some general information on the concrete mixes used in masonry dams at the
beginning of the last century and up to recent times has been provided.

Acknowledgements The authors would like to express their thanks to Brown and Root Services for their
support in the preparation of this paper. Thanks are due to Yorkshire Water PLC
for providing a selection of historical photographs for the authors to select from.

References 1. Rankine, W.J.M., A Manual of Civil Engineering. Charles Griffen: London, 1865.
2. Scott, H.V.D. Account of the manufacture of a new cement. Corps of Royal Engineers,
Papers , X, 1861, 136.
3. Hill, N., Holmes, S., Mather, D., Lime and Other Alternative Cements. I T Publica-
tions, 1992. ISBN 1 85339 178 6.
4. Doddiah, D., Selection, processing and specification of stones and mortar for the con-
struction of Surki mortar masonry dams in India. In: Proc. 7th ICOLD, 1961: Q24,
R32, 525–44.
5. Sims, G.P., Ageing of masonry dams. In: Proc. Instn Civ. Engrs, Wat. Marit. Energy,
1994: 106, 61–71.
6. Murthy, Y.K., Mane P.M., Pant B., Tensile failures in some stone masonry gravity dams
in India. In: Proc. 13th ICOLD, 1979: Q49, R31, 461–77.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:30:31
17.qxd 15/11/2001 16:01 Page 357

Historic concrete in dams 357

7. Eckel, E.C., Cements, Limes and Plasters, their Materials, Manufacture and Proper-
ties. John Wiley, New York, 1928.
8. Schnitter, N.J., A History of Dams, the Useful Pyramids. A.A. Balkema: Rotterdam,
Brookfield, 1994.
9. Crossthwaite, C.D., Hunter J.K., The deterioration of concrete dams, 40 years expe-
rience in North Wales. In: Proc. 9th ICOLD, 1967: Q34, R13, 207–26.
10. Price, W.H., Higginson E.C., Bureau of reclamation practices in mass concrete. In:
Symp. Mass Concrete, Am. Concr. Inst., 1963, Special Publication SP-6, Paper No. 4,
77– 87.
11. Kollgaard, E.B., Chadwick, W.L., Development of Dam Engineering in the United
States. Pergamon Press, New York, 1988.
12. The Engineer, 18 July 1958.
13. Féret, L.R., Sur la compacité des mortiers hydrauliqes, 1892.
14. Abrams, D.A., Design of Concrete Mixtures. Structural Materials Research Laboratory:
Chicago, 1918.
15. Davis, R.E., Historical Account of Mass Concrete, Am. Concr. Inst., 1963, Special
Publication SP-6, Paper No.1, 1–35.
16. Chadwick, L.W., In: Kollgaard, E.B., Chadwick, L.W. (eds), Influence of Some Related
Technologies on the Technology of Dams, Development of Dam Engineering in the
United States. Pergamon Press, New York, 1988.
17. Raphael, J.M., The optimum gravity dam, rapid construction of concrete dams. In:
Proc. Eng. Res. Conf., Asilomar, 1970: 221–47.
18. Dunstan, M.R.H., The state-of-the-art of RCC dams. Int. J. Hydropower Dams,
1994, 1(2).
19. Dunstan, M.R.H., Recent developments in RCC dams, Int. J. Hydropower Dams, 1999,
6(1), 40–45.
20. (a) Schnitter, N.J., The evolution of buttress dams. In: Studies in the History of Civil
Engineering, Vol. 4, Dams, Ashgate-Variorum, Aldershot, Brookfield USA, Singapore,
Sydney, 1997; (b) Schnitter, N.J., The evolution of the arch dam. In: Studies in the
History of Civil Engineering, Vol. 4, Dams, Ashgate-Variorum, Aldershot, Brookfield
USA, Singapore, Sydney, 1997.
21. Tolle, J.M., Simard P.W., Brown L.A., Modern Engineering Saves Troubled Dam, Civil
Engineering, New York, 1979, June, 78–80.
22. Berdal, A.B., Slab and buttress type dams. In: Concrete Dams in Norway, Norconsult
AS, Oslo, 1968: 5–8.
23. Kennard, M.F., Owens, C.L., Reader, R.A., Engineering guide to the safety of concrete
and masonry dam structures in the UK, Report 148, CIRIA, 1996.
24. USBR, Trial Load Method of Analysing Arch Dams, United States Bureau of Recla-
mation: Denver, 1938.
25. Pelletreau, A., Barrages cintrés en forme de voûte. Annales ponts et chaussées, 1879,
1er sém., 198–218.
26. Cole, K.L., Neal, P.C., The Victoria Project, Sri Lanka: Construction of Victoria Dam.
In: Proc. Instn Civ. Engrs, Part 1, Vol. 90, Paper 9528, 1991.
27. ICOLD, Roller Compacted Concrete for Gravity Dams — State of the Art. Bulletin
No. 75, 1989.
28. Kamel, K.F.S., Babá, L.J.N., Machado, B.P., Salto Caxias: Brazil’s largest RCC Dam.
Intern. J. Hydropower Dams, 1999, 6(1), 56–60.
29. Develay, D., Hagen, R.J., Bestagno, R., Lesotho highlands water project — design and
construction of Katse dam. Suppl. Civil Eng., 1997, Vol. 120, Special Issue 1.
30. Arthur, L.J., Wagner, C.M., Hein, B., Lesotho highlands water project — design of
the ‘Muela’ hydropower station. In: Proc. Instn. Civil Engrs, Suppl. Civil Eng., 1997,
Vol. 120, Special Issue 1.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:30:31
18.qxd 15/11/2001 16:02 Page 359

18 Concrete roads

Graham West
Synopsis Considered as historic artefacts, concrete roads lack much of the visual appeal of
other concrete structures such as dams and bridges. Nevertheless, concrete roads
are a part of our civil engineering heritage and have an interesting history which
is worth studying in its own right. This chapter deals with the history of concrete
roads from earliest times to the present day, but before dealing with their history
a basic description of the structure and composition of concrete roads, including
some definitions and explanations of the technical terms used, will be given so that
the reader may understand the descriptions that follow.

Structure and The basic structure and main components of a concrete road are shown dia-
grammatically in Figure 18.1. The slab and the sub-base taken together are referred
composition to as the pavement. The soil on site beneath the road is referred to as the subgrade.
The main structural function of the pavement is to spread the load applied by
the traffic over a sufficiently wide area of the subgrade such that the bearing cap-
acity of the soil is not exceeded. The principle element of a concrete road, the
concrete slab, is not generally cast directly on the subgrade, but on a thin layer of
granular material (e.g. gravel, sand) called the sub-base which is placed on the soil
first. (In the road industry, concrete pavements are sometimes called rigid pave-
ments, in contrast to bituminous pavements which are called flexible pavements — this
terminology reflects their different material properties.) The load-carrying cap-
acity of the concrete road structure depends mainly on the structural rigidity of
the slab and so long as a concrete slab remains in sound condition it will perform
its structural function satisfactorily. Since the late 1980s, if the bearing capacity of
the subgrade soil is very low (California bearing ratio of 5 per cent or less) a capping
layer is spread on the subgrade before the sub-base is laid. The capping layer is a
material of lower quality than the sub-base but of higher quality than the subgrade.
As well as assisting in load-spreading, the capping layer provides a working plat-
form for construction of the road. One section of concrete slab is separated from
another by regularly spaced, vertical joints and the traffic load is transferred across
the joints by horizontal steel connectors called dowel-bars.
Concrete is strong in compression but weak in tension. Because of this plain
concrete slabs tend to be prone to cracking, and to prevent this from happening
the concrete slab is often reinforced with steel. Traditionally, therefore, there have
been two basic methods of constructing concrete road pavements, and these are

Slab
Pavement

Sub-base

Capping layer
Figure 18.1 Concrete road
(if required)
structure (diagrammatic). Subgrade

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:32:34
18.qxd 15/11/2001 16:02 Page 360

360 Historic concrete

Figure 18.2 Jointed


unreinforced (top) and jointed
reinforced (centre) concrete
road pavements. Unjointed
continuously reinforced
concrete road pavement
(bottom).

the jointed unreinforced concrete road pavement and the jointed reinforced concrete road
pavement: they are shown in Figure 18.2, and make up the majority of concrete
roads in the United Kingdom. Joints tended to be troublesome, so that in addi-
tion to these two methods there is another type of construction that came into use
in the 1980s, the unjointed continuously reinforced concrete road pavement, which is
also shown in Figure 18.2; as its name indicates, this slab has no joints.1 (It should
be added that roads made from bituminous surfacing materials sometimes
have a lean concrete roadbase but this use of concrete will not concern us in this
chapter.)

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:32:34
18.qxd 15/11/2001 16:02 Page 361

Concrete roads 361

A typical road-slab concrete would have had mix proportions of 1 : 2 : 4 (1 part


cement : 2 parts fine aggregate : 4 parts coarse aggregate), a water : cement ratio
of 0.5 : 1 and a 28-day compressive strength of about 35 MN/m2. The thickness of
the slab of a concrete road pavement depended on whether it was reinforced or
not and on the traffic loading, but for an unreinforced slab it varied from 230 mm
for light traffic to 330 mm for heavy traffic. All types of concrete road pavements
can be laid using either fixed-form or slipform pavers. Traditionally, roads made
of unreinforced or reinforced concrete were designed to have a life of 40 years.

Precursors of Concrete has been known since about 1350 BC when it was used to construct the
foundations for the city of Akhet-Aten in ancient Egypt.2 The Romans were great
concrete roads users of concrete, employing it in building and civil engineering to great effect
almost everywhere in their empire where they could.3 A cross-section through the
pavement of a Roman road is shown in Figure 18.3, with dimensions given for
the various layers as measured in the archaeological excavation. The Romans’
names for the layers of the pavement are also given.
The section of road is the Fosse Way near Radstock in Somerset, and shows that
for this important road the Romans had used concrete for both the rudus and nucleus.
The cement used to make the concrete was probably lime-based and it is likely that
it was manufactured locally. The total thickness of pavement was a massive 863 mm,
of which the concrete comprised 635 mm. The Fosse Way was a Roman main road
and the pavement structure shows that it was constructed to their highest standard.
The use of concrete almost died out with the decline of the Roman Empire, and
apart from some isolated instances, it was not until the middle of the 18th cen-
tury, with the construction of Smeaton’s Eddystone Lighthouse, that a revival in
the use of concrete came about. With introduction of improved cements in the
19th century, concrete as a construction material began its steadily rise in import-
ance which has continued to the present day.

Modern history of The first concrete road in Great Britain were constructed in Inverness and
Edinburgh, Scotland, in 1865–66; the concrete was laid fairly dry and compacted
concrete roads by roller, and performed satisfactorily until 1875 when the concrete was proba-
bly damaged by frost. In the early years of the 20th century experimental lengths
of concrete pavement were constructed by County and Port of London Authority
engineers. The first concrete road in the United States was constructed in 1893
at Bellefontaine, Ohio. This was followed in 1909 by a concrete highway built at
Wayne County, Michigan; and another in 1912 at Saltney in Chester City, Ohio.
However, in the United States by 1920, concrete pavements were being laid at the

Pavimentum

101 mm lime-grouted polygonal slabs

Nucleus 254 mm fine concrete

Rudus 381 mm coarse concrete

Figure 18.3 Section through 127 mm rubble


a Roman road, the Fosse Way
Solum
at Radstock, Somerset. Statumen

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:32:34
18.qxd 15/11/2001 16:02 Page 362

362 Historic concrete

rate of 2000 miles per year, and the Experimental Branch of the Ministry of Trans-
port thought that this form of construction would be suitable for use in southern
England where sand and gravel aggregates were plentifully available. As a result,
a number of by-passes and other roads were constructed in concrete, often incorp-
orating experimental features aimed at developing a general specification for
concrete pavements.4 Some of these pre-war concrete roads were:
• Kingston Bypass, Kingston-upon-Thames, Surrey, A3 (Figure 18.4);
• Great West Road, west of London, A4;
• Caterham Bypass, Caterham, Surrey, A22;
• Great Chertsey Road, near Chiswick, Middlesex, A316;
• Colnbrook Bypass, Middlesex, A4.
Many of these bypasses were still satisfactorily carrying traffic after over 40 years
service, although they have for various reasons been overlaid with bituminous sur-
facing, and it is not now readily apparent that these were once concrete roads.
Notably, the Lancashire County Council constructed many concrete roads in the
1930s, such as the Maghull Diversion, built in 1936, and pictured in Figure 18.5.
Many of the early concrete road slabs were laid and finished predominantly by

Figure 18.4 Kingston


Bypass, Surrey, 1934.

Figure 18.5 The Maghull


Diversion, Lancashire, 1936.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:32:34
18.qxd 15/11/2001 16:02 Page 363

Concrete roads 363

Figure 18.6 Finishing with


smoothing board on the
Rickmansworth to Denham
Road, Buckinghamshire,
pre-1935.

Figure 18.7 The Dingler


compaction and paving train
at work on the Mickleham
Bypass, Surrey, in the early
1930s.

hand methods (Figure 18.6), but on some roads a forerunner of the modern paving
train was making its appearance (Figure 18.7).
The service experience of these early concrete roads, however, cannot have been
wholly satisfactory because when the motorway network was designed and con-
structed in the 1950–60s; the Ministry of Transport did not allow concrete-surfaced
roads to be built. However, during this period the Road Research Laboratory
carried out a number of full-scale concrete road experiments which resulted in
renewed interest in concrete roads.5 The concrete industry made successful rep-
resentations against the Ministry’s policy, and after 1969 contractors were allowed
to use either concrete or bituminous construction for the Department of Transport’s
trunk roads and motorways. The proportions, and corresponding lengths, of con-
crete roads in the United Kingdom, as estimated in 1986, are given in Table 18.1.
In July 1992, following concern about the high level of noise generated by
vehicle tyres running on concrete road surfaces, the Department of Transport once
again imposed a ban on new concrete-surfaced construction for heavily trafficked

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:32:34
18.qxd 15/11/2001 16:02 Page 364

364 Historic concrete

Table 18.1 Concrete roads in the United Kingdom

Category of road Proportion (%) Length (km)

All roads 5 17 500


All trunk roads and motorways 20 3000
Trunk roads and motorways constructed since 1969 22 530
M25 motorway 46 90

motorways and trunk roads, although it still allowed concrete roadbase to be used
with an asphalt surfacing.6 In 1996 the Department of Transport announced that
an exposed-aggregate concrete surface, which was quieter than the conventional
brushed concrete surface, would be an allowable option for new road construction.
The situation in the United Kingdom, where the minority of trunk roads and
motorways roads are of concrete, can be contrasted with that in the United States
where 53 per cent of interstate highway pavements are of concrete.

Examples of Five examples of historic concrete roads are now given: the first is from Italy and
the remainder are from England.
historic concrete
roads
Milan to Lake Maggiore autostrada

The first motorway (autostrada) in Europe was planned in 1921 by an Italian engin-
eer, Piero Puricelli. His idea was to connect the city of Milan in northern Italy
with the neighbouring towns of Como, Varese and Lake Maggiore with a road
tailored specifically to the needs of the motor car. The Italian government approved
the project, construction work began in June 1923 and the whole project was com-
plete in September 1925; the total length being 84 km. The motorway consisted
of a single two-way carriageway varying in width from 8.5 to 10.5 m, separation
of opposing traffic being by a continuous dividing line of white paint — looking
back it is rather amusing to think that this was considered adequate. After stud-
ies of the various kinds of pavement used for motor roads in the United States,
the Italian motorway was constructed with a concrete pavement, special continu-
ous mixers and pavers laying the whole width of the carriageway in one pass. The
mix proportions were such that each cubic metre of concrete contained 0.75 m3
crushed stone aggregate, 0.5 m3 of sand and 350 kg of cement.7

Alconbury Hill

In 1957 the Road Research Laboratory conducted a full-scale pavement design


experiment on a section of dualling on the Great North Road, A1, at Alconbury
Hill, Huntingdonshire. Part of the experiment involved the construction of 1.6 km
of concrete pavement. The main variables studied were: thickness of slab
(127–279 mm); thickness (76–229 mm) and type (natural sandy gravel and lean
concrete) of sub-base; amount of reinforcement; type of aggregate (crushed rock
and natural gravel); and strength of concrete (44 and 64 MN/m2 compressive
strength at 28 days). A central batching and mixing plant was erected on site for
preparing the concrete, which was delivered by side-tipping lorries to a train or
rail-mounted concrete-laying machinery spanning the 7.9 m width of the car-
riageway. The reinforcement in the reinforced concrete slabs consisted of oblong
mesh placed at a depth of 64 m below the surface. The subgrade was a heavy clay
having a California bearing ratio of 5–9 per cent at the time of construction.
Figure 18.8 shows the concrete slab under construction, and Figure 18.9 the

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:32:34
18.qxd 15/11/2001 16:02 Page 365

Concrete roads 365

Figure 18.8 Laying the


concrete slab from a
rail-mounted paving train
at Alconbury Hill,
Huntingdonshire, A1, 1957.

Figure 18.9 Near side: the


just-completed new concrete
carriageway at Alconbury Hill.
Far side: the old Great North
Road still carrying two-way
traffic, 1957.

completed carriageway just before opening to traffic. The main conclusions after
six years of traffic (4000 commercial vehicles per day) were that only reinforced
slabs of 203 mm thickness and over, laid on a granular sub-base at least 76 mm
thick were likely to give a satisfactory long-term performance. Also, it was found
that satisfactory pavement-quality concrete could be made using either crushed
rock or natural gravel aggregate.8

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:32:34
18.qxd 15/11/2001 16:02 Page 366

366 Historic concrete

St Albans Bypass

The St Albans Bypass in Hertfordshire, 27 km long, was part of the M1


motorway, Britain’s first full-length motorway, and built in 1958–59. Although the
rest of the M1 (92 km) was of flexible construction, the St Albans Bypass was of
rigid construction so that the characteristics and performance of the two types of
pavement under traffic could be compared. The pavement consisted of a 279 mm
thick reinforced concrete slab overlying a 178 mm thick hoggin sub-base, giving
a total pavement thickness of 457 mm. Figure 18.10 shows a train of concreting
plant in operation laying the slab on the St Albans Bypass, and Figure 18.11 gives
a general view of the completed road.

Padiham Bypass

The 1 km long section of concrete pavement on the Padiham Bypass, A6068,


Lancashire, was constructed in 1963 to provide the Lancashire County Council
with experience in preparation for the forthcoming construction in concrete of
the Broughton to Hampson Green section of the M6 motorway, using the same
paving machine and source of aggregates as were to be used for the motorway.
The total slab thickness was 254 mm, made up of a 178 mm thick layer made with a
siliceous limestone coarse aggregate, and a 76 mm thick upper layer made with

Figure 18.10 Train of


concreting plant in operation
on the St Albans Bypass,
Hertfordshire, M1, 1958–59.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:32:34
18.qxd 15/11/2001 16:02 Page 367

Concrete roads 367

Figure 18.11 General view


of the completed St Albans
Bypass, 1958–59.

siltstone coarse aggregate. The fine aggregate used in both layers was a sea-dredged
natural sand. The reinforcement in the slab consisted of a mesh laid between the
two layers, 9.5 mm diameter cranked tie bars 0.76 m long, and dowel-bars at the
contraction joints. The joint spacing was 12 m. The slab was laid in two 3.65 m wide
runs, using an SGME box spreader, a compactor and Jaeger finisher, and was
placed on a 152 mm thick sub-base of wetmix crusher-run limestone. The two car-
riageway slabs were held together by the tie bars running across the longitudinal
joint. The completed road, therefore, consisted of a two-lane single carriageway,
7.3 m wide. Map-cracking was noticed on the road surface in 1979, but by 1989
had become serious enough, particularly near the transverse joints, to merit an
investigation, which was carried out by the Transport Research Laboratory.9 Cores
were taken and thin-sections of the concrete were examined under the polarizing
microscope (Figure 18.12). It was found that the slab was affected by alkali–silica
reaction, both the coarse aggregates being reactive. The reactive constituent was
found to be microcrystalline quartz which was present in both coarse aggregates.
The source of alkali was identified as the deicing salt applied to the road surface
in the winter.10

M25 motorway

The London orbital motorway, M25, constructed between 1976 and 1986, and
195 km long, was the last full-length motorway to be built in Great Britain. The
Department of Transport decreed that for the M25 they would allow contractors
to submit tenders for either flexible or rigid construction for the road pavement.
The outcome was that 54 per cent of the motorway has a flexible pavement and
46 per cent has a rigid pavement — this result is due to the very close tenders
which in turn must reflect the very close costs of the two forms of construction a
the time. On the Epping section, in the north-east quadrant of the M25, the road
pavement was an example of rigid construction. There, the continuously-rein-
forced concrete slab had a thickness of 250 mm with the reinforcement placed at

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:32:34
18.qxd 15/11/2001 16:02 Page 368

368 Historic concrete

Figure 18.12 Photomicro-


graph of thin-section of
concrete from the Padiham
Bypass, Lancashire, A6068.
A microcrack can be seen
passing across the field of
view. The reactive constituent
is microcrystalline quartz in
the siltstone aggregate.

mid-depth, the concrete was made using flint gravel aggregate, and the cement
content was 320 kg/m3. Figure 18.13 shows the Westerham to Sundridge section
of the M25 where, in 1979, the unreinforced concrete slab was laid with a con-
ventional rail-mounted paving train; and Figure 18.14 shows a fixed-form paving
train being used on the Wisley to Leatherhead section of the M25 in 1984. Some
of the concrete on the M25 was later given a bituminous surface dressing so that
its concrete nature is not obviously apparent now.

The future At the risk of oversimplification, it can be said that experience with increasingly heavy
traffic has shown that flexible pavements need to be stiffer to resist deformation,
but that rigid pavements are, perhaps, too stiff and liable to crack. One solution to

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:32:34
18.qxd 15/11/2001 16:02 Page 369

Concrete roads 369

Figure 18.13 Westerham to


Sundridge section of M25,
1979. The unreinforced
concrete slab was laid with
a conventional rail-mounted
paving train.

Figure 18.14 Fixed-form


paving train on the Wisley to
Leatherhead section of M25,
1984.

Figure 18.15 Composite


concrete and bituminous
pavement.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:32:34
18.qxd 15/11/2001 16:02 Page 370

370 Historic concrete

these problems would be to combine the best elements of rigid and flexible pave-
ments in a single composite pavement: namely a concrete roadbase with hot-rolled
asphalt surfacing, as shown in Figure 18.15, and this may well be the road pave-
ment of the future. In which case, concrete roads will still be with us in the new
millennium, but they will be hidden beneath the surface.

Acknowledgement The British Cement Association is thanked for providing Figures 18.2, 18.4 –18.11
and 18.13–18.15.

References 1. British Cement Association, Concrete Pavements for Highways, 2nd edn. Publication
46.030. British Cement Association: Crowthorne, 1994.
2. Hammond, N., Excavation shows pharaohs knew the secret of concrete. The Times,
11, October 1993.
3. Stanley, C.C., Highlights in the History of Concrete. Publication 97.408. Cement and
Concrete Association: Slough. 1980.
4. Croney, D., The Design and Performance of Road Pavements. HM Stationery Office:
London, 1977.
5. Road Research Laboratory, Concrete Roads: Design and Construction. HM Stationery
Office: London, 1955.
6. Department of Transport, Minister announces steps to tackle road noise problem. Press
Notice No 204. Department of Transport: London, 1992.
7. Mangarano, A., Pellizzi, G., Italy’s first motorways. AIPCR-PIARC 1909–1969. In: Per-
manent International Association of Road Congresses, Paris, 1970, 129–40.
8. Croney, D., Loe, J.A., Full-scale pavement design experiment on A1 at Alconbury Hill,
Huntingdonshire. Paper No 6848. In: Proceedings of the Institution of Civil Engineers,
1965, 30, February, 225–70.
9. Sibbick, R.G., West, G., Examination of concrete from the Padiham Bypass, Lancashire.
Research Report 304. Transport Research Laboratory: Crowthorne, 1992.
10. West, G., Alkali-Aggregate Reaction in Concrete Roads and Bridges. Thomas Telford:
London, 1996.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:32:34
19.qxd 15/11/2001 16:03 Page 371

19 Military

John Weiler
Introduction Military uses of concrete in construction were shaped by circumstances that dif-
fered in at least three important respects from those prevailing in civil projects.
Most obvious was the overriding concern for national defence and the achieve-
ment of war aims. Then, there was the regimen of military discipline under which
the armed services worked coupled with the government bureaucracy that con-
trolled military contracts awarded to the private sector. Finally, by its very nature,
military construction engaged in a continuous game of technological leap-frog with
the fire-power of weaponry and its means of delivery by land, sea and air.
The story of concrete in military works extends from early 19th century experi-
ments to the Second World War and its aftermath. Royal Engineers figured pro-
minently in developments, both as individuals and as a corps. Most often they
worked in collaboration with civil engineers and others engaged in the business
of construction. During the 20th century, the private sector took increasingly
greater responsibility in the design and construction process. Although the clear
emphasis in this narrative is on British achievements, frequent reference is made
to significant developments in other countries to provide historical perspective.

Early experiments In the early 19th century, Charles William Pasley and other Royal Engineers joined
a controversy surrounding the introduction of concrete and artificial stone
with concrete in England, particularly concerning the work of William Ranger. Pasley was direc-
tor of the Royal Engineer Establishment at Chatham and a prodigious experi-
menter and recognized authority on cementitious materials as well as developer
of an artificial Roman cement that bore his name. What gave particular impetus
to the Royal Engineers’ involvement in the issue was a directive from the Board
of Ordnance to ascertain the fitness of Ranger’s patent concrete for casemate arches.
On orders from the Inspector-General of Fortifications, Lieutenant Colonel
George Harding oversaw the construction of a model vault by Ranger at Woolwich
Marshes in 1835 (Figure 19.1). Concrete specifications were seven parts gravel and
sand mixed with one part Dorking lime and one part and a half of boiling water
(a peculiarity of Ranger’s formula). The model vault had a span of 17 ft and a rise
of 9 ft and contained 5947 ft3 of concrete. It stood up well under bombardment
notwithstanding the fact that the arch had already cracked because of inadequate
foundations on the marshy soil and even though the core of the structure was still
soft (Ranger’s concrete was quick setting but slow hardening). Harding reported
that he was prepared to recommend Ranger’s concrete be adopted in the arches
of small magazines and casemates by virtue of its strength and economy, but he
was doubtful about using it for the core of brick piers because of the difference
of compression and expansion in the two materials; he also concluded that a 4 ft
thick arch of concrete would be bombproof.1 There is no evidence that Harding’s
recommendations were accepted or implemented.
Pasley argued that in the face of artillery fire brick casemates laid in cement mor-
tar and coated with pure cement would be stronger than concrete and less prone
to dampness.2 In 1836 he began experiments to determine how any given lime
was fit for making concrete on Ranger’s principle, and his curiosity induced him

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:33:02
19.qxd 15/11/2001 16:03 Page 372

372 Historic concrete

Figure 19.1 Method of


constructing Harding’s
concrete model casemate,
1835. (Source: Ref. 1,
Plate 1.)

to investigate cement concrete. Strength tests on concrete using various materials


and methods led him to the following conclusions: concrete, especially Ranger’s,
was inferior to all natural building stones and even well-made bricks; concrete
should be confined to building foundations, backing of wharf walls and the for-
mation of retaining walls; and cement should not be used for concrete because
the adhesion of mortar was reduced by a greater amount than was lime mortar
by the inclusion of sand.3 It is interesting to note that in 1836 General Joseph
Gilbert Totten of the US Army Corps of Engineers undertook experiments on
limes, cements and concrete that included artillery resistence trials on model struc-
tures in which these materials were used.4 Pasley’s fellow officer William Denison,
superintending engineer at Woolwich naval dockyard, also investigated concrete.
In 1837 Denison experimented with making concrete blocks using slightly
hydraulic lime to see if the blocks would break down under the action of sea water.
Later he sent these blocks to Pasley who performed crushing tests. Denison also
observed the performance of Ranger’s concrete both in blocks and in mass that
had been used in the construction of wharf walls at Woolwich. He concluded that
exposed concrete, especially Ranger’s, was unsuitable in marine works.5
Pasley and Denison were not alone in their views on concrete. Robert Smirke
restricted the material to building foundations.6 Another authority who concurred
was Dennis H. Mahan, professor of military and civil engineering at West Point;
his 1837 textbook was a standard reference for the US Army Corps of Engineers.7
Royal Engineers only began to change their opinions about concrete in the 1860s.
Henry Scott, instructor at the Royal Engineer Establishment and the inventor of
selenitic cement (1854), refuted Pasley’s earlier teachings by demonstrating through
experiment that higher proportions of sand weakened both lime and cement mor-
tars equally; in 1862 he offered a design for a cellular revetment in concrete.8

The forts of the At the end of the 1850s, Britain faced an arms race with France in fast moving,
increasingly manoeuvrable, ironclad warships equipped with ever more accurate
1859 Royal and powerful guns. The weak link in Britain’s defence was inadequate coastal
Commission fortifications and obsolete methods of constructing batteries. The Palmerston

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:33:02
19.qxd 15/11/2001 16:03 Page 373

Military 373

government’s response was to appoint a Royal Commission. W.F.D. Jervois, Assis-


tant Inspector-General of Fortifications, was secretary to the commission and he
largely wrote its report of 1860. Recommended were the fortification of the Royal
Navy dockyards and Woolwich Arsenal and the harbours of Portland, Dover and
Cork, and the construction of coastal batteries to resist sea attack and a ring-fortress
system to defend against land assault. The Commission forts were designed by the
Royal Engineers under the overall direction of Jervois.9
Newhaven Fort, begun in 1865, included the first use of mass concrete above
foundation level in permanent fortifications in Britain (Figure 19.2). It was
designed and constructed under Lieutenant John Charles Ardagh. Early in 1864,
Gerald Graham, Commanding Royal Engineer at Brighton, experimented with
a variety of limes and cements to determine which material would be best for mass
concrete in Newhaven’s revetments, and Scott’s selenitic cement was selected.10
The revetments contained some 20,000 m3 of concrete; specifications called for one
part Scott’s cement to one part course sand, shingle or flint.11 This accomplish-
ment had actually been anticipated the year before by Royal Engineers stationed
at Halifax, Nova Scotia. Under Lieutenant Colonel F.C. Hassard, they used
Portland cement concrete in the construction of an escarp wall at Fort Ogilvie, one
of the outlying defences of Halifax harbour.12
Except for Newhaven Fort, the Commission Forts used concrete in conventional
ways, for a seal to brick casemate arches, for floors in magazines, in foundations,
for bombproofing works from plunging fire and other minor applications.13 The
most extensive use was in Portland cement concrete blocks for the foundations of
the great ironclad coast forts of Plymouth Breakwater, Portland Breakwater, and
the Spithead Forts, Portsmouth (Spit Bank, Horse Sand, St Helens and No Man’s
Land Fort). Foundations for the Spithead Forts (Figure 19.3) were massive rings
of stone work in the sea bed executed by engineering contractors and filled in with
concrete blocks by the Royal Engineers afterwards. Horse Sand Fort, for exam-
ple, contained 15,000 t of concrete blocks weighing 3–7 t each.14 Iron shields on
the Spithead Forts had Portland cement concrete backing between the shields’
armour plate layers; in some shields, iron concrete was used.15 Portland cement
concrete was also used to fill the wrought iron pier casings which supported
the armour on the Spithead Forts.16 The first large scale use of Portland cement
concrete in fortification superstructures in the British Isles was in the defences of
Cork Harbour, beginning about 1873; this included sea walls, casemated batter-
ies, powder magazines and dwelling houses.17 The Chatham ring of land forts,
begun in 1876 but not completed until 1899, were built primarily of Portland
cement mass concrete but casemates and tunnels still had brick walls with concrete
arches.18 More or less contemporaneously Portland cement concrete was being

Figure 19.2 Plan and section


of Newhaven Fort, 1865.
(Source: Ref. 11, p. 14.)

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:33:02
19.qxd 15/11/2001 16:03 Page 374

374 Historic concrete

Figure 19.3 Elevation and


cross section for one of the
proposals for the Spithead
Sea Forts. (Source: J. Roy.
United Serv. Inst., 1869, XII,
Plate XXXV.)

employed in the Chatham Dockyard extension works,18a the dock walls being
entirely of Portland cement concrete, of varying mix proportions. Viewed in
context, the Royal Engineers were late and conservative in their adoption of struc-
tural concrete. From 1859 the US Army Corps of Engineers used natural hydraulic
cement concrete for superstructure construction, in the strategically important forts
Richmond and Tomkins at New York harbour.19

The First New technological developments in weaponry, particularly the explosive shell, led
to the replacement of brick and masonry in fortifications, first by mass concrete
World War and soon after by reinforced concrete. The French forts at Verdun were the most
advanced with thick concrete of superior quality and reinforced.20 The Germans
were the first to use reinforced concrete in fieldwork emplacements and shelters.21
Indeed, the use of reinforced concrete by Germany on the Western Front was prob-
ably the first use of the material on such a vast scale.22 In addition to this, as part
of a series of submarine shelters on the Belgian coast, Germany built at Bruges a
forerunner to the famous submarine pens of the Second World War.23
British concrete shelters were well below the standards set by the Germans.
Nonetheless, Britain’s adoption of prefabricated defences and the air-space
theory deserve comment. Corrugated iron was used for permanent formwork on
concrete emplacements. The Ryes plate, bolted together in a part-hexagon shape,
was the most sophisticated development.24 The Moir Pill Box (Figure 19.4), the
ultimate British prefabricated design, was developed at the end of the First World
War but did not see service. It was constructed from interlocking concrete blocks
and steel parts.25 Application of the air-space theory was in machine gun posts
and other shelters. In this approach, a reinforced concrete slab detonated the shell
and the air space or soft absorptive material such as earth in a concrete sandwich
construction, cushioned the blast.26

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:33:02
19.qxd 15/11/2001 16:03 Page 375

Military 375

Figure 19.4 ‘Moir’ Pill Box.


(Source: Ref. 24, p. 80.)

The inter-war Although there was no significant British military construction in the inter-war
period, notable developments elsewhere deserve brief mention. Begun in 1930,
period the Maginot Line built by France adopted the air-space theory used by the British
in the First World War. The final thickness of reinforced concrete was 3.5 m but
exposed concrete was limited to combat blocks; 1.5 million cubic meter and 150,000 t
of steel were used in this unique feat of engineering.27 The West Wall or Siegfried
Line constructed by Germany along its western border in 1938–39 drew on
the diverse experiences of the First World War concrete defences and the con-
struction organization developed by Dr Fritz Todt for the autobahn system built
in the 1930s. Within 18 months, 6 million tons of concrete were used to construct
22,000 individual works.28 Finally, with the growing threat from Japan in the
Pacific, the Americans reinforced their defences in the Philippines through forti-
fications and large, tunnelled underground installations. Construction began in
the early months of 1941. Amongst these works, the US Army Corps of Engineers
built Fort Drum on El Fraile Island in Manila Bay. Shaped like a battleship, it had
reinforced concrete with thicknesses of between 5 and 9 m.29

The Second The Second World War sparked a number of innovations in military concrete con-
struction. As early as October 1939, the Royal Engineers began building reinforced
World War concrete pill boxes for the British Expeditionary Force that featured standard
steel formwork and reinforcing bars bent and bundled into ‘pill box packs’ for
delivery to sites. The designs and methods were utilized for standard hexagonal
pill boxes in British coastal defences in 1940 and 1941.30
War also called for an immediate and drastic change in the construction of
Britain’s military barracks. Concrete became the almost necessary solution with
the shortage of timber and steel. During the winter of 1939–40, investigations were
made by the Directorate of Fortifications and Works and it adopted as the stan-
dard design for small single huts of short span, pre-cast concrete units for roof
trusses and purlins.31 Patented pre-cast unit designs were soon approved by
the War Office.32 By 1942, the British Concrete Federation had developed a largely
prefabricated standard design for a concrete hut (Figure 19.5). The design fea-
tured a lightweight raft foundation into which were fitted reinforced concrete
posts, slotted for the insertion of wall panels, and a roof of reinforced concrete
beams and breeze concrete slabs.33 Also in 1942, the Ministry of Works and Plan-
ning developed the MOWP Standard Hut which provided a basic prefabricated,
reinforced concrete frame structure in which several types of cladding could be
used (Figure 19.6).34 Finally, Major Waller designed the C’tesiphon hut that used

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:33:02
19.qxd 15/11/2001 16:03 Page 376

376 Historic concrete

Figure 19.5 British Concrete


Federation prefabricated
barrack hut design.
(Source: Ref. 33, p. 236.)

Figure 19.6 Assembly of


an MOWP prefabricated
standard hut. (Source:
Ref. 34, p. 194.)

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:33:02
19.qxd 15/11/2001 16:03 Page 377

Military 377

for the first time in hutting an all compressive thin concrete shell structure; cement
grout render of 50 mm was applied on fabric that completely avoided the use
of steel reinforcement.35 A system similar to this hut was developed during the
war by Waller and Kurt Billig for the construction of ends to blister hangers of
90 ft span.36
A great leap forward with air power created a corresponding need to protect
civilian populations as well as military installations against the devastating effects
of areal bombing. The Blitz and the public rush in London to the Underground
for protection eventually shook the British government out of its lethargy
concerning the development of effective air raid shelters. Both reinforced con-
crete and reinforced concrete block work were employed in shelter construction.
The Borough of St Pancras used an innovative constructional process using a
system of steel shuttering originally designed for building reinforced concrete
culverts on a travelling basis; sections of rectangular concrete tube were laid in
the streets to form air raid shelters.37 Some compartmental cellular shelters were
built at Finsbury despite the government’s official rejection of the design. These
were constructed in two stages. The first was a reinforced concrete slab shelter of
minimum thickness placed about 1.5 m below ground level and covered over with
earth; when the opportunity afforded itself, the initial shelter was re-excavated
and a second stage of reinforced concrete protection added.38 In contrast to
British efforts, the Germans constructed (from 1940) extensive, heavily protected
reinforced concrete bombproof bunkers for the civilian population. Original roof
slabs had 1.5 m of concrete but this was later increased to 2.5 m.39
The design and construction of Britain’s sea forts in 1941–42 stands as a cele-
brated wartime engineering achievement (Figure 19.7). Four naval forts were
erected in the North Sea, opposite the Thames estuary (Figure 19.8); six army forts
were constructed, three in the Thames estuary and three in the Mersey estuary.
A prefabricated, amphibious type of fort solved the problems of time, labour, mate-
rials and the risk of enemy attack during construction. G.A. Maunsell, a civil engi-
neer, designed the sea forts and the projects were executed under his supervision

Figure 19.7 British Naval


Fort under tow to action
station. (Source: Ref. 40,
pp. 132–62.)

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:33:02
19.qxd 15/11/2001 16:03 Page 378

378 Historic concrete

Figure 19.8 British Army


Fort being lowered to sea bed
in Thames estuary. (Source:
Ref. 40, pp. 132–62.)

together with the Civil-Engineer-in-Chief of the Admiralty; reinforced concrete


design was worked out and supervised by G. Smith, a structural engineer.40 The
naval forts comprised a reinforced concrete pontoon from which rose twin cylin-
drical reinforced concrete towers surmounted by a steel superstructure. These forts
were floated to and sunk at their respective action stations. An innovation in
construction practice was the pre-assembly of reinforcing steel in mats (slabs) and
cages (towers) that were partly wired and partly spot welded to form rigid units;
plain round bars were used and hook ends and bends avoided.41 The army forts
comprised seven separate towers interconnected by walkways. Each tower had a
hollow reinforced concrete base consisting of four members in the form of an
Oxford picture frame. At the base’s four intersections rose four hollow reinforced
concrete cylindrical legs (each leg pre-cast in three separate sections), surmounted
by a concrete cap in which main joists were embedded to support a triple-decked
steel house. Because the foundation design was not self-buoyant, army forts needed
special plant to float the bases out and deposit them on the sea bottom.
Another British accomplishment was with the amphibious ‘Phoenix’ reinforced
concrete caissons that formed the major part of the ‘Mulberry’ Harbour break-
waters used to support the D-day invasion (Figure 19.9). The ‘Phoenix’ units were
built at eight sites in Britain, and laid down in 12 specially excavated basins along
the Thames. Responsibility for design rested on a War Office team headed by
Captain W.J. Hodge that included leading designers drawn from outside.42 A fast-
track construction system based on standard designs was essential. Six types were
originally constructed with two others later for the supplementary units. Con-
struction began in December 1943 and 213 units were completed in 150 days; the
first caisson arrived at Arromanche in Normandy on D-day plus one. The caissons
were of plain symmetrical cellular reinforced concrete construction; reinforcing was
almost entirely of straight bars to save time and labour; internal walls had large
openings to lighten the structures; caissons were divided into 22 compartments and
solid concrete blocks were formed at the four corners to take towing or mooring
strains; and torsional diagonal tie bars were provided across the tops of each open
compartment (substituted by concrete roofs in supplementary units).43 A gale which
broke on June 19 and lasted for three days damaged or destroyed several of the
original units; supplementary units were strengthened accordingly.44
Notwithstanding what Britain accomplished with concrete in the Second World
War, it is important to see this in perspective. Germany lead the way in both
the scale and innovative uses of the material. Submarine pens were particularly

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:33:02
19.qxd 15/11/2001 16:03 Page 379

Military 379

Figure 19.9 ‘Phoenix’ units


for Mulberry Harbour
breakwaters.

outstanding. Construction started with the Deutsche-Werft pens at Hamburg, and


involved extensive use of precast, prestressed concrete trusses manufactured by
Wayss and Freytag. After June 1940, pens using the same system were built on
the Atlantic coast in France and other captured territories. At Brest these featured
closely spaced trusses with up to 5.5 m of concrete in a roof slab reinforced with
steel bars, and fortified with a 1.5 m ‘burster’ slab over part of the roof.45

Epilogue During and after the Second World War, British experts observed wartime bomb
damage and carried out trials on the effects of bombs and other explosive devices
on structures, including those constructed of mass concrete and reinforced
concrete. The Research and Experiments Department of the Ministry of Home
Security handled most of the work, and trials ranged from simple to sophisti-
cated structures. British researchers shared findings with their American allies.
Amongst the more interesting experiments were Anglo-American bombing trials
on the roofs of German submarine pens.46 American observations in the Pacific
theatre included those on unusual failures of reinforced concrete caused by atomic
bombs at Hiroshima and Nagasaki. Atomic bombs were detonated at a consider-
able height above the buildings and blast effects were distributed fairly evenly as
opposed to the localized patterns seen with conventional explosives.47 All the same,
hardened fortifications had become redundant in mobile war; the focus shifted
to hardened protection for civilian populations in the event of nuclear war and
to nuclear missile silos.48 The contest between concrete and weaponry was virtu-
ally over.

References 1. Harding, G.J., Description of a concrete bomb-proof erected at Woolwich with detailed
experiments as to the effect produced on it by artillery fire. Papers on subjects
connected with the duties of the Corps of Royal Engineers, 1837, 1, 33–38.
2. Pasley, C.W., Observations on Limes, Calcareous Cements, Mortars, Stuccos and
Concrete etc. London, Weale, 1838: 22.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:33:02
19.qxd 15/11/2001 16:03 Page 380

380 Historic concrete

3. Pasley, C.W., Observations on Limes, Calcareous Cements, Mortars, Stuccos and Con-
crete etc. London, Weale, 1838: 85–86, 144–45.
4. Gillmore, Q.A., Practical Treatise on Limes, Hydraulic Cements and Mortars. New York,
1874: 80, 184 –85, 252–53.
5. Denison, W., Notes on concrete. Papers on Subjects Connected with the Duties of the
Corps of Royal Engineers, 1838, 2, 266.
6. Pasley, C.W., Observations on Limes, Calcareous Cements, Mortars, Stuccos and Con-
crete etc. London, Weale, 1838: vii.
7. Mahan, D.H. and Barlow P. (eds). An Elementary Course of Civil Engineering. Edin-
burgh, MacMarton, 1845.
8. Scott, H.Y.D., Observations on limes and cements, their properties and employments.
Papers on Subjects Connected with the Duties of the Corps of Royal Engineers etc.
New Series, 1862, XI, 15–94.
9. Powter, A., Conservation of concrete in fortifications and gun batteries. Unpublished
thesis, Diploma in Conservation Studies, University of York, 1979, 40–46.
10. Graham, G., Experiments on limes and cements. Papers on Subjects Connected with
the Duties of the Corps of Royal Engineers etc. New Series, 1865, XIV, 155–61.
11. Hamilton-Baillie, J.R.E., Nineteenth century concrete and the Royal Engineers. Con-
crete, 1980, 14, 15–16.
12. Vincent, E., Substance and Practice: Building Technology and the Royal Engineers
in Canada. Canada Communications Group — Publishing, Supply and Services
Canada: Ottawa, 1993, 73–75.
13. Powter, A., Conservation of concrete in fortifications and gun batteries. Unpublished
thesis, Diploma in Conservation Studies, University of York, 1979, 40.
14. The Engineer, 1864, 18, 163.
15. The present condition of our armoured defences. The Engineer, 1881, 51, 326, 334.
16. The present condition of our armoured defences. The Engineer, 1881, 51, 326, 334.
17. Maj , M., Notes on Portland cement concrete. Papers on Subjects Connected with the
Duties of the Royal Engineers etc. New Series, 1874, XXII, 149–53.
18. Powter, A., Conservation of concrete in fortifications and gun batteries. Unpublished
thesis, Diploma in Conservation Studies, University of York, 1979, 46.
18a. Bernays, E.A., Portland cement concrete and some of its applications. Min Proc. Ice,
62, 1880, 87–95.
19. Gillmore, Q.A., Practical Treatise on Limes, Hydraulic Cements and Mortars. New York,
1874: 246.
20. Mallory, K., Ottar, A., The Architecture of War. Pantheon Books: New York, 1973: 29.
21. Mallory, K., Ottar, A., The Architecture of War. Pantheon Books: New York, 1973: 45.
22. Mallory, K., Ottar, A., The Architecture of War. Pantheon Books: New York, 1973: 51.
23. Mallory, K., Ottar, A., The Architecture of War. Pantheon Books: New York, 1973: 69.
24. Taylor, G., The general design and use of hardened defences in twentieth century
warfare. Royal Engineers Journal, 101, 1987, 87.
25. Taylor, G., The general design and use of hardened defences in twentieth century
warfare. Royal Engineers Journal, 101, 1987, 87.
26. Mallory, K., Ottar, A., The Architecture of War. Pantheon Books: New York, 1973: 59.
27. Taylor, G., The general design and use of hardened defences in twentieth century
warfare. Royal Engineers Journal, 101, 1987, 80–83.
28. Taylor, G., The general design and use of hardened defences in twentieth century
warfare. Royal Engineers Journal, 101, 1987, 87–88.
29. Dodd, K.C., The Corps of Engineers: the war against Japan. United States Army in
World War II, Technical Services. Office of the Chief of Military History, United States
Army, Washington, DC, 1966.
30. Taylor, G., The general design and use of hardened defences in twentieth century
warfare. Royal Engineers Journal, 101, 1987, 90.
31. Singer, C.M., Notes on alternative materials and methods of construction for war
hutting. Roy. Engrs J. 1940, 54, 180–87.
32. Precast unit construction for hutting. The Builder, 1941, 160, 90–91.
33. Clear span hostels of prefabricated type. The Builder, 13 March 1942, 236.
34. The M.O.W.P standard hut. J. Roy. Inst. Br. Archit. September 1942: 193–94.
35. Mallory, K., Ottar, A., The Architecture of War. Pantheon Books: New York,
1973: 197.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:33:02
19.qxd 15/11/2001 16:03 Page 381

Military 381

36. Billig, K., Concrete shell roofs with flexible moulds. J. Inst. Civil Engrs, 1945–1946,
25, 228–31.
37. Mallory, K., Ottar, A., The Architecture of War. Pantheon Books: New York, 1973: 233.
38. Mallory, K., Ottar, A., The Architecture of War. Pantheon Books: New York, 1973: 225.
39. Mallory, K., Ottar, A., The Architecture of War. Pantheon Books: New York, 1973:
223–43.
40. Posford, J.A., The construction of Britain’s sea forts. The Civil Engineer in War, Vol.
2. Institution of Civil Engineers: London, 1948: 133, 161.
41. Posford, J.A., The construction of Britain’s sea forts. The Civil Engineer in War, Vol.
2. Institution of Civil Engineers: London, 1948: 137, 161.
42. Wood, C.R.J., Phoenix. The Civil Engineer in War, Vol. 2, Institution of Civil Engi-
neers: London, 1948: 342.
43. Wood, C.R.J., Phoenix. The Civil Engineer in War, Vol. 2, Institution of Civil Engi-
neers: London, 1948: 342–46.
44. Jellett, J.H., The lay-out, assembly and behaviour of the breakwaters at
Arromanche Harbour (Mulberry ‘b’). The Civil Engineer in War, Vol. 2. Institution
of Civil Engineers: London, 1948: 300–307.
45. Mallory, K., Ottar A., The Architecture of War. Pantheon Books: New York, 1973:
252–55.
46. Walley, F., The effect of explosions on structures. Proc. Inst. Civil Engrs, Struct. Build.,
1994, 104, 325–34.
47. Thomas, W.N., The effects of impulsive forces on materials and structural members.
J. Inst. Civil Engrs, The Civil Engineer in War, 3, 1948, 72.
48. Taylor, G., The general design and use of hardened defences in twentieth century
warfare. The Civil Engineer in War, 99.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:33:02
20_app_I.qxd 16/11/2001 13:11 Page 383

Appendix I: sources of further


information on historic concrete

Concrete as a material has been in use for thousands of years yet the term does
not appear to have come into use in England until the late 1820s, having been
rediscovered by English architects and engineers a decade or so earlier. Since that
time, as with so much technical literature, published information on the use of
concrete has grown exponentially, making it very difficult for the inexperienced
researcher to identify and locate useful sources of information. This book is
intended to fill this gap, but inevitably many readers will need to research fur-
ther. In terms of further reading some indication is provided by the references
cited in individual chapters, and some authors, notably Bussell, have provided fur-
ther guidance. Having identified likely references one has still to find them, and
what follows is intended to provide guidance on the most important collections,
as well as a general reading list.
It is important to distinguish the purpose of this reading list from that of the
references at the end of each of the 19 chapters in this book which are seen as
windows opening on to further information on specific points in the text, whereas
the general reading list is intended not only to broaden the outlook of engineers
dealing with specific tasks but to appeal to historians or to the general reader who
may be unaware of the great impact of concrete on almost every aspect of life today.
No list can possibly be complete but it is hoped that this reading list will at least
act as a starting point.
With dual readership in mind the list has been split into three sections. The first,
based on a broad readership, gives some of the key historical books or articles on
the adoption of concrete as a material and on the introduction of both reinforced
and prestressed concrete. The second, intended mainly for engineers involved
in appraisals of existing construction, provides a short list of sources of informa-
tion on the historical investigation and assessment of actual concrete structures.
Most of these sources have extended bibliographies. The final section covers pub-
lished guidance on practical methods of appraising, repairing and maintaining
existing concrete structures. The contents of these three sections are summarized
below:
A General reviews of the growth of understanding of structural concrete
1 Broad historical reviews
2 Books and papers on cements, mass concrete and fireproof flooring up to
about 1900 (before the reinforced concrete period)
3 Books and papers on pioneer reinforced concrete 1900–20
4 Books and papers on reinforced concrete 1920–45
5 Books and papers on early prestressed concrete 1945–65
6 Periodical publications: 1830s to the present day
B Practical guidance on the identification of concrete structures
1 Sources of information on buildings and structures
2 General directories/guides to archives, etc.
3 Organizations holding archives on individual structures as well as general
records (possibly including drawings)
4 Organizations which may be worth consulting for general technical and his-
torical data on concrete structures
5 Websites and bulletin boards

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:33:23
20_app_I.qxd 16/11/2001 13:11 Page 384

384 Historic concrete

6 Trade literature and published synopses of details of proprietary concrete


systems
C Publications giving guidance on methods of appraisal, repair and mainte-
nance of existing concrete structures

A: General A1 Broad historical reviews


reviews of the
Texts on cement and concrete in the first half of the 19th century were descrip-
growth of tive rather than analytical, but nonetheless the better examples such as Pasley’s
understanding of provide a wealth of information on contemporary knowledge and practice. By the
structural concrete 1870s increasing data were available on the strength and chemistry of concrete.
Similarly the earliest textbooks on reinforced concrete had much useful descrip-
tive material in them — Christophe in France, and Marsh in England being obvi-
ous examples. These do contain some general analytical material, but the reinforced
concrete textbooks most would recognize today generally appear after the First
World War, consolidating a lot of early research on the continent and in the USA,
as well as design experience. For those who are able to tackle some German
the Handbuch der Eisenbetonbau — a multi-volume encyclopaedia, begun by
Emperger is a fascinating insight into continental practice — generally years ahead
of Britain. This comprehensive approach was reflected in the Concrete Publica-
tions series which began publication in the 1920s (qv).
American Concrete Institute. A selection of historic American papers on concrete
1876–1926. Detroit: ACI, 1976 (ACI Special Publication 52, includes papers
by J.O. Draffin, T. Hyatt, W.E. Ward, A.N. Talbot, A.R. Lord, C.A.P. Turner,
E.L. Ransome, A. Saurbrey and D.A. Abrams)
Hamilton, S.B. A note on the history of reinforced concrete in buildings, National
Building Studies, Special Report, 24, 1956.
Idorn, G.M. Concrete Progress from Antiquity to the Third Millennium. Telford, 1997.
Mainstone, R.J. Developments in Structural Form, 2nd edn. Architectural Press, 1998.
Revolution or evolution? Historic concrete: vital information for all involved in
carrying out appraisals and alterations. Catalogue for the exhibition at the Insti-
tution of Civil Engineers. ICE, 1996.
Stanley, C. and Bond, G. Concrete Through the Ages. BCA, 1999 (New edition of:
Stanley, C.C. Highlights in the History of Concrete. Cement and Concrete
Association, 1979.)

A2 Books and papers on cements, mass concrete and fireproof


flooring up to about 1900 (before the reinforced concrete period)

Baker, I.O. Treatise on Masonry Construction. New York, various editions.


c. 1890–1910.
Bristow, I.C. Exterior renders designed to imitate stone. Trans. ASCHB, 1997, 22,
pp. 13–30.
Grant, J. Experiments of the Strength of Cement, Chiefly in Reference to the Port-
land Cement Used in the Southern Main Drainage Works. London: Spon, 1875.
Hamilton, S.B. A short history of the structural fire protection of buildings. National
Building Studies Special Report 27, London: HMSO, 1958.
Lawford, G.M. Fireproof floors. Trans. Soc. Engrs, 1889, pp. 43–70.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:33:23
20_app_I.qxd 16/11/2001 13:11 Page 385

Appendix I: sources of further information on historic concrete 385

Pasley, C.W. Observations on Limes, Calcareous Cements, Mortars, Stuccos, and Concrete.
London: Weale, 1838.
Pasley, C.W. Observations on Limes, Calcareous Cements, … Part 1 only, 2nd edn, 1847.
Pasley, C.W. Outline of a Course of Practical Architecture … Royal Engineers Estab-
lishment, repr. 1862 of 1828 lithographed notes. Reprint in facsimile by Donhead,
2001 (see Hurst).
Potter, T. Concrete: Its Uses in Buildings. Various editions. 1870–1907.
Sutcliffe, G.L. Concrete, its Nature and Uses. London: Crosby Lockwood, 1893.
Vicat, L.J. A Practical and Scientific Treatise on Calcareous Mortars and Cements, Arti-
ficial and Natural … translated by J.T. Smith. London: J. Weale, 1837 (also later
French edn, 1865). Reprint in facsimile by Donhead, 1997.
Webster, J.J. Fire proof construction. Instn Civil Engrs, Min. Proc., 1891, CV,
pp. 249–88.
White, G.F. Observations on artificial, hydraulic or Portland cement; with an
account of the testing of the brick beam erected at the Great Exhibition, Hyde
Park, Instn Civil Engrs, Min. Proc., 1852, 11, pp. 478–502.

A3 Books and papers on pioneer reinforced concrete 1900–20

Berger, C. and Guillerme, V. La Construction en Ciment Armé. Paris: Dunod, 1902.


Christophe, P. Le Béton Armé et ses Applications, 2e edn. Paris: Béranger, 1902 (orig-
inally published in Annales des Travaux Publics de Belgique).
Coignet, E. and Tédesco, N. de. Du calcul des ouvrages en ciment avec ossature
métallique. La Societé des Ingènieurs Civils de France, Paris, 1894.
Considère, A. Experimental Researches on Reinforced Concrete. Trans. L.S.
Moisseiff. New York: McGraw-Hill, 1903.
Emperger, F. Handbuch für Eisenbetonbau. Berlin: Ernst, various editions.
c. 1908–39.
Marsh, C.F. and Dunn, W. Reinforced Concrete, 3rd edn. London: Constable, 1906
(excellent description of systems and early structures).
Marsh, C.F. and Dunn, W. Manual of Reinforced Concrete and Concrete Block Con-
struction. London: Constable, 1908 and various other editions.
Morsch, E. Concrete-Steel Construction. Trans. from 3rd (1908) edn revised and
enlarged by E.P. Goodrich. New York: Engineering News, 1909.
Tédesco, N. de and Forestier, V. Manuel Théorique et Pratique du Constructeur en
Ciment Armé. Paris: Béranger, 1909.
Twelvetrees, W.N. Concrete-Steel Buildings. London: Whittaker & Co., 1907.

A4 Books and papers on reinforced concrete 1920–45

A very large number of publications on reinforced concrete were issued in this


period amongst which the series published by Concrete Publications is outstand-
ing. A list of these prepared by Dr L.G. Booth with dates of revisions is given
below. This was first published in Proc. Instn Civ. Engrs Structs & Bldgs, 1998, 128,
pp. 398–99.
Concrete Publications advertised the ‘Concrete Series’ as ‘practical and useful
books on various branches of concrete and reinforced concrete design and con-
struction, estimating, cost-keeping, pre-cast concrete, etc.’ It appears that the Series
began in 1918 with Concrete Cottages, Bungalows, and Garages by Albert Lakeman, a

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:33:23
20_app_I.qxd 16/11/2001 13:11 Page 386

386 Historic concrete

Lecturer at Woolwich Polytechnic. During the next 50 years, until it ceased pub-
lication, the company published some 85 titles (see table), most of which went into
several editions or reprints.
For the first decade, the books tended to concentrate on the practical applica-
tions of the material, particularly pre-cast work. The results of research and the
current design of reinforced concrete beams were covered by Faber (1924) and
the design and construction of formwork by Wynn (1926). During the 1930s the
emphasis moved to design manuals (the first of the ‘Charley’ Reynolds in 1932)
and the design of special structures: e.g. Gray (1931) on reservoirs and Terrington
(1937) on arches and domes. The first of the commentaries on the codes (Scott
and Glanville in 1934) also appeared. A notable immediate post-war addition to
the list was a translation of Magnel’s pioneering work on Prestressed Concrete (1948).
Ultimate load methods were introduced by Baker (1949), who was professor of
concrete technology at Imperial College, London.
Structural analysis was never strongly represented: the well-established publishing
houses provided stiff competition. There were, however, minor contributions from
Shepley (1942), Manning (1952) and Smolira (1955), and a major (unappreciated)
contribution from Gartner (1942). Gartner’s book introduced the so-called ik (influ-
ence coefficient) method, but the full potential of the method was only fulfilled when
Jenkins introduced matrices to the structural analysis of shells and skeletal frames
and when Henderson began to spread the gospel through the Concrete Technol-
ogy diploma course (DIC) in the late 1940s at Imperial College.
Some later well-known names appeared as authors early in their careers: (Sir)
Ove Arup (1939), (Professor) A.L.L. Baker (1937), (Sir) Charles Davis (1924), (Dr)
Oscar Faber (1924) and (Sir) William Glanville (1934).
The writer’s search for a complete set of titles has been made over many years.
It is hoped that the first edition titles and dates are complete and correct (a math-
ematics book — an unusual venture — has been omitted). Some of the dates of
later editions are missing despite a thorough trawl of the book reviews in the pub-
lisher’s monthly magazine Concrete and Constructional Engineering (C&CE) and an
examination of the catalogues of various libraries. In the table opposite, titles in
the ICE Library are marked with an asterisk (*) and those in the British Library
by a dagger (†). Reprints of books have not been included nor have all reprints
of articles. The writer has not seen copies of all the titles in the table.
Care must be taken in defining the dates of the editions of a book. The title page
is never dated but since about 1950 its reverse has been dated. If the reverse is
dated, it is used to date the book: if this date is not given then the date of the pref-
ace (or foreword), is quoted and it is given in square brackets. These dates may,
or may not, be the same. A code and its commentary may have the same or dif-
ferent dates. A book may be a reprint of articles that had appeared in previous
issues of C&CE: they would usually have different dates. The nomenclature for
the different editions is also confusing. A history of printing is sometimes given
on the reverse of the title page: it is possible for a book to be described as 2nd
edn, 2nd edn revised, 2nd edn revised and enlarged, or even as reprinted.
It would be interesting to learn if anybody has a complete set. It may be that
the ICE Library with about 60 first editions has the most. Additions to the list would
be welcome.
The conventional alphabetical list has been reorganized as a tool for the
refurbishing engineer by presenting it in chronological order under 11 headings.
The method of use is to work backwards in time to the first edition devoted to the
topic of interest and then to work forwards through the later editions to the date
in question and so find the current practice. (For example, a chimney in 1965:
working backwards yields a first edition of Taylor and Turner in 1940: working
forward from 1940 yields a second edition dated 1960: both editions are in the
ICE Library (*) and the British Library (†).)

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:33:23
20_app_I.qxd 16/11/2001 13:11 Page 387

Appendix I: sources of further information on historic concrete 387

Titles published by Concrete Publications Ltd in the Concrete Series. Chronological classified list of authors, titles
and dates of editions

Bridges
Anon., Road Bridges in Great Britain, 1939*
Legat, A.W., Dunn, G., Fairhurst, W.A., Design and Construction of Reinforced Concrete Bridges, [1948*†]

Cement
Davis, A.C., A Hundred Years of Portland Cement: 1824–1924, 1924*
Kuhl, H., Cement Chemistry in Theory and Practice, 1931*
Davis, A.C., Portland Cement, 1934*; 2nd edn, 1943*
Platzmann, C.R., International Dictionary of Cement. English–French–German–Spanish, [1935*†]
Watson, W., Cradock, Q.L., Cement Chemists’ and Works Managers’ Handbook, [1940*†]; 2nd edn, 1952*†
Davis, Sir Charles, Portland Cement, 1948

Commentaries on Codes of Practice


Scott, W.L, Glanville, W.H., Explanatory Handbook on the Code of Practice for Reinforced Concrete as Recommended by the
Reinforced Concrete Structures Committee of the Building Research Board [of the DSIR], 1934*; 2nd (rev.) edn, 1939*; 2nd (further
rev.) edn, 1948*
Scott, W.L., Glanville, W.H. Thomas, F.G., Explanatory Handbook on the BS Code of Practice for Reinforced Concrete Recommended by
the Codes of Practice Committee Formed Under the Aegis of the Ministry of Works, 1950*† [commentary on CP 114: 1948 code];
1957*†; 1961*† [commentaries on CP 114: 1957 code]; 1965*† [commentary on CP 114: 1965 code]
Walley, F., Bate, S.C.C., A Guide to the B.S. Code of Practice for Prestressed Concrete: No. 115: 1959, 1961*†

Concrete making and finishes, and pre-cast products


Wynn, A.E., Andrews, E.S. Modern Methods of Concrete Making, 1926*; 2nd edn, 1928; 3rd edn, 1939*
Childe, H.L., Manufacture and Uses of Concrete Products and Cast Stone, [1927*]; 5th edn, 1930; 6th edn, 1939; title changed in 1940
to Concrete Products and Cast Stone*, 8th edn, 1949*; 9th edn, 1961
Childe, H.L., Pre-cast Concrete Factory Operation and Management, [1929*†]
Anon., Moulds For Cast Stone and Pre-cast Concrete. 1930. 1937*
Wynn, A.E., Estimating and Cost Keeping for Concrete Structures, [1930*]; 2nd edn, 1944*
Wynn, A.E., Making Pre-cast Concrete for Profit: A System of Cost-Keeping and Determining Profits, 1930*
Anon., Concrete Surface Finishes, Renderings and Terrazzo, 1935. Gray W.S., H.L. Childe, 2nd edn, 1948*
Baumgarten, R.H., Childe, H.L., Manufacture of Concrete Roofing Tiles, [1936*†]; 2nd edn, 1947
Burren, F., Gregory, G.R., Moulds for Cast Stone and Concrete Products, [1936*]
Walsh, H.N., How to Make Good Concrete, [1939*†]; 2nd edn, 1955*
Fielder, F.H., Estimating and Costing Pre-cast Concrete Products and Cast Stone, 1943*; [2nd edn, 1963*†]
Gray, W.S., Childe, H.L., Concrete Surface Finishes, Renderings and Terrazzo, 2nd edn, 1948*
Pennington, A.M., Concrete Fences, 1950*†
Childe, H.L., Concrete Finishes and Decoration, 1964*
Nesbit, J.K., Structural Lightweight-Aggregate Concrete, 1966*

Design and special structures


Faber, O., Reinforced Concrete Beams in Bending and Shear, [1924*†]
Lakeman, A., Elementary Guide to Reinforced Concrete, 1925†; 6th edn, 1930*; 12th edn, 1950†
Gray, W.S., Reinforced Concrete Reservoirs and Tanks, [1931*†]; 2nd edn, 1942*; 3rd edn, 1954*†; Revised by G.P. Manning
4th edn, 1960*†
Reynolds Chas. E., Reinforced Concrete Designers’ Handbook, [1932*†]; 2nd edn, 1939*; 3rd edn, 1946; 4th edn, 1948*;
5th edn, 1957*; 6th edn, 1964†
Adams, H.C., The Elements of Reinforced Concrete Design, [1933*†]; 3rd edn, 1947*; 1950
Gray, W.S., Reinforced Concrete Water Towers, Bunkers, Silos and Gantries, [1933*†]; 2nd edn, 1944*; 1953* (with G.P. Manning);
4th edn, 1964*, see Gray, Manning
Terrington, J.S., Design of Domes, 1937*
Terrington, J.S., Design of Arch Roofs, 1937*
Reynolds, C.E., Practical Examples of Reinforced Concrete Design, [1938*†]
Arup, O.N., Design, Cost, Construction and Relative Safety of Trench, Surface, Bombproof and Other Air-Raid Shelters, [1939*†]
Terrington, J.S., Design of Pyramid Roofs, 1939*
Lee, D.H., Design and Construction of Air-Raid Shelters, [1940*†]
Taylor, C.P., Turner, L., Reinforced Concrete Chimneys, [1940*†]; 2nd edn, 1960*†
Pennington, A.M., Concrete Farm Silos, Granaries and Tanks, [1942†*]
Fairhurst, W.A., Arch Design Simplified, [1945*†]; 1954
Baker, A.L.L., Reinforced Concrete, 1949*†
Terrington, J.S., Design of Arch Ribs for Reinforced Concrete Roofs, 1950*
Ashdown, A.J., Design of Prismatic Structures, 1951†; 2nd edn, 1958*†
Reynolds, C.E., Examples of the Design of Reinforced Concrete Buildings, 1952*†
Eriksen, B., Theory and Practice of Structural Design Applied to Reinforced Concrete, 1953*†

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:33:23
20_app_I.qxd 16/11/2001 13:11 Page 388

388 Historic concrete

Continued from previous page

Manning, G.P., Reinforced Concrete Arch Design, 1st edn, 1933*; 2nd edn, 1954*
Pennington, A.M., Concrete Farm Structures, 1954†
Baker, A.L.L., Ultimate-Load Theory Applied to the Design of Reinforced and Prestressed Concrete Frames, 1956*†
Childe, H.L., Editorial Notes from Concrete and Constructional Engineering, 1958*
Bennett, J.D., Design of Eccentrically Loaded Columns by the Load-Factor Method, 1958*
Bennett, J.D., Reinforced Concrete Members Subjected to Bending and Direct Force: Design by the Load-Factor Method, 1962*
Reynolds, C.E., Basic Reinforced Concrete Design, 2 vols, 1962*†
Gray, W.S., Manning, G.P., Watertowers, Bunkers, Silos and Other Elevated Structures, 1964*
Terrington, J.S., Turner, F.H., Design of Non-planar Concrete Roofs, 1964*†
Pannell, F.N., Design Charts for Members Subjected to Biaxial Bending and Thrust, 1966*†
Baker, A.L.L., The Inelastic Space Frame, 1967*
Manning, G.P., Concrete Reservoirs and Tanks, 1967*

Formwork and construction


Wynn, A.E., Design and Construction of Formwork for Concrete Structures, [1926*†]; 2nd edn, 1930*; 3rd edn, 1939*†; revised by
G.P. Manning, 5th edn, 1965†*
Turner, L., Lakeman, A., Concrete Construction Made Easy, [1929*]; 3rd edn, 1942†
Reynolds, C.E., Concrete Construction, [1938*†]; 2nd edn, 1945*; 3rd edn, 1967*
Childe, H.L., Introduction to Concrete Work, 1943*†; 3rd edn, 1951†
Hunter, L.E., Construction with Moving Forms, 1951*†
Disney, L.A., Steel Reinforcement, 1954*†
Gill, H.R., Concrete Formwork Designer’s Handbook, 1960*

Foundations
Baker, A.L.L., Raft Foundations: The Soil-line Method of Design, 1937; 2nd edn, 1942*; 3rd edn, 1957*†
Wentworth-Sheilds, F.E., Gray, W.S., Reinforced Concrete Piling, [1938*†]; with H.W. Evans, Reinforced Concrete Piling and Piled
Structures, 2nd edn, 1960*†
Lee, D., Sheet Piling, Cofferdams, and Caissons, 1945*
Manning, G.P., Design and Construction of Foundations, 1961*†
Szechy, K., Foundation Failures, 1961*†
Irish, K., Walker, W.P., Foundations for Reciprocating Machines, 1969*

Prestressed concrete
Magnel, G., Prestressed Concrete, 1948*†; 2nd edn, 1950*†; 3rd edn, 1954*†
Abeles, P.W., Turner, F.H., Prestressed Concrete Designer’s Handbook, 1962*†
Abeles, P.W., An Introduction to Prestressed Concrete, 2 vols, 1964*†

Roads
Anon., Concrete Roads and Their Construction, 1920*; 2nd edn, 1923*
Smith, R.A.B., Design and Construction of Concrete Roads, [1934*†]; with T.R. Grigson, 2nd edn, [1946*†]

Small buildings
Lakeman, A. (ed.), Concrete Cottages, Small Garages, and Farm Buildings, 1918†
Lakeman, A., Concrete Cottages, Bungalows, and Garages, 1918; 2nd edn, 1924*†; 3rd edn, 1932*†; title changed to Concrete
Houses and Small Garages, 4th edn, 1949*†
Structural analysis
Eriksen, B., Influence Lines for Thrust and Bending Moments in the Fixed Arch, 1939*
Shepley, E., Continuous Beam Structures, [1942*†]; 2nd edn, 1962*
Gartner, R., Statically Indeterminate Structures, [1944*]; 2nd edn, 1947*; 3rd edn, [1957*†]
Manning, G.P., Displacement Method of Frame Analysis, 1952*†
Smolira, M., Analysis of Structures, 1955*†
Rygol, J., Nomograms for the Analysis of Frames, 1957*†

Yearbook
Concrete Yearbook, 1924 onwards

* Titles held by Institution of Civil Engineers Library.



Titles held by British Library.

The history of the publishers themselves, with their office initially off Aldwych
and from the early 1920s in Dartmouth Street off Victoria Street, the consultant’s
home, should also be recorded. It is hard to think of any other house that had
such an influence on structural engineering: particularly important were the com-
mentaries on the codes and Reynolds’ books. The contribution of H.L. Childe,

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:33:23
20_app_I.qxd 16/11/2001 13:11 Page 389

Appendix I: sources of further information on historic concrete 389

who was managing editor from 1922 to 1959, as well as being a regular author,
was equally noteworthy.
It is very likely that every structural engineer who practised between 1925 and
1975 owned at least one of the dark-blue-backed series and used it heavily:
fortunately they belonged to an era when things were built to last and many of
them — books and readers — still survive.
Other relevant publications on reinforced concrete issued in this period and since
can usually be found without difficulty in technical libraries. No attempt has been
made to cover them here.

A5 Books and papers on early prestressed concrete 1945–65

The following publications together with references in Dr Francis Walley’s chap-


ter give a good idea of the state of knowledge on prestressed concrete as first intro-
duced in Britain.
Freyssinet, Eugene. Prestressed concrete: principles and applications. Instn Civ.
Engrs, J., 1949–50, 33(4), pp. 331–80.
Guyon, Y. Prestressed concrete. Contractors Record, 1955.
Prestressed concrete: proceedings of the conference held at the Institution,
February, 1949. ICE, 1949.
Magnel, Gustave. Applications of prestressed concrete in Belgium. Instn Civ. Engrs,
J., 32(6), 1948–49, pp. 161–74.
Walley, Francis. The Progress of Prestressed Concrete in the United Kingdom.
Lecture, Institution of Civil Engineers, 3 July 1962, C&CA, 1962.
As with reinforced concrete, there should be no difficulty in locating current books
and papers on prestressed concrete in technical libraries.

A6 Periodical publications: 1830s to the present day

With the introduction of reinforced concrete in the 1890s the general engineer-
ing literature of the time was soon challenged by the specialist engineering press —
both from specialist suppliers such as Hennebique supplying information to
their licensees and commercial publishers. This was a world-wide phenomenon
and in the UK the lead was taken by Concrete and Constructional Engineering, the
Transactions of the Concrete Institute (founded 1908, name changed to Institution of
Structural Engineers 1922) and Ferro-concrete, the latter being a promotional mag-
azine for L.G. Mouchel and Partners and the Hennebique system. Whilst one can-
not ignore the more general magazines, such as Engineer, Engineering, The Builder
or the proceedings of the learned societies such as the Institution of Civil Engi-
neers (founded 1818), the Society of Engineers (founded 1854), the Royal Insti-
tute of British Architects (founded 1834), the Institution of Water Engineers
(founded 1895), etc., they were not specialist publications and their coverage of
the rapid developments in reinforced concrete in the first two decades of the 20th
century was inadequate. In recognition of these developments another general
weekly, the Builder’s Journal issued special ‘Concrete and Steel’ supplements. Impor-
tant at the time they were soon overtaken by Concrete and Constructional Engineer-
ing which remained the chief trade periodical for 60 years. After the Second World
War it was joined by the more attractive architectural approach of Concrete Quar-
terly but ceased publication in 1966. Concrete, the journal of the Concrete Society
commenced publication in 1967.
Major early titles thus are:

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:33:23
20_app_I.qxd 16/11/2001 13:11 Page 390

390 Historic concrete

• Concrete and Constructional Engineering (1906–66);


• Transactions of the Concrete Institute (1909–22);
• Ferro-concrete (1909–39);
• Concrete (1967 onwards);
• Concrete Quarterly (1947 onwards);
• Civil Engineering and Public Works Review (1906–89).
Despite their often inadequate coverage of concrete, the more general architec-
tural journals published at the end of the 19th and during the 20th century often
include useful building studies and an excellent introduction to them is included
in the publication Victorian periodicals and Victorian society edited by J. Don Vann
and Rosemary T. VanArsdel, Aldershot, Scolar, 1994 and Toronto, University of
Toronto Press, 1995. The paper on architecture by Ruth Richardson and Robert
Thorne includes a bibliography, locations and a select annotated list. The titles which
follow are fully described in that list and are arranged in order of publication:
• The Architectural Magazine (1834–39);
• [Royal] Institute of British Architects’ Publications: Transactions 1835–92; Pro-
ceedings 1878–93; merged and recast as Journal of the RIBA (1894 onwards);
• Civil Engineer and Architect’s Journal (1837–68/9);
• The Surveyor, Engineer and Architect (1840–43) entitled Architect, Engineer and Sur-
veyor 1843;
• The Builder (1842/3–1966) entitled Building thereafter;
• The Architect and Building Operative (1847–68) became Architect and
Building Gazette in 1850, was then absorbed by Civil Engineer and Architects’ Jour-
nal;
• Building News (1854–1926): originally Freehold Land Times and Building News
(1854–56) continued as Building News and Architectural Review (1860–62); then
Building News and Engineering Journal (1863–1962) absorbed by Architect and
Building News, 1926;
• The Engineer (1856 onwards);
• The Builders’ Weekly Reporter (1856 onwards) absorbed Building and Engineering
Times in 1886 and renamed Builders’ Reporter and Engineering Times;
• Engineering (1866 onwards);
• The Architect (1869 onwards) entitled The Architect and Contract Reporter
(1893–1918) then Architect and Building News (1919–70; new series 1971–80);
• The British Architect (1874 onwards) absorbed by The Builder in 1919;
• Illustrated Carpenter and Builder (1877 onwards);
• Architectural Association Journal (1887 onwards) entitled Notes (1887–1905);
• The Builder’s Journal and Architectural Record (1895–1905); became in succession:
The Builders Journal and Architectural Engineer (1906–1909); The Architects and
Builders Journal (1910–19); and then The Architects Journal (1919 onwards);
• The Architectural Review (1896 onwards).

B: Practical B1 Sources of information on buildings and structures


guidance on the Listed below are some of the sources which can be used when trying to track down
identification of the records of a building or structure:
concrete
• Local Authorities:
structures — Building Control Department;
— Architects Department;
— Planning Department;
• Members of the original design team or their successors: architects; engineers;
surveyors; quantity surveyors; party wall surveyors; contractors; specialist sub-

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:33:23
20_app_I.qxd 16/11/2001 13:11 Page 391

Appendix I: sources of further information on historic concrete 391

contractors. (For many buildings and structures, the designers’ etc. names may
be found in contemporary technical periodicals.)
• Owners: current; previous; original, e.g. Railtrack for railway structures;
• Occupiers: current; previous; original;
• Owners’ and occupiers’ professional advisers: solicitors; managing/estate agents;
architects; engineers; surveyors, etc.;
• Maintenance manuals and Health and Safety files;
• Public utilities and statutory undertakers;
• Public libraries:
— Reference libraries;
— Local studies libraries;
• Local museums;
• Local history societies;
• Regional record offices such as the London Metropolitan Archives;
• National record offices such as the Public Record Office; English Heritage
National Monuments Record;
• Libraries with national coverage such as the Institution of Civil Engineers,
the Institution of Structural Engineers, the Royal Institute of British
Architects;
• Museums with national coverage such as the Science Museum and its library;
the Victoria and Albert Museum (housing the RIBA Drawings Collection);
• Museums with subject coverage such as The Museum of Docklands.

B2 General directories/guides to archives, etc. (most published


works will be available in large public libraries)

The Aslib Directory of Information Sources in the United Kingdom, edited by


Keith W. Reynard and Jeremy M.E. Reynard, 11th edn. Aslib, 2000 (biennial).
A guide to libraries, record offices and organizations with contacts, information
on coverage, special collections and publications.

The Directory of Museums and Special Collections in the United Kingdom,


edited by Peter Dale, 2nd edn. Aslib, 1996.
Similar in format to the Aslib Directory.

Directory of British Associations & Associations in Ireland, 14th edn. CBD


Research Ltd, 1998.
With brief details on each organization: contact details, information resources,
publications.

The Business Archives Council


101 Whitechapel High Street London E1 020 7247 0024.
Does not hold archives or carry out research. Publishes a journal which includes
location lists of archives deposited in public collections.

National Register of Archives


Quality House, Quality Court, Chancery Lane, London WC2A 1HP; Tel: 020 7242
1198; website: www.hmc.gov.uk
A source of information on archives held outside public collections maintained by
the Royal Commission on Historical Manuscripts.
Visitors can consult the indexes in person, or searches can be carried out via the
website. Telephone enquiries are not undertaken, although written enquiries are.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:33:23
20_app_I.qxd 16/11/2001 13:11 Page 392

392 Historic concrete

Record Repositories in Great Britain, Royal Commission on Historical Manu-


scripts. TSO (updated at intervals).
Guide to publicly available record collections in local and national record offices,
universities and research centres.

The Building Museums Guide, Interbuild Publications for the Building


Museum Project, 1993
Now very dated but still useful. Brief details of each museum are given.

Museums around the UK on the Web can be found at www.mda.org.uk/vlmp/

B3 Organizations holding archives on individual structures as well as


general records (possibly including drawings)

Cadw
Welsh Historic Monuments
Brunel House, 2 Fitzalan Road, Cardiff, Wales CF2 1UY; Tel: 01222 465111.
Publications and documentation on histotic buildings in Wales.

Concrete Archive
See Institution of Civil Engineers Library and Archives.

Construction Industry Resource Centre Archive (CIRCA)


Curator: John Keenan, 3 Cranleigh Gardens, Stoke Bishop, Bristol BS9 1HD; Tel:
0117 968 7850.
Aims to collect drawings and documentation including text books and trade lit-
erature which might otherwise be lost.

English Heritage: National Monuments Record Centre


Main offices: Kemble Drive, Swindon, Wilts SN2 2GZ; Tel: 01793 414600; Lon-
don search room: 55 Blandford Street, London W1H 3AF; Tel: 020 7208 8200;
website: www.english-heritage.org.uk
Drawings, photographs, publications and documentation on historic buildings of
all ages.

Historic Scotland
Longmore House, Salisbury Place, Edinburgh EH9 1SH; Tel: 0131 668 8668;
website: www.historic-scotland.gov.uk
Publications and documentation on historic buildings in Scotland.

Institution of Civil Engineers Library and Archives


Great George Street, London SW1P 3AA; Tel: 020 7222 7722;
website: www.ice.org.uk
The ICE is the oldest established library in the construction industry, and it has
material on the use of cements and concrete reflecting this. These traditional
strengths have been enhanced by the development of the ‘concrete archive’ with
the support of the Concrete Society. In addition to published material, drawings,
specifications, calculations in the original and microfilm, have been placed in the
archive, together with samples of concrete and reinforcement.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:33:23
20_app_I.qxd 16/11/2001 13:11 Page 393

Appendix I: sources of further information on historic concrete 393

London Metropolitan Archives (formerly the Greater London Record Office)


40 Northampton Road, London EC1R 0HB; Tel: 020 7332 3820; Fax: 020 7833
9136; website: www.cityoflondon.gov.uk/archives/lma
No entry formalities, but there is usually a wait for documents.

Public Record Office


Ruskin Avenue, Kew, Surrey, TW9 4DU; Tel: 020 8876 3444;
website: www.pro.gov.uk
Holds the official records for all government owned buildings in the United King-
dom and much material deposited by government departments dealing with build-
ings. Indexes can be searched on-line. A reader’s ticket is required and can be
obtained on site. There is usually a wait for documents.

Royal Institute of British Architects


66 Portland Place, London W1B 1AD; Tel: 0906 302 0400 (premium rate public
information line); website: www.architecture.com
The early users of concrete were often architects rather than engineers, and whilst
modern textbooks on reinforced concrete tend to focus on its engineering analy-
sis, it is a basic building material so, inevitably, the RIBA Library has an impor-
tant collection covering developments from earliest times to the present day. There
is a charge for use of the Library but the library catalogue (covering accessions
since 1984) is available online.
The Drawings Collection of the RIBA is being rehoused and will be available in
the Victoria & Albert Museum from 2004.

Science Museum
Exhibition Road, London SW7; Tel: 020 7942 4000; website: www.nmsi.ac.uk
The Museum has a collection of artefacts connected with construction and its
Library has a good collection of technical journals, publications and trade literature.

Transport Research Laboratory


Old Wokingham Road, Crowthorne, RG45 6AU; Tel: 01344 773131;
website: www.trl.co.uk
Originally established in 1933 as the Road Research Laboratory, the Government
research body producing specifications and guidance on roads and pavements,
bridges, etc.

B4 Organizations which may be worth consulting for general


technical and historical data on concrete structures

Amberley Museum (formerly known as the Chalk Pits Museum)


Amberley, Arundel, West Sussex BN18 9LT; Tel: 01798 831370;
website: www.amberleymuseum.co.uk
Houses a display on the use of concrete, including exhibits and artefacts on loan
from the Science Museum.

British Cement Association


Century House, Telford Avenue, Crowthorne RG45 6YS; Tel: 01344 762676;
website: www.bca.org.uk

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:33:23
20_app_I.qxd 16/11/2001 13:11 Page 394

394 Historic concrete

The BCA, successor to the Cement and Concrete Association which for 50 years
was the major source of information and research on cement and concrete, houses
an important specialist collection, not only of its own published literature, but
on cement and concrete generally, together with an impressive archive of audio-
visual material. Although it charges for its services it is an essential and helpful
resource.

British Library
96 Euston Road, London NW1; Tel: 020 7412 7676; website: www.bl.uk
Coverage is national and includes books, maps and patents. A reader’s ticket is
required and can be obtained on site. There is usually a wait for documents, but
this can be avoided if they are requested before visiting.

British Standards Institution


389 Chiswick High Road, London W4 4AL; Tel: 020 8996 9001;
website: www.bsi-global.com
For copies of current and earlier codes and standards. Their online catalogue shows
their status.

Building Research Establishment


Garston, Watford WD2 7JR; Tel: 01923 664000; website: www.bre.co.uk
The BRE was for approximately 80 years the Government’s centre of excellence
for building research, a position which enabled it to collect information on con-
crete on a world-wide basis. Their own publications are a mine of useful infor-
mation and pioneering research on concrete in the early years. For material eluding
you elsewhere it is still worth trying.

Construction History Society


c/o The Chartered Institute of Building, Englemere, King’s Ride, Ascot, Berks
SL5 8BJ; Tel: 01344 23355; website: www.construct.rdg.ac.uk/chs/
A grouping of people interested in the history of construction.
Publishes a Journal which includes abstracts of periodical literature. Newsletter
includes a digest of major accessions to Repositories relating to building and con-
struction from the Royal Commission on Historical Manuscripts annual listing.
Newsletter may be used to request information of members.

English Heritage (Headquarters)


23 Savile Row, London W1X 1AB; Tel: 020 7973 3000;
website: www.english-heritage.org.uk
The head office of English Heritage. Source for regional office addresses,
publications.

Institution of Structural Engineers Library


11 Upper Belgrave Street, London SW1X 8BH; Tel: 020 7235 4535;
website: www.istructe.org.uk
With its origins as the Concrete Institute the IStructE is particularly strong in the
development of structural concrete from the late 19th century onwards. Whilst
material may also be held at ICE, IStructE often has material which has disap-
peared from ICE over the years, particularly standards.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:33:23
20_app_I.qxd 16/11/2001 13:11 Page 395

Appendix I: sources of further information on historic concrete 395

The Lime Centre


Long Barn, Morestead, Winchester, Hants SO21 1LZ; Tel: 01962 713636;
website: www.thelimecentre.co.uk

Newcomen Society
Science Museum, Exhibition Road, London SW7 2DD; Tel: 020 7371 4445;
www.newcomen.demon.co.uk
A society for the study of engineering and technology.

Twentieth Century Society


70 Cowcross Street, London EC1M 6BP; Tel: 020 7250 3857;
website: www.c20society.demon.co.uk

Victorian Society
1 Priory Gardens, Bedford Park, London W4 1TT; Tel: 020 8994 1019;
website: www.victorian-society.org.uk

B5 Websites and bulletin boards

Website addresses have been included for many of the organizations listed. How-
ever, it must be stated that many organizations change their servers and their
addresses and therefore the accuracy of the addresses given cannot be assured for
the future. Being able to carry out searches at a distance is extremely convenient
and many more organizations will doubtless soon have their own sites with links
from them to other related sites.
The online catalogues of major universities and national libraries such as the
British Library and the Library of Congress are easily accessible and are excellent
guides to published information.
When the major sources have been exhausted, a medium which should not be
ignored is bulletin boards on the internet. Many organizations and groups which
have a website also have bulletin boards on which enquiries can be posted. These
may form part of a service to members on a professional institution’s website or
may be open to a wider participation.
Links from professional institutions websites can also be a more efficient way of
searching the internet than simply using the major search engines. However,
amongst the major search engines, the Google site, which scans and rates other
sites, is an excellent place to start: www.google.com

B6 Trade literature and published synopses of details of proprietary


concrete systems

Useful as textbooks and periodicals are, they rarely provide details on the many
commercial reinforced and prestressed concrete systems employed in buildings
and other structures from the middle of the 19th century onwards. Details of
these must be sought in trade literature issued by the patentees or their licensees.
Unfortunately, the survival of such material is unpredictable, although better
for the more recent technology of prestressed concrete; firms tend to discard
old trade literature as new revised literature arrives, whilst most libraries have
never collected promotional literature, unless it has a high technical content. Well-
organized firms with a long history of involvement in reinforced concrete such as
L.G. Mouchel and Partners have excellent collections, but for every Mouchel there

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:33:23
20_app_I.qxd 16/11/2001 13:11 Page 396

396 Historic concrete

are dozens of firms who have ceased trading or reorganized so frequently as to


destroy the integrity of their records. In the past the Science Museum has collected
trade literature, but this collection contains relatively little on reinforced concrete.
In these circumstances one can pursue information contained in original patents,
obtainable in the British Library or the various local depositories, or seek busi-
ness records in libraries and record offices local to the firms concerned.
Some information on systems can be obtained from Specification (published annu-
ally since 1898); compilations such as Diamant’s on Industrialised Buildings of the 1960s,
the BCSA publications on Prefabricated Floors for Use in Steel Framed Buildings (1964,
1965, 1977), and various other 19th and early 20th century texts on concrete and
fireproof construction. For the inter-war period Jaggard and Drury’s Architectural
Building Construction, 3 vols, CUP, 1916–47, is a good starting point. Unfortunately
these rarely provide ‘safe load tables’ or the like and if they do, they should be
applied with caution as the criteria used for testing are unlikely to be known.
There is good coverage of prestressing systems incorporating hac and used in the
1960s in Addendum 2 of Building Regulations Advisory Committee Report by Sub-
committee P — High Alumina Cement Concrete, Dept of the Environment, 1976.
The Institution of Civil Engineers Library (ICE) is a good first point of contact
for finding trade brochures, followed by the Institution of Structural Engineers
(IStructE), The British Library (BL) and the Science Museum Library.
Trade literature can be found in these major libraries as follows:
Reinforced concrete systems
• Mouchel-Hennebique: Betón Armé (ICE some issues; BL) Ferro-concrete (ICE
incomplete; BL);
• Monier: 19th century brochures (ICE; BL);
• BRC: Early brochures and journals (ICE);
• Considere: Early brochure (ICE);
• Coignet systems: Early brochure (IStructE);
• Expanded metal: No known source for 19th century; various booklets (ICE;
IStructE);
• Indented Bar Company: Brochures (IStructE, complete set; ICE some);
• Kahncrete/Truscon (TrussedConcrete Steel): Kahncrete periodical (BL, 1 issue
ICE);
• Kleine (ICE archives);
• Ritchie system (ICE archives);
• Truscon leaflets (ICE, incomplete; IStructE).
Prestressed concrete: the early systems are described by Francis Walley
• PSC-Freyssinet: Brochures (ICE; IStructE); House journal (ICE, incomplete,
IStructE);
• Gifford-Udall: Book (ICE; IStructE);
• Hoyer;
• Laingspan: Brochures (ICE; IStructE);
• Lee-McCall/McCalloy: Brochures (ICE; IStructE);
• Rylands: Brochures (ICE; IStructE).
Piling systems
• See also ‘reinforced concrete systems’ above;
• Frankipile: Books (ICE; IStructE);
• Lee-McCall: (ICE).

This select short list is intended to provide sources for engineers and others
involved with existing concrete structures (principally buildings).

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:33:23
20_app_I.qxd 16/11/2001 13:11 Page 397

Appendix I: sources of further information on historic concrete 397

C: Publications C1 General: investigation and appraisal


giving guidance on The Institution of Structural Engineers. Appraisal of Existing Structures. ISE, 2nd
the investigation, edn. 1996.
appraisal, repair Building Research Establishment. Digest 366. Structural Appraisal of Existing Build-
and maintenance ings for Change of Use. BRE, 1991.
of existing Beckmann, P. Structural Aspects of Building Conservation. McGraw-Hill, 1995.
concrete Health & Safety Executive. Guidance Note G58. Evaluation and Inspection of Build-
structures by ings and Structures. HMSO, 1990.
Michael Bussell Institution of Structural Engineers. Guide to Surveys and Inspections of Buildings and
Similar Structures. ISE, 1991.

C2 General: repair and maintenance

Currie, R.J., Robery, P.C. Report BR254. Repair and Maintenance of Reinforced Con-
crete. BRE, 1994.
Fédération Internationale de la Précontrainte. Inspection and Maintenance of Re-
inforced and Prestressed Concrete Structures. FIP, 1986.

C3 Information on original construction

The Building Research Establishment has published numerous reports and pam-
phlets on proprietary, prefabricated and large-panel housing systems; these
describe the construction, defects and inspection of systems used in the UK since
the Second World War.

C4 Investigation, testing and repair of concrete


General
Bungey, J.H. Technical Note 143. Testing Concrete in Structures: A Guide to Equip-
ment for Testing Concrete in Structures. CIRIA, 1992.

Strength
British Standards Institution. BS 6089: Guide to Assessment of Concrete Strength in
Existing Structures. BSI, 1981.
Concrete Society. Technical Report No 11. Concrete Core Testing for Strength. Con-
crete Society, 1987.
Watkins, R.A.M., McNicholl, D.P. Statistics applied to the analysis of test data from
low-strength concrete cores. Struct. Engr., 1990, 68(16), pp. 327–32.

Cracking
Concrete Society. Technical Report No 22. Non-structural Cracks in Concrete. Con-
crete Society, 1992.

Corrosion
Building Research Establishment. Digest 444. Corrosion of Steel in Concrete:
Part 1: Durability of Reinforced Concrete Structures; Part 2: Investigation and
Assessment; Part 3: Protection and Remediation. BRE, 2000.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:33:23
20_app_I.qxd 16/11/2001 13:11 Page 398

398 Historic concrete

Building Research Establishment. Digest 434. Corrosion of Reinforcement in Concrete:


Electrochemical Monitoring. BRE, 1998.
Building Research Establishment. Digest 389. Concrete Cracking and Corrosion of Rein-
forcement. BRE, 1993.
Lea, F.M. Building Research Technical Paper 7. Slag, Coke, Breeze and Clinker as
Aggregates. HMSO, 1929.
Brady, F.L. Building Research Special Report 15. The Corrosion of Steel by Breeze
and Clinker Concretes. HMSO, 1930.
Concrete Society. Technical Report No 26. Repair of Concrete Damaged by Re-
inforcement Corrosion. Concrete Society, 1984.
Pullar-Strecker, P. Corrosion Damaged Concrete: Assessment and Repair. CIRIA/
Butterworth, 1987.

Carbonation
Building Research Establishment. Digest 405. Carbonation of Concrete and its Effect
on Durability. BRE, 1995.
Parrott, L.J. Report BR114. A Review of Carbonation in Reinforced Concrete. Build-
ing Research Establishment/Cement and Concrete Association, 1987.
Roberts, M.H. Information Paper 6/81. Carbonation of Concrete Made with Dense Nat-
ural Aggregates. BRE, 1981.

Chlorides
Roberts, M.H. Information Paper 21/86. Determination of the Chloride and Cement
Content of Hardened Concrete. BRE, 1986.

High alumina cement


Building Research Establishment. Digest 392. Assessment of Existing High Alumina
Cement Concrete Construction in the UK. BRE, 1994.
Bate, S.C.C. Report BR235. High Alumina Cement Concrete in Existing Building Super-
structures. HMSO, 1984, repr 1992.
Dunster, A.M. and others. Report BR386. Durability of Ageing High Alumina
Cement Concrete. A Literature Review and Summary of BRE Research Findings. CRC, 2000.

Alkali silica reaction


Institution of Structural Engineers. Structural Effects of Alkali–Silica Reaction: Tech-
nical Guidance on the Appraisal of Existing Structures. ISE, 1992.
Hobbs, D.W. Alkali Silica Reaction in Concrete. Thomas Telford, 1988.

Mundic
Royal Institution of Chartered Surveyors. The Mundic Problem: A Guidance
Note: Recommended Sampling, Examination and Classification Procedure for Suspect Con-
crete Building Material in Cornwall and Devon. RICS Books, 1994.

Load testing
Building Research Establishment. Information Paper 2/95. Guidance for Engineers
Conducting Static Load Tests on Building Structures. BRE, 1995.
Building Research Establishment. Digest 402. Static Load Testing: Concrete Floor and
Roof Structures within Buildings. BRE, 1995.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:33:23
20_app_I.qxd 16/11/2001 13:11 Page 399

Appendix I: sources of further information on historic concrete 399

Fire damage
Concrete Society. Technical Report No 33. Assessment and Repair of Fire-
Damaged Concrete Structures. Concrete Society, 1990.

Surface treatment
Construction Industry Research And Information Association. Technical Note 130.
Protection of Reinforced Concrete by Surface Treatments. CIRIA, 1987.

Bridges
Mallett, G.P. Repair of Concrete Bridges: A State of the Art Review on the Repair, Main-
tenance and Protection of Concrete Bridges. Telford, 1994.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:33:23
21_app_II.qxd 16/11/2001 13:04 Page 401

Appendix II: 19th century proprietary


floor systems

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:34:26
21_app_II.qxd 16/11/2001 13:04 Page 402

402 Historic concrete

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:34:26
21_app_II.qxd 16/11/2001 13:04 Page 403

Appendix II: 19th century proprietary floor systems 403

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:34:26
21_app_II.qxd 16/11/2001 13:04 Page 404

404 Historic concrete

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:34:26
21_app_II.qxd 16/11/2001 13:04 Page 405

Appendix II: 19th century proprietary floor systems 405

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:34:26
22_app_III.qxd 16/11/2001 16:42 Page 407

Appendix III: proprietary reinforcing


and flooring systems
Introduction This appendix describes and illustrates many of the proprietary reinforcing and
flooring systems used in Britain during the greater part of the 20th century, exclud-
ing more recent developments. It aims primarily to help those dealing with
concrete structures to identify such reinforcement and flooring systems. Identifi-
cation is a necessary preliminary to seeking further information on the particu-
lar system, using specific chapter references and the more comprehensive advice
on information sources listed in Chapters 3–5 and in Appendix E.
In addition, it is hoped that the general reader will find interest in the sheer
variety of such systems, as compared with the relatively limited range of rein-
forcement and flooring types in common use today.
The systems listed here include most of those known to have been in general
use, but certainly not all of the many systems that were patented, advertised, and
promoted. Quite a few of these, particularly from the earlier part of the century,
in practice were used only rarely, and some not at all.
As explained in Chapter 4, the early concrete specialists had their own rein-
forcing systems, and often provided much more, culminating in what today can
be described as a complete ‘design and build’ service, as exemplified by the Hen-
nebique system. Reinforcement for such systems came in many shapes and sizes
in the early 20th century, before patents expired and the round or square bar
became the norm for reinforcement, specified by designers and used by contrac-
tors without the constraints and costs imposed by patent protection. Such changes
in reinforcement types are described more fully in Chapter 5.
Early 20th century flooring systems evolved from those of the second half of
the 19th century, which saw the development, patenting, and use of many so-called
‘fireproof ’ floor systems. The commonest of these are described in Chapter 3,
which also lists some specific references to aid their identification.
Such floors commonly used beams or joists of cast or wrought iron (later steel)
at relatively close spacing, between which spanned infill blocks, timber joists or
lathing, or flat or shallow-vaulted (but essentially unreinforced) concrete. The
blocks, of fired clay, or clinker or breeze concrete, were often hollow to save both
weight and cost. They often had ‘tongues’ at either end; these butted together to
protect the beam flanges from fire. Alternatively, the timber joists and the beams
were fire-protected by plastering the floor soffit. Concrete might be laid as ‘top-
ping’ over the beams and blocks to form a level floor.
These forms of flooring construction continued into the 20th century. However,
as the use of reinforced concrete increased, the infill blocks or timbers were
replaced by longer-span proprietary precast concrete floor units, themselves often
used in conjunction with infill blocks spanning between the units, and usually with
an in situ concrete topping. In situ concrete slabs were a widely-used alternative.
These were typically solid for shorter spans and ribbed for longer spans, the lat-
ter often employing hollow blocks or ‘pots’ to infill the spaces between the ribs.
The supporting beams and columns could be of steel, or equally of reinforced con-
crete.
Many building contractors produced their own flooring units and – together
with specialist precast concrete manufacturing firms – offered ‘off-the-shelf ’ stan-
dard units for floor, roof, and stair construction as well as other elements (includ-
ing architectural cladding).
The remainder of this appendix comprises three illustrated lists. The first two,
of early reinforcing systems and early-20th century flooring systems, are

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:35:21
22_app_III.qxd 16/11/2001 16:42 Page 408

408 Historic concrete

arranged alphabetically. To aid identification, each particular system is illustrated,


described, and/or cross-referred to similar systems. The third list, of mid-20th cen-
tury flooring systems, is arranged within generic types of floor construction, in
view of the proliferation of such systems; to describe and illustrate each is beyond
the available space in this volume. For each type, a representative proprietary floor
is illustrated, and the larger or more widespread manufacturers are listed.

Early
reinforcement
systems

Figure III.1. Bonna (also known as Columbian or Feresco): (a) sizes of ‘Latin Cross’ sec-
tions used in building frames; (b) ‘Double Cross’ or ‘Cross of Lorraine’ and also ‘triple
cross’ bars used in floors (see also Figure III.29)

Figure III.2. Coignet: (a) spiral links in columns; (b) stirrups sometimes formed of spiral
rod wired to top and bottom bars; (c) bent-up bars used in small diameters as shear
reinforcement; (d) beam and slab; (e) column. Columbian: see Bonna, Figure III.1

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:35:21
22_app_III.qxd 16/11/2001 16:42 Page 409

Appendix III: proprietary reinforcing and flooring systems 409

Figure III.3. Considère: showing spiral binding to main bars

Figure III.4. Types of Corr or Corrugated bars (Indented and


Johnson similar or identical); some were early deformed Figure III.5. Cottancin: woven interlinked mat of small diam-
bars of high carbon steel with a yield strength of eter rods; system closer to ferro-cement than ‘true’ rein-
65,000–70,000 lb/in2 (448–482 N/mm2) (see also Unit, Figure forced concrete, with thin slabs reinforced by ribs often
III.24) acting as folded plates

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:35:21
22_app_III.qxd 16/11/2001 16:42 Page 410

410 Historic concrete

Figure III.7. Expanded metal (Expamet): John F. Golding


patented ‘slashed metal screening’ in 1883 in the USA. Fer-
esco: See Bonna, Figure III.1

Figure III.6. Diamond: rolled ribbed bar with constant


cross-sectional area

Figure III.8. Hennebique: very widely used system, with round main bars with ‘fishtail’
ends, thin slabs and walls, distinctive deep narrow beams, often square columns, and
chamfers on beam and column arrises: (a) stirrup round tension bar; (b) tension bars
and stirrup; (c) bottom bars cranked up to form top steel over supports in continuous
spans; (d) stirrups and links originally flat strips, later round bars; (e) columns linked by
‘sausage’ bars around main bars in some cases

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:35:21
22_app_III.qxd 16/11/2001 16:42 Page 411

Appendix III: proprietary reinforcing and flooring systems 411

Figure III.9. Hyatt: early American reinforcement, flat bars pierced with round bars, used
in slabs and later in beams (also known as Jackson bars)

Figure III.10. Hy-rib: also known as herring-bone metal lath; folded expanded metal that
could act as both permanent formwork and reinforcement Improved Construction: plain
round bars with spiral shear reinforcement (not widely used). Indented: see Corr, Fig-
ure III.4. Jackson: see Hyatt, Figure III.9. Johnson: see Corr, Figure III.4

Figure III.11. Johnson’s lattice: triangular or rectangular woven steel fabric

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:35:21
22_app_III.qxd 16/11/2001 16:42 Page 412

412 Historic concrete

Figure III.12. Kahn (also Truscon): unique ‘trussed bars’ with fins rolled on opposite cor-
ners of plain square bar that could be bent up to improve anchorage: (a) paired fins;
(b) staggered fins; (c) typical bar profile; (d) larger bar section; (e) a later and more con-
ventional profile was the Kahn cup bar, with yield strengths 33,000–37,000 lb/in2 (227–255
N/mm2); (f) fins bent at 45° as beam stirrups; (g) and (h) fins bent at 90° as column links

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:35:21
22_app_III.qxd 16/11/2001 16:42 Page 413

Appendix III: proprietary reinforcing and flooring systems 413

Figure III.13. Kahn expanded metal: a sheared expanded rolled profile with the main bars
normal to the secondary steel (unlike expanded metal, cf. Figure III.7)

Figure III.14. Keedon: stirrups and links mechanically ‘keyed on’ to main bars by wedges inside fully enclosing loops (a)
stirrups inclined to improve shear resistance in beams; (b) some bottom bars bent up in continuous spans; (c) column
links

Figure III.15. ‘Lock-woven’ or ‘lock-tied’ mesh: (a) wires secured by mechanically-fixed


wires or (b) by staples, in lieu of welding to provide an anchored mesh system

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:35:21
22_app_III.qxd 16/11/2001 16:42 Page 414

414 Historic concrete

Figure III.16. Monier: plain round bars, in beams bent up over continuous supports; more
commonly used in Germany, where known as Wayss und Freytag system

Figure III.17. Monolith: a grooved bar into which angled stirrups could be fitted, being
held in place by swaging
(a)

(b)

Figure III.18. Ransome: (a) the archetypal square-twisted cold-worked bar (also
hot-rolled), with a yield strength of 53,000 lb/in2 (379 N/mm2) or better, much employed
in the USA; (b) early use of coffered slab construction for longer slab spans

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:35:21
22_app_III.qxd 16/11/2001 16:42 Page 415

Appendix III: proprietary reinforcing and flooring systems 415

Figure III.19. Ridley-Cammell: corrugated ‘dovetail’ steel sheeting [sic.], with angle bars
as main steel in beams, and vertical sheeting panels for shear reinforcement riveted or
bolted to main steel; sheeting also used in columns and walls; served also as perma-
nent formwork where required

Figure III.20. Spiral: similar to Ransome, but with corner ‘beads’, with a yield strength
65,000–80,000 lb/in2 (448–551 N/mm2)

Figure III.21. Square twisted Lug: a further deformed (if not tortured) version of the Ran-
some square-twisted bar with additional lugs to improve bond

(a) (b)

Figure III.22. Thacher: deformed bar; (a) constant cross-sectional area without sharp re-
entrant corners; (b) an early form was of flat strip with indentations

Figure III.23. Truscon: a deformed bar with notches to accommodate washers to provide mechanical anchorage; also
used in the Kahn reinforcement system (see Figure III.12)

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:35:21
22_app_III.qxd 16/11/2001 16:42 Page 416

416 Historic concrete

(a)

(b)

Figure III.24. Unit: (a) rolled bars (two or four) within linking web; (b) prefabricated reinforcement system using first Corr
bars (see Figure III.4), later plain round bars

Figure III.25. Universal: an indented bar (not widely used)

(a) (b)

(c)

(d)

Figure III.26. Wells: (a) reinforcement profile originally a single round bar profile with a
flat continuous fin, which could be perforated to accommodate transverse bars (spac-
ers in beams, links in columns, etc); (b) later developed into twin round bar section joined
in an hour-glass profile; (c) perforated to receive indented or perforated hanger bars;
(d) or slit longitudinally allowing one bar to be cranked up over supports as anti-crack
steel; the system itself later used part-precasting of beam soffits as permanent form-
work, with roughened top surface for keying to in situ topping

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:35:21
22_app_III.qxd 16/11/2001 16:42 Page 417

Appendix III: proprietary reinforcing and flooring systems 417

Figure III.27 Williams: filler joist floors reinforced with draped transverse steel rods

Early 20th century


flooring systems

Figure III.28. Armoured Tubular: (a) precast concrete ribs with corrugated strip rein-
forcement; (b) fired clay, or coke or breeze concrete hollow infill blocks; also known as
Herbst floor. Caxton: see Figure III.35

Figure III.29. Columbian (known also as Feresco, using Bonna bars, see Figure III.1): distinctive cross-profiled rolled bars
used as filler joists: (a) and (b) supported on hangers or (c) by bolting to beams; (d) used to form single and (e) double
floors, the latter giving excellent sound-proofing and an early ‘services void’; (f) bars in beams were spaced and frame-
work could be supported by cross braces

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:35:21
22_app_III.qxd 16/11/2001 16:42 Page 418

418 Historic concrete

Figures III.30, III.32, III.34,


and III.35 are reproduced
from Jaggard, W. R. and
Drury, F. E., Architectural
Building Construction,
Volume 3, 1937 (2nd edn) by
courtesy of the publishers,
Cambridge University Press.
Figure III.30. Cullum: variant of Kleine floor (see Figure III.35) with sound-proof raised
timber floor)

Figure III.31. Dentile: triangular clay mitre tiles interlocked to form voids within one-way
or two-way spanning floor: (a) section; (b) plan on part of section; (c) single tile

Figure III.32. Diespeker: variant of Kleine floor (see Figure III.35)

Figure III.33. Fawcett (Monolithcrete): unusual floor with early perforated-web steel
beams with enlarged bottom flange and transverse reinforcement from twisted flat bars
(a form of filler joist construction, successor to the Fawcett hollow clay block and
iron/steel joist system, see Chapter 3)

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:35:21
22_app_III.qxd 16/11/2001 16:42 Page 419

Appendix III: proprietary reinforcing and flooring systems 419

Figure III.34. Helicon: hollow tile floor similar to Kleine (see Figure III.35). Herbst: see
Figure III.28

Figure III.35. Kleine: early hollow clay tile floor, first used in Britain by Higgs and Hill
c.1905; (a) reinforcement was flat steel on edge in cement mortar joints between blocks
with coke-breeze concrete topping; (b) later, clay blocks and slip tiles were used with
in situ concrete ribs and topping, usually with steel beams, often concrete-cased; Cax-
ton floor similar

(a) (b)

Figure III.36. Mushroom: originally developed by C.A.P. Turner, US pioneer of flat slab construction c.1906, notable both
for this and for four rather than two layers of slab steel: (a) plan with broken lines suggesting ‘mushroom heads’; (b)
perspective indicating draping of slab reinforcement

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:35:21
22_app_III.qxd 16/11/2001 16:42 Page 420

420 Historic concrete

Figure III.37. Siegwart: among the first hollow precast concrete units, originally reinforced
with corrugated sheeting, later with round bars

Figure III.38. Stolte floor: German system using reinforced hollow concrete blocks with
flat bar embedded reinforcement

Typical proprietary Various generic types of floors were in use during the middle part of the 20th cen-
tury. Typical floors are shown for each type, and makers are identified. Some of
concrete floors the types and their makers remain active today.
used in the Prestressing was often used in precast ribs, beams, and planks.
mid-20th
century
Figures III.39–III.48 are
reproduced from
Prefabricated Floors for Use in
Steel Framed Buildings (2nd
edn, 1965) by courtesy of the
publishers, the British
Constructional Steelwork
Association Limited, and
Figure III.49 is reproduced
from Mitchell’s Structure and
Fabric: Part 2 (3rd en, 1983)
by courtesy of Longman
Scientific & Technical.

Figure III.39. Precast ribs or joists with/without infill blocks (usually of fired clay or con-
crete) and/or composite topping: (i) Inverted T-sections – Armocrete shown; (ii) Joists
or X-joists – Lyncrete shown; (iii) X-joists with thin precast planks and composite in situ
topping – Pierhead shown. Makers: (i) Allisons, Anglian, Armocrete, Blockspan, Bradley,
Dabro, Flooring Contracts, Francis, Greenwood, HBS Self-centring, Invictus, Musker,
Quikspan, Rose Green, Tilecast, Trent. (ii) Fram, Lyncrete, Pierhead (later Booth), Rapid,
Trent. (iii) Pierhead.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:35:21
22_app_III.qxd 16/11/2001 16:42 Page 421

Appendix III: proprietary reinforcing and flooring systems 421

Figure III.40. Precast solid planks with/without composite topping – Bison (composite)
shown. Makers: Bison, Francis, GKN, Omnia, Parcrete, Quikspan, Stotam

Figure III.41. Precast slabs or ribs with lattice reinforcement and in situ concrete to form
composite structure, often with infill blocks between ribs – Filigree beam and block floor
shown. Makers: Atlas-Omnia (with trough formers), Combric, Filigree, GKN, Katzenberger,
Omnia, Sindall, Spancrete, Spanlite, Stahlton (prestressed clay block planks – see Chap-
ter 13), Triad. Variant with wide precast planks and lightweight void formers with in situ
topping: Combric Wideslab, Filigree Wideslab, Katzenberger Wideslab

Figure III.42. Precast inverted trough sections with/without structural topping – Siegwart
shown. Makers: Blatchford, Bradford, Ebor, Evercrete, Greenwood, Helicon, Invictus,
Matthews & Mumby, Milbank (see Chapter 13), Siegwart, Stuarts, Truscon (sometimes
used with timber bearers and floor boarding over concrete)

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:35:21
22_app_III.qxd 16/11/2001 16:42 Page 422

422 Historic concrete

Figure III.43. Precast single-void hollow cored sections (basically rectangular profile),
with/without structural topping – Dow-Mac shown. Makers: Arrow, Bison, Bradford,
Corite, Croft, Dow-Mac, Ebor, Evanstone, Framlite, Girling, Greenwood, Helicon, Invic-
tus, Matthews & Mumby, Raphcon, Siegwart, Stuarts, Viking

Figure III.44. Precast twin- or multi-voided hollow-cored sections (basically rectangular


profile), with/without structural topping – Truspan prestressed floor shown. Makers:
Bison, Dow-Mac, Evercrete, Francis, Marley, Shockcrete, Spiroll, Stotam Roth, Tembo,
Trent, Truspan, Viking

Figure III.45. Precast double tee units with structural topping – Blatchford shown. Mak-
ers: Blatchford, Dow-Mac

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:35:21
22_app_III.qxd 16/11/2001 16:42 Page 423

Appendix III: proprietary reinforcing and flooring systems 423

Figure III.46. In situ slabs composite with profiled metal decking permanent formwork – Holorib shown (note decking
profiles vary: typically either ‘dovetail’ re-entrant or trapezoidal). Makers: Expamet, Holorib (later Richard Lees), Robert-
son Thain (Q- and QC-Floor), etc. (many more profiles and makers from 1980s onwards)

Figure III.47. In situ slabs on permanent lightweight formers (e.g. woodwool) – Chipcrete shown. Makers: Chipcrete, Gyp-
klith, Marlith, Neolith, Spanform, Thermaflor, Wylam

Figure III.48. In situ ribbed slabs on metal formers (permanent or removable) – Kaiser shown. Makers: BRC, Kaiser. Vari-
ant with ribbed or coffered slabs on removable non-metallic formers: GKN, Mills

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:35:21
22_app_III.qxd 16/11/2001 16:42 Page 424

424 Historic concrete

Figure III.49. In situ ribbed slabs with permanent infill blocks, the ‘hollow pot’ floor with
hollow clay tiles, often used with thin ‘slip tiles’ to rib soffits to form a continuous plas-
ter key; generic form, but particularly associated with Phorpres (London Brick Co.) blocks.
Hollow clay tile makers: London Brick Co. Variant with hollow concrete blocks Smith’s
(with telescopic formwork for in situ ribs)

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:35:21
23_app_IV.qxd 16/11/2001 16:40 Page 425

Appendix IV: assessing ancient


mortars
There is often a need to determine the ‘mix proportions’ of old or even ancient
mortars, plasters, renders, and stucco – without resorting to the costly techniques
sometimes employed by archaeologists.1
Engineers need to know mortar ‘strengths’ when involved with structural alter-
ations to old buildings; as do conservators repairing or replacing eroded parts of
historic structures, and those researching the use of early materials, need to know
the essential makeup of mortars. Because of the wide variety of possible binders
and aggregates in these materials, together with problems relating to the limita-
tions of chemical analysis, their assessment requires careful thought. The chemist,
however, must be brought in early enough to glean valuable background infor-
mation in terms of sampling and seeing the material in its context.
For visual examination, discrete pieces of material must be supplied; powdered
material being virtually useless. Render materials, for example, need each layer
to be separately considered. External layers in particular, due to weathering, can
suffer loss of binder constituents by leaching and erosion. Colour photography is
useful as a visual record, although the colours may not always be reliable.
Where mortars have been used as a bedding material, the aggregate will usu-
ally have been chosen for practical reasons. However, for visual effect as in point-
ing or matching mortars to types of stone or brickwork, the builder may have gone
to some lengths to obtain a suitable aggregate and/or pigment.
Early colouring additives include crushed coal and ash, chalk, brick dust and
naturally coloured cements (such as the pinkish-brown ‘Roman’ cements from the
septaria of Harwich2). Information regarding type, shape and colour of the aggre-
gate particles can be gained from acid wash separation (for non-calcareous types),
sieve grading and visual examination with a hand lens or low-powered microscope.
Detailed investigation may be made by petrographic means after preparing a thin
section of the original material. The final colour of the mortar will have derived
from a combination of aggregate/additives and binder.
Today, synthetically prepared pigments are available in a wide range of colours
obtained from, for example, iron oxides, manganese and chromium oxides and
cobalt and ultramarine blues.
Excluding Portland cement, present from the mid-19th century onwards, the
most common binders2 used were limes of non-hydraulic, semi and eminently
hydraulic types. Gypsum was used mostly, but not exclusively, for indoor work.
Natural cements react essentially as eminently hydraulic limes but can contain mat-
ter which, dependent on source, give rise to natural pigmentation. Binder char-
acteristics are as follows:

• Portland cement: invented by Joseph Aspdin resembles Portland stone in appear-


ance and was patented in 1824, this hydraulic cement’s grey colour is roughly
proportional in intensity to its iron oxide content, helping with identification
in set materials.
• Non-hydraulic or high calcium limes: contain more impurities than modern build-
ing limes which are used mainly as workability aids, and had low proportions
of soluble silica and alumina, hardening purely by drying-out and normal atmos-
pheric carbonation.
• Hydraulic limes: these have chemical compositions not unlike Portland cements,
the components of silica and alumina imparting the ability to set under water.
Moderately hydraulic limes exhibit mid-range proportions of silica and alumina

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:33:48
23_app_IV.qxd 16/11/2001 16:40 Page 426

426 Historic concrete

which vary widely depending on the limestones used.


• Magnesian limes: essentially limes deriving from dolomitic limestones and with
a capability of hydraulic properties.
• Natural (Roman) cements: composition of these materials is a function of the clay
and mineral contents of the limestones used. Setting and hardening mechanisms
resemble hydraulic limes. Because of their colours, they were used extensively
for stucco and decorative features in neat form or in rich mixes with fine aggre-
gates.
• Gypsum: commonly used as Plaster of Paris, the partly dehydrated form of gyp-
sum employed as binders in mortars and plasters. Retarders were needed to
give flexibility in working practice (keratin from animal hooves and horns being
an early example); in early building work plaster frequently contained impuri-
ties and was used neat or with lime in the mix. Because plaster is sparingly sol-
uble in water is more often, but not exclusively, used for indoor work. An
alternative form is anhydrous gypsum (exemplified by Keene’s Cement) which
may need an accelerating agent.

The most common additives in non-hydraulic lime-bound materials are poz-


zolans, fine siliceous materials capable of reacting with lime in a mix to form cal-
cium silicate cementing compounds. Pozzolanic materials derive from compact
volcanic rocks, volcanic glasses and tuffs but where these were not available brick
dust was often used. Pulverized fuel ash (pfa) from power stations is the best known
modern form of pozzolan.
Where these materials are present in mortars they make identification of
hydraulic of non-hydraulic limes impossible by chemical analysis, as a non-
hydraulic lime/pozzolana mortar assumes a similar chemical composition, in sol-
uble component terms, as a hydraulic lime mortar.
In addition to these additives, natural organic substances were frequently added
to mortar to impart particular physical properties. These included animal fats, or
blood, egg white and beeswax, while hair and straw were commonly used as rein-
forcement.
Such substances would generally be present in significant amounts, enabling sim-
ple inexpensive qualitative laboratory tests to identify their presence. Quantities
can be determined using solvent extraction methods or comparative total organic
matter once the additive has been characterised. Sometimes infrared spectroscopy
is needed for positive identification. Visually, mortars do not always show the pres-
ence of additives.
Chemical analysis can be carried out using methodology based on that given in
BS 45513, considered to be equally applicable to old lime-based mortars as to mod-
ern Portland cement ones.
Interpreting such chemical analyses without visual examination would only in
a few cases lead to the correct mix being chosen.
Examples of different types of mortar analyses are as follows:

• External pointing mortar was creamy white with a discrete black surface coat-
ing. Analysis, by virtue of the high sulphate content, indicated a gypsum binder
with an excess of calcium assumed to be a lime addition rather than calcium
from the aggregate. The black coating consisted of a fine, essentially carbon black
material with a very high lead content, which gave the conclusion that the mor-
tar joints had been painted with black lead to protect the gypsum from weath-
ering.
• Taken from the same building, bedding mortar showed a very different analy-
sis with much lower content of sulphate in spite of its similar appearance. A low
soluble silica ruled out the presence of hydraulic lime. Mix proportions were
reported as non-hydraulic lime and sand, which assumed that the sulphate was

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:33:48
23_app_IV.qxd 16/11/2001 16:40 Page 427

Appendix IV: assessing ancient mortars 427

present as a natural impurity in the lime of aggregates rather than as a delib-


erate addition.
• Creamy white in appearance, old bedding mortar from a canal tunnel contained
sizeable particles of unmixed lime, almost white. Aggregate contained a pro-
portion of crushed brick debris. The high magnesium content indicated the use
of a magnesium lime. Qualitative analysis of separated lumps of lime confirmed
the high magnesium content but low silica. The mortar was judged to have been
made with a non-hydraulic magnesian lime, a high soluble silica content being
attributed to clay brick particles.
• A stucco sample was notable for its very hard dense nature and for the absence
of visible aggregate particles. Analysis suggested hydraulic lime, the silica and
alumina content being high. Additionally, a high soluble iron content indicated
a pigment or iron-rich substance, almost certainly related to the sample’s brown
coloration. Composition was assumed to be of a neat natural cement with an
inherent brown mineral coloration.
• Analysis of a quite firm, greenish-grey mortar gave no reaction with acids and
alkalis. On heating, all of the mortar burned away. More detailed organic analy-
sis revealed the presence of denatured protein suggesting eggs. In such an
instance, bearing in mind the age of the mortar (400 years) chemical analysis
could not give the exact original composition. Perusal of the parish records from
June 1594 eventually solved the mystery: ‘Mortar to be made up of 40 gallons
of worte and 3s of eggs’ (worte is unfermented beer).

Mix proportions are traditionally expressed on the basis of volume ratios, to cal-
culate which of the bulk densities of material must be known. These values are
known for modern materials, but not for the variable aggregates and aggregate
mixtures used in old mortars and plasters. Thus calculations of volume propor-
tions have to be based largely on guesswork and it is desirable to express mix pro-
portions both by weight and volume.
Where no background information is forthcoming and several mixes are pos-
sible, alternative mixes are presented together with a ‘balance of probabilities assess-
ment’ as to the most likely mix.
I have tried to demonstrate that for a relatively modest amount of work and
cost, significant information can be gained about the makeup of old mortars and
plasters.
As a cautionary note, it should be stressed that possession of this information is
only a small step in terms of carrying remedial works such as repointing and re-
rendering on old structures. The preparation of effective replacement mortars it
is said, is 20% knowledge of the original mix and 80% the technique in deciding
on effectively preparing the material. Deciding whether a more modern material
might be the best option demands great skill and should be entrusted to those
qualified to make such decisions.

References 1. An investigation of the chemical constituents of some Renaissance plasters, Newton, R.G.
and Sharp, J.H, Studies in Conservation, 32 (1987), pp. 163–175.
2. Ashurst, J. and N. Practical Building Conservation.
3. BS 4551: Part 2: 1998 Analysis of mortars, screeds and plasters.

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:33:48
24_index.qxd 16/11/2001 16:22 Page 429

Index

Headings in italics refer to publications. Page numbers in italics refer to illustrations, a ‘t’ suffix indicates a table
of information.

aggregates bearing capacities, foundations, Dover viaduct, 216, 218


alkali reactions, 109, 241, 345, 367 152–153, 152t, 153t, 154t expansion joints, 237, 237, 268
breeze, 47 bentonite, diaphragm walls, 146 epoxy resin, 264
clinker, 78, 79, 83–84 boats, ferro-cement, Nervi, 31 Farnworth 217
combustability, 47, 78, 79, 83–84 Bonna system, sewer pipes, 324–325, Findhorn 218
lime concrete, 47 325, 326 Floriston 217
‘no fines’, 97–98 Bournemouth foot-, 228–229, 230, 231, 232
precast concrete, 234 Alum Chine 229, 230 Ganter 28, 29
waterproofing, 324 Bus Garage 173, 174 Hartlake Bridge 218
air raid shelters, 92, 94, 377 breeze, combustability, 47 hinges
aircraft hangers bridges concrete, 267, 267
cylindrical shell, 167, 173, 174 A9: 29 Considère, 235
precast beams, 34, 35 Aberuthven 217 lead strip, 234
ribbed shell, 31, 32 aesthetics, 234, 241–242 Mesnager, 267, 268
airship hangers, corrugated vault, 21, Alum Chine 230 uneven loadings, 234–235
22 Bargoed 220 Holmersfield, 212, 212
alkali aggregate reaction [AAR], 345, beam and slab, 213t, 218, 258, 259 incremental launching, 261, 265
367 bearings Kahn system, 216, 218, 219
alkali-silica reactions [ASR], 109, 241, rocker, 235, 236 Kew Gardens 230
367 roller, 235, 236 Kincardine-on-Forth 237
aqueducts, prestressed concrete, 30, rubber, 267 King’s, Belfast 219
30 sliding, 235, 267 King George V 224
arch, spandrel Berw, 216 Lake Maracaibo 39
open, 214t, 215–216, 216, 221–222, Borrodale 15 listed buildings, 241, 270
222, 264 bowstring, 213t, 215, 225 loading
solid, 214t, 216, 217, 222, 223, 227 box girder, 227–228, 229, 263–264, inter-war criteria, 238–239, 238,
architects 268–269 239
and engineer-designers, 43, 70–71, cable-stayed, Ganter, 39, 40 inter-war tests, 239–240, 240
94 cantilever London 152
engineers as, 45 balanced, 214t, 224–225, 225 local authority designed, 219–220,
architecture prestressed, 256, 258 221
and engineers, 43, 45, 70–72 reinforced, 52, 53 Maracaibo, L, 39, 39
and structure, 28, 34 suspended span, 261, 263, 263 mass concrete, 15, 15, 211, 212
Arup, Ove, 28, 29, 90, 90, 91 Charing Cross 152 Mizen Head, 232, 233, 240
see also Ove Arup & Partners Cherwell 257 Monier 16
asbestos-cement sheets, 79 Clifton Bridge 263 Montrose 225
Aspdin, Joseph, 13, 46 Chiswick, 222, 223 motorways, 263–264, 265–266, 265,
Association of Portland Cement Chiswick flyover, 264–265 266
Manufacturers, 105 Clifton, Nottingham, 261, 263, 263 Mouchel-Hennebique system,
Coignet, 212, 218 69–70, 215–216, 215, 216, 275
barrel vaulted roofs, 178, 179, conservation, 240–241, 270 Northam, 261, 262
180–181, 180, 187, 187 Considère system, 216, 217 Nunn’s Bridge 253
design parameters, 178, 179, 179t, Crewe Park, 216, 217 Ogmore 217
180 decking, beamless, 19 partial prestressed, 259, 260
roof lights, 181 design Plougastel 22, 23
stiffening beams, 180, 180 analysis, 268–269 portal frame, 214t, 220, 221, 225,
valleys, 180 tools, 268 226, 256, 257
Bauhaus movement, 27 trends, 221–222, 269 propped, 266

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:34:05
24_index.qxd 16/11/2001 16:22 Page 430

430 Historic concrete

bridges (cont’d) York Racecourse, 218, 219 churches


post-tensioned, 39 Ystradgynlais 226 barrel vaulted, La Raincy, 23, 25
precast British Concrete Steel Co, indented hypar form, Mexico, 38, 38
decoration, 234 bar use, 73 in situ walls, Anerley New, 59
foot-, 230, 231, 232 British Fire Prevention Committee, mass concrete, Le Vesinet, 14, 14
in situ cast, 255, 256 58–59, 301 Unity Temple, 26, 26
non-system, 233 Concrete Aggregates Committee, CLASP [Consortium of Local
standard designs, 267 83–84, 302t Authority Schools Projects]
prestressed, 39, 39 British Precast Concrete Federation, 77 system, 95, 97
beam and slab, 258, 259 British Reinforced Concrete clinker concrete
cantilever/suspended span, 261, Engineering [BRC] combustability, 79, 83–84
263, 263 foundation, 72 reinforcement, corrosion, 79
decks, 252, 252 New Royal Horticultural Hall, 85, Coignet, Edmond, 68
early UK, 252, 253, 254, 254 85 see also Edmund Coignet & Co
in situ cast, 255–256, 256 piles, 123, 126, 127 piles, 126, 127, 130, 133–134, 134
portal frame, 256, 257 shell roofs, 35 reinforcement, 69t
post-war development, 251 water towers, 331, 333 Coignet, François
problems, 269 British Standards bridges, concrete arched, 212
war-time stockpile, 252 BS 12:1904–91 Portland cement, 83, patents, surface finishing, 13–14
railway, 229, 231, 231 105, 106 walls, exposed, 13–14
foot, 229, 230, 231, 232 BS 785:1938 Reinforcement, 94 Collins, A.E., 305
over-, 258–260, 259 BS 8110: 1985 Structural Use of columns
under-, 258, 259 Concrete, 99 compressivity, 84
West Highland line, 15, 15, CP 110, Unified Code 1972, 99 coverage, 85
212 CP 114 Structural use of Concrete foundations, 122, 122
reinforced concrete, 18–19, 221 1948, 95 heads, flat slabs, 18, 86, 87, 88
Continent, 18–19, 19, 20, 21 1957, 98 modular ratio design, 88–89
earliest UK, 212, 212, 215 1965, 98 composite construction, 34, 98
reinforcement, expanded metal, 1969 Metric, 98 compressive strengths, concrete,
211 reinforcements 1933–85, 93–94t 1915–85, 100t
Rhinefield 254 structural concrete 1907–85, 111t concrete
Royal Tweed, 221–222, 222, 241 British Steel Piling, 142, 143 see also aggregates; cements;
rubble concrete, 211–212 Brunel, Marc Isambard, 46, 52, 53 concrete mixes; lime concrete;
Schwandbach 21 building systems, precast panel, 7 mass concrete; prestressed
service life, inter-war built, 240–241, concrete; reinforced concrete;
241t calcium chloride, concrete corrosion, shotcrete
Souillac 211 113 admixtures, 113
slab vault, 213t, 220, 240 Candela, Felix, 37, 173 air raid shelters, 92, 94, 377
spine box, edge cantilevers, 264 CCL prestressing system, 199, 203t blocks, 79
St James’s Park 258 Cementation piling systems, 142, 144 compacting, 77
St John’s Hill 233 cements compressive strengths, 1915–85,
Stakeford 219 see also concrete; hydraulic cements; 100t
Stow 227 Portland cement; Roman cement creep, 192
Sutton Drain 215 alkali-silica reactions, 109, 241, 367 cyclopean, 346
Tavanasa 19, 42 asbestos, 79 definitions, 11, 46, 47
Thorverton 220 Aspdin 13, 46 degradation, 113–114, 114t
Tower 152 blast furnace, 106t calcium chloride, 113–114
Twickenham 4, 223 composite, 106t, 110 sea-water, 275–276, 278, 297
unequal loadings, stiffness, 19 European standards, 106–107t sewers, 324
Valtschielbach 19, 20 glass reinforced, 79 durability, 113–114, 293–294, 372
variable depth trusses, 213t high alumina, 92, 98, 114–115, 206 expansion, long-term, 110
Verdre 20 low heat, 345 ferro-, 31, 275, 335
Vierendeel girder, 214t, 223, 227 pozzolanic, 46, 106t, 118, 345 gypsum binder, 47
Water Eaton 226 pulverised fly ash content, 110 heat of hydration, 110
Waterloo, 227–228, 229 selenitic, 372, 373 lattices, tall buildings, 40
Westminster 152 slag content, 110 lime ash, 49
White’s 222 water ratio, 77, 109, 348 minimum standards, 99
Williams designed, 27, 27 marine environments, 278 Mouchel specification, 76
Wisbech 226 vibrated concrete, 95 placement rates, 348
Withycombe Road 221 woodwool, 79 procurement procedures, 112

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:34:05
24_index.qxd 16/11/2001 16:22 Page 431

Index 431

reactivity, 110 Considère and Partners, bridges, 218, delayed ettringite formation [DEF],
ready-mix, 112 222–223, 223 110
roller compacted [RCC], 349, Considère system design
355–356, 355t, 356t bridge hinges, 235 aesthetics, 4, 234, 241–242
setting bridges, 216, 217 bridges
temperature control, 348–349 piles, 125, 127 post-war, 251
times, 109 reinforcement, 69t stress analysis, 268–269
sewers, 324–325 foundations, 136, 138 testing, 269
shrinkage stresses, 191–192 construction and build services, 70–72
strength bridges, incremental launching, code adoption, 99
changes over time, 107–108, 107t, 261, 265 computerization, 99–100, 251, 268
111–112 concrete concrete specialists, 72, 75–76
Pasley tests, 47 air expulsion, 77 and construction, 72
time-dependent effects, 112 customer demands, 113 dams
Concrete and Constructional Engineering, mechanical vibration, 77, 89, 94, finite element, 352
80, 100, 301 95 trial load, 352
concrete frames punning, 77, 77 disguising concrete, 75
butt-connections, 73, 74, 75 ready-mix, 112 exposed aggregate surfaces, 5
flats, Paris, 23, 24 site-batching, 76 legislation, 84
Newcastle NER depot, 132, 132 maintenance clauses, 95 limit state, 99
prestressed, 208 specialization, 111 process, 72
retaining wall integration, 141 worker training, 94 shells
Concrete Institute, The [after 1923 see contractors, 71–72, 178 barrel vaulted, 178, 179, 179t, 180
Institution of Structural creep, 20, 22–23 detailing, 184–185
Engineers] crib walling, 151 specialization, 111
building regulations, contributions culverts, 325–326, 327 surface textures, 4–5
to, 307, 309 cyclopean concrete, 346 designers, engineers as, 46
Council, 1908, 303t diaphragm walls, 147–148
cross-committee representation, Dagenham, May and Baker Canteen dams, 146, 148
310–311 166 slurry trench, 146
entrance examinations, 306 dams domes
foundation, 84, 301, 302t arch see also shells
Honorary Members, 306t double curvature, 352, 353, 356, arch supported, 26, 26
membership, 304, 304, 307, 308–9 356t column supported, 29, 30
notation, standardization, 309–310 multiple, 350–351, 350, 352 Roman, 12–13
objectives, 302, 304, 305 thin, 352, 353 spherical, bending effects, 166
papers, 306, 306t, 309t variable radius, 352 dry docks, 288, 289, 293–294
publications, 311, 311t buttress, 349–350
title change, 307 cements, early, 343–345 Edmund Coignet & Co, bridges, 212,
Transactions, 80, 310 concrete 218
concrete mixes facing, 354–355, 355t piles, 126, 127, 130, 133–134,
see also aggregates hearting, 354–355, 355t 134
dams, 353–56, 355t, 356t mixes, 353–356, 355t, 356t elastic modular ratio theory, 83
grades [1933], 89, 100t placement rates, 348 engineer-designers
procurement, 112 roller compacted [RCC], 349, 355 as architects, 45
roads, 361 contiguous, 351, 351 and architects, 43, 70–71
water-cement ratio design project involvement, 94
durability, 109 finite element, 352
strength, 77, 109, 348 trial load, 352 Faber, Oscar, modular ratio design,
vibrated, 95 diaphragm walls, 146 88–9
cone anchorages, 39 flat slab, 351 factories
conservation deterioration, 351 Boots ‘wets’, 86, 87, 88
bridges, listing, 241, 270 foundations, 120 Colodense foundations 95, 97
deicing penetration, 240, 241 gravity, 349 iron framed, 16, 16
Gunite repairs, 240, 278 Hoover, 348–349, 348, 349 northlight roofs, Cadbury Bros,
re-alkalization, Roman concrete, masonry, 346–347, 346, 347 186, 186, 187
62 prestressed, 352–353, 354 Perret designed, 23
spalling, 6 deicing salts planform dome, Brynmawr, 173,
Considère, Armand, column binding, concrete bridges, 240, 241 173
72–73 concrete roads, 114, 367 prestressed arch frame, 95, 97

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:34:05
24_index.qxd 16/11/2001 16:22 Page 432

432 Historic concrete

factories (cont’d) tile creased, 54–55, 55, 56 variable overburden, 131, 131
raft foundations, Rowntrees, unreinforced, 53–54, 78–79 viaducts, railway, 119
132–133, 133 isostatic ribbed, 32–33, 33 Fowler, Charles, 54–55, 55, 56
reinforced concrete, Weaver’s Mill, plaster, 49, 49 Fox & Barrett flooring, 49–50, 50, 51
68, 70 precast, early C20, 78, 78 Fox, Henry Hewes, 49–50, 50, 51
reinforced concrete foundation, prestressed, 195, 203 Franki piles, 127–129, 130
J.C.&J.Field, 133–134, 134 Stahlton, 206–207, 207 Freyssinet, Eugène
Simplex piled, Tranmere Bay, Ransome rib, 17, 17 bridges, 20–21, 22, 23
134–135, 136 floor systems concrete, vibrating, 77
fat lime, 343t, 344 Bunnett 404 creep, overcoming, 22–23
Fawcett, fireproof floors, 57, 58 Dawnay 402 piles, prestressed concrete, 145, 145
Ferro-Concrete, 80 Dennett & Ingle 403 prestressing system, 38–39, 191,
ferrocement, 31, 32 Fairbairn, Sir W 401 197, 197, 198, 199, 200t, 252,
see also reinforced concrete Fox & Barrett 401, 402 254
Festival of Britain, prestressed Hornblower 404 vaults, 21–22, 22
concrete, 34 Hyatt 404 conoid, 22, 22
fire resistance, 50–51, 84 Lindsay, W H 403 Frost’s cement see Roman cement
fireproofing Liverpool "Flags" 402
breeze aggregate, 47 Maurer 405 gasholder tanks
embedded iron joists, 14–15 Moreland 401, 403 clay bases, 336
and fire resistance, 50–51 Pioneer 405 early, 335
floors see floors, fireproof Raritan 405 Portland concrete
public concern, 50, 56–57 Scott 401 footings, 335
flat slabs Vaux & Thuasné 403 tanks, 336
building structures, 86 Whichcord 403 reinforcement, 323, 337
columns Woodhouse & Potts 401 Gifford-Udall/Burrow prestressing
heads, 86 foundations systems, 199, 205t
mushroom heads, 18, 87, 88 see also piled retaining walls; piles GKN Reinforcements [GKNR],
compression, behaviour, 70 bearing capacities, 152–153, 152t, 177–189
dams, 351 153t glass-reinforced cement [grc], 79
floors, 1, 3, 70 dams, 120 Glenfinnan Viaduct 15
Stahlton, 206–207, 207 design, 152–153 Goldsmith, Myron, 39–40, 40, 41
tile creased, 54–55, 55, 56 grillage over concrete, 131, 131 Gropius, Walter, 27
unreinforced, 53–54, 78–79 legislation, 120 gypsum, concrete binder, 47
‘lift’ construction system, 209 raft
Maillart, 20, 42 nineteenth century specifications, halls
reinforcement layers, 86–87 120–122 arched, New Royal Horticultural,
flats first, 117 85, 85
box-frame ICE building, 132 domes, arch-supported, 26, 26
Highpoint, 28, 90, 91, 92 Rowntrees Works, 132–133, 133 Harris, A.J., 34, 35
Rosebury Avenue, 95, 96 War Office, 131 Hennebique, François
concrete framed, Paris, 23, 24 reinforced concrete, J.C.&J.Field’s bridges, 69–70, 215–216, 215, 216,
Eldon Street 1, 2 factory, 133–134, 134 275
precast panel, 1, 2, 3, 7 reinforced footings, NER depot factories, Weaver’s Mill, 68, 70
floors Newcastle, 132, 132 Paris house, 17
aggregate combustablity, 78, 79 reinforcement piles, reinforced concrete, 123, 124,
composite construction, 98 BRC systems, 123 125, 125, 130, 134, 280
filler joist, 78–79, 78 nineteenth century, 120, 121 port works, 275
fireproof Considère system, 69t, 136, 138 reinforcement system, 15–16, 16,
Dennett’s arch, 50, 52 construction, 122, 122 67, 68, 69t
Fawcett, 57, 58 CWS Newcastle, 130 retaining walls, 136, 136
Fox’s patent, 49–50, 50, 51 settlement, differential, 120 specifications, 68–69, 69t
Homan & Rogers, 57, 58 soil pressures, allowable, 152, 154t warehousing, CWS Newcastle, 70,
Hyatt’s, 57, 59, 60 specifications, nineteenth century, 71
incidence, 59, 62 119, 120 high alumina cements [HAC], 92, 98,
King’s fireclay tube, 57, 57 tall buildings 144–145, 150 114–115, 206
reinforcing cover, 50, 57–58 timber/fascine, 117 hogging moments, reinforcing bars,
theatres, 14 underpinning 67, 89–90
flat slab, 1, 3 Chatham Dockyard, 119–120, 119 Homan & Rogers fireproof floors, 57,
load behaviour, 70 Custom House, 46, 48, 48, 118 58

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:34:05
24_index.qxd 16/11/2001 16:22 Page 433

Index 433

hoop iron early use, 13, 371–372 loadings, 285


brickwork bonds, 52, 53, 54 foundations, 13, 48, 48, 118 slab and beam, 276–278
reinforcing, 52, 62, 337 manufacture, 47 specifications, 279t, 280
hotels, foundations, Ritz, 131, 131 maritime structures, 372 lighthouses
Hoyes prestressing system, 252 listed buildings, bridges, 241, 270 Eddystone, 13, 46
Hyatt, Thaddeus Liverpool Kish Bank, 293, 294
floors, fireproofing, 57, 59, 60 gas works 326 Middle East, 296–297
reinforced concrete, heat expansion, Royal Liver Building 1, 74, 76 Parkeston quay, 281–282, 281
83 Eldon Street flats 1, 2 piles
hydraulic cements Seaforth Docks 147 braces, 282–283, 283
effect of lime, 46, 371 shell roof testing 183 precast, 278
placing machine, 118 Local Government Board reinforcement, 280–281
Portland, 120, 273 funding, 221, 332–333 Screwcrete, 286–287
pozzolans, 13, 46, 118, 345 reinforced concrete, attitude, 324 steel, 292
Roman, 13, 117–118, 120, 344 swimming baths, 332–333 Purfleet pier 215, 216
Smeaton’s experiments, 46 London Clay, piles, 131, 131, 144, precast, prestressed, 288–290, 290,
strength, 345 145 291
Telford’s experiments, 117–118 London County Council reinforcement, corrosion, 273–274,
terras, 46, 118 Building Construction Guide 275–276, 291–292
[1938], 91 sea forts, WWII, 377–378, 377, 378
Indented Bar Co, 218–219, 331 Reinforced Concrete Regulations Southend Pier, 283, 284, 285
Institution of Structural Engineers [1915], 81, 84–85 Tema type walls, 274, 274
[before 1923 see Concrete London County Council [General U-boat pens, 378–379
Institute], 94, 307 Powers] Act [1909], 46, 84, 153, MARS Group, 28–29
Institution of Civil Engineers [ICE] 307 Marsh, C.F., Reinforced Concrete, 80
Concrete Institute, 301 Lubetkin, Berthold, 28 mass concrete
Piling Committee, 145 bridges, 15, 15
Reinforced Concrete Committee, 84 Macalloy prestressed concrete piles, definition, 273
Sea Action Committee, 276 146, 147, 148, 149, 291 dock walls, London, 120, 121
Intergrid construction system, 208, Magnel-Blaton prestressing system, flooring, 14
209 197, 201, 202–203t, 260 fortifications, 373–374, 373, 374
International Congress of Modern Maillart, Robert, bridges, 18–19, 19, in situ, early, 59, 60
Architecture [CIAM], 28 20, 42 retaining walls, 135–137, 136, 137,
maintenance, concrete structures, 6, 141
Johnson, Thomas Marr, 211 95 swimming baths, 333–334
Johnson’s bar system, 69t, 73 maritime structures May and Baker Canteen 166
Joint Committee on Reinforced berthing forces, 282–283, 283 Metropolis Management and
Concrete, 84, 304t, 307 caissons Buildings Act [1878], 120
corrosion, 291–292 military structures
Kahn bar system Phoenix, 287–288, 287, 378, 379 air raid shelters, 92, 94, 377
bridges, 216, 218, 219 concrete blast resistance, air-space theory,
patent, 72 degradation, 275–276, 278, 297 374, 375
specification, 68, 69t durability, 293–294, 372 fortifications, 373–374, 373, 374
water towers, 331, 332, 334 mix specifications, 278, 285–286, Maginot Line, 375
Kahn, Fazlur, 40, 41, 43 289t, 290t, 293–294 Moir pill boxes, 374, 375
Kew Garden footbridge 229, 230 Condeep platforms, 294–295, 295 Mulberry Harbour, 287–288, 287,
King, J.A., fireclay tube floor, 57, 57 dry docks, 288, 289, 292–293 378, 379
Ekofisk artificial island, 294, 295 nuclear blast, 379
Laingspan construction system, 208, Erith jetty, 288–290, 290, 291 off-shore fortifications, 377–378,
208 foundations 377, 378
Lascelles, William, precast Roman Chatham Dockyard, 119–120, Portland concrete blocks, 373–374
concrete, 61–62, 61 119, 373–374 precast huts, 375, 376, 377
Lee-McCall prestressing system, 196, London docks, 120, 121 Ranger’s concrete, tests, 371–372,
199, 203t, 252 gravity walls, 274 372
lighthouses Grimsby Fish Dock, 283, 283, reinforced concrete
Eddystone, 13, 46 284t WWI, 374
Kish Bank, 293, 294 Hennebique, 275 WWII, 375, 376, 377–379
lime ash concrete, 49 jetties Siegfried Line, 375
lime concrete decks, 276–278, 277 modular ratio method, concrete
aggregates, 47 deep water, 288 strength, 68, 88–89

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:34:05
24_index.qxd 16/11/2001 16:22 Page 434

434 Historic concrete

Monier, Joseph Ove Arup & Partners Prestcore 142, 143


bridges, 212 factories, Brynmawr, 173, 173 Raymond, 127, 128
reinforced concrete, patent, 15, 68, flats Rotinoff system, 144
83 Highpoint, 28, 90, 91, 92 Screwcrete system, 144, 286–287
sewer pipes, 324 Rosebury Avenue, 95, 96 Simplex, 127, 128, 135, 135
mortars foundation, 34, 95 Vibro, 142
hydraulic, 46, 344 Penguin Pool 28, 90, 91 jetties, 276–278
Roman cement, 46, 117–118 London clays, 131, 131, 144, 145
L.G.Mouchel & Partners Pantheon, The, 12–13, 12 marine environments, 280–281,
bridges Paris 280, 281
foot-, 228, 230 Exhibition [1900], 16–17 precast
Hennebique system, 215–216, Rue Franklin 23, 24 advantages, 278
215, 216, 221–222, 222 Parker, James, 13, 46 BRC, 123, 126, 127
portal frame, 225, 226 Pasley, Charles William Coignet, 126, 127, 130, 133–134,
railway, 229 brick beam testing, 52, 53 134
concrete, specification, 76 cement testing, 47 Considère, 125, 127
Hennebique system, 69–70, 275 military testing, 371–372 Hennebique, 123, 124, 125, 125,
piles, hollow, 123, 125, 125, 275, patents 130, 134, 280
275, 280 see also piles Mouchel, 123, 125, 125, 275, 275,
reservoirs, 338 beamless decking, Maillart, 19 280
Royal Liver building, 1, 1, 24, facings, Coignet, 13–14 prestressed concrete, 145–146, 145,
74–75, 76 fireproof flooring, Fox, 49, 50 209
sewage tanks, 338 flat slab mushroom heads, Turner, Macalloy, 146, 147, 148, 149, 291
swimming baths, 335, 338 18 plaster floors, 49, 49
water towers, 330, 330, 331, 331, helical binding columns, Considère, Portland cement
338 72–73 alkali content, 109, 110t
Mouchel, L.G., 68 Kahn bar, Trussed Concrete, 72 alternative forms, 106t
precast concrete, Lascelles, 61–62, 61 British Standards, 83, 105, 106,
Nervi, Pier Luigi reinforced concrete, 59 106t
hangers, ribbed shell, 31, 32 Hennbique, 67 C3S/C2S content, 108, 108t
roofs, ferro-cement, 31–32, 32, 42 Monier, France, 15, 68 changes
staircases, reverse spiral beam, 31, Ransome, USA, 68 causes, 108
31 Wilkinson, 15, 55, 67 effects, 108–109
‘no fines’ construction, 97–8 Roman cement, Parker, 13, 46 strength, 107–108, 107t
northlight roofs, 168, 170, 181–183, shuttering, Tall, 14, 59 contents, 106
182, 183t, 186, 186, 187 Perret, Auguste, 23, 24 corrosion resistance, 109
notation, standardization, 84, 309–310 pile driving, 129, 135 discovery, 13, 46
nuclear power industry, 209 compression stresses, 141–142 durability, 109
dynamic formulae, 153 foundations, nineteenth century,
off-shore structures hard strata, 141 120
Brent Field, 42, 42 stress waves, 142 gasholder tanks, 336–337
Condeep, 294–295, 295 piled retaining walls manufacture, 344–345
Ekofisk, 294, 295 bored, 150 maritime use, 120, 273
WWII sea forts, 377–378, 377, 378 contiguous, 150 temperature shocks, 337
offices secant, 150–151 post-tensioning, 39, 203
concrete framed, Royal Liver Peine sheet, 292–293 pozzolans
building, 1, 1, 24, 74–75, 76 precast sheet, 148–150 fly ash, 345
in situ, early, 59, 60 piles Portland cement, 106t
raft foundation see also foundations; pile driving; Roman cement, 118
ICE building, 132 piled walls Roman usage, 12–13
War Office, 131 bored, large diameter, 144–145 Smeaton’s experiments, 46
Ransome system, Ingalls building, braced, 282–283, 283 uses, 345
17, 18, 75 foundation caps, 122, 122, 143 precast concrete
retaining walls, RAC London, 140, impact strength aggregates, 234
141 reinforcement, 141–142 beams, aircraft hangers, 34, 35
Roman concrete, Royal Albert Dock, wet curing, 142 bridges see bridges, precast
61–62, 61 in situ early use, 77–78
shell and core construction, Cementation systems, 142, finishes, 234
Brunswick building, 40, 41 144 floors, early C20, 78, 78
tall, 39, 40 Franki, 127–129, 130 piles see piles, precast

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:34:05
24_index.qxd 16/11/2001 16:22 Page 435

Index 435

postwar growth, 33, 193 relaxation stresses, 196 reinforcement


Roman, 60–62, 61 settling foundations, 22–23, 193 see also prestressing; reinforced
walls, flats, 1, 2, 3 project costings, whole-life, 114 concrete
Prestcore piling system, 142, 143 pulverised fly ash [PFA] content, anchorage, 73, 84
prestressed concrete cements, 110 bar design, 73
see also prestressing puzzolana see pozzolana bending stresses, permissible, 89
acceptance of, 23 bonding, 73
aqueducts, Alloz, 30, 30 rafts see foundations, rafts carbonation, 7
bridges see bridges, prestressed railway bridges Coignet system, 69t
CCC prestressing system 199, and electrification, 259, 260 cold-worked, 99
203t erection, service disruption, 261 compression stresses, permissible,
creep, 192 prestressed, 258–260, 259, 260 89
dams, 352–353, 354 partial, 259–260, 260 corrosion
early experiments, 191–192 railway stations carbonation, 7
Festival of Britain, 34, 256 Newcastle NER Goods, 132, 132 clinker concrete, 79
floors, 195, 203 Victoria extension, 130–131, 131 iron, 47
Stahlton, 206–207, 207 Ranger, William, 371–372 seawater, 273–274, 275–276
frames, 208 Ransome, Ernest, 17, 17, 18, 68, 75 slag content, 292
Freyssinet 38–39, 191, 197, 197, Raymond piles, 127, 128 cover
198, 199, 200t, 252, 254 reinforced concrete fire resistance, 84
Gifford Udall 199, 205t see also reinforcement importance of, 50
GKNR 177–189 adoption, 68, 73 lack of, 6–7, 6
ground anchors, 209 Bonna system 324–325, 325, 326 vulnerability, 67
Lee Maccall 196, 199, 203t, 252 BRC 123 drawings, 8, 9
‘lift’ slabs, 209 building regulations, 73 factors of safety, 89, 98
Magnel-Blaton 197, 201, 202–203, coefficients of thermal expansion, foundations
260 83 BRC systems, 123
nuclear power industry, 209 Coignet 69t, 126, 127, 130, nineteenth century, 120, 121
piles, 145–146, 145, 209 133–134, 134 Considère system, 69t, 136,
Macalloy, 146, 147, 148, 149, 291 Considère system 69t, 125, 127, 138
precast units, 203 136, 138, 216, 217, 235 construction, 122, 122
railway sleepers, 206 construction, as specialist activity, differential loads, 133–134,
roofs, inverted trough, 206, 206 72 134
shrinkage, 192 cottage, Wilkinson, 55, 56 grouting failures, 291
trusses, 208, 208, 209 elastic modular ratio theory, 83 hogging moments, resistance, 67,
UK adoption of, 194–195 flexural cracking, 191 89–90
X-joists, 206, 206 Government buildings, 73 hoop iron, 52, 62, 337
prestressing in situ, early, 59 brickwork, 52, 53, 54
application, 192 Indented Bar 73, 218–219, 331 hot-rolled, 99
CCL system, 199, 203t international conference on [1930], indented bar, 136, 137
circular tanks, 203 27 iron, 15, 47
concrete strength, 196 Kahn system 72, 216, 218, 219 operation, 67
corrosion, 269 Local Government Board, attitude, shear stresses, 67, 89
definition, 11 324 shells, 177–178, 184–185, 184t
ducts, grouting, 203 Mouchel Hennebique 67, 69–70, specifications 1915–85, 93–94t
Freyssinet system, 38–39, 191, 197, 215–216, 215, 216, 271 tensile stress, 67, 83
197, 199, 200t, 252, 254 oil rigs, Brent Field, 42, 42 anchorage, 84
Gifford-Udall/Burrow systems, 199, patents, 15, 55, 59, 67, 68 torsional stresses, 90
205t Penguin Pool, London Zoo, 28 unnecessary use, 6–7, 6
grouting failure, 269 Ridley Cammell 695 yield stresses, 94
high-tensile wire, 195 reservoirs, 328 repairs see conservation
Hoyer system, 252 sinple 134–135, 136 reservoirs
invention, 191 specifications, 69–70, 69t covered
Lee-McCall system, 196, 199, 203t, strength, modular ratio method, 68, barrel vaulted, 326, 328
252 70 jack arched, 328
Magnel-Bleton system, 197, 201, swimming baths, 334–335, 335 hybrid construction, 328
202–203t, 260 testing, deflection, 70 reinforced concrete use, 328
partial, 259, 260 Reinforced Concrete, C.F.Marsh, 80 sloping sided, 326, 329
post-tensioning, 39, 203 Reinforced Concrete Regulations, LCC uncovered, 326
pre-tensioning methods, 196 [1915], 81 vertical sided, 328

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:34:05
24_index.qxd 16/11/2001 16:22 Page 436

436 Historic concrete

retaining walls mortars, 46 hypar see hyperbolic paraboloidal


bored piled, 148–150, 150 precast hyperbolic paraboloidal, 38, 38,
contiguous, 150 blocks, 60–62 188–189, 188
secant, 150–151 building decoration, 60–62, 61 loadings
cantilever, 136, 140, 141 colouring, 61–62 charactersitics, 164
concrete framing, 141 roofs deflections, 185
counterfort, 136 see also domes; shells design calculations, 37
crib walling, 151 counterbalanced, 35, 35 northlight construction, 168, 170,
gravity, 274 ferro-cement, 31–32, 32 181–183, 182, 183t, 186, 186, 187
reinforced earth, 151 folded plate, 36–37, 36 reinforcement
reinforcement inverted trough, 206, 206 specifications, 184–185, 184t
Considère system, 136, 137 isostatic, 37, 37 suppliers, 177–178
indented bar, 136, 137, 141 Rotinoff piling system, 144 stress systems, 164, 170
Ridley-Cammell system, specification, Royal Institute of British Architects complex curves, 168, 168, 169,
69t [RIBA], 84, 304t, 307 170
road bridges rubble concrete, bridges, 211–212 design analysis, 167–168, 183
loading, inter-war criteria, 238–239, edge bending effects, 166–167,
238, 239 Sachs, Edwin O., Concrete Institute, 166, 167
motorways, 263–266, 265, 266 foundation, 301, 301 reinforced concrete, 166
standard designs, 266 safety factors support, early concepts, 166
roads load-factor approach, 98 thickness, 183–184
Alconbury Hill, 364, 365, 366, 367 reinforcement, 89, 98 valleys, 180, 184
capping layers, 359, 359 F.J. Samuely & Partners, 37, 37, 95, ship test tanks, 337–338, 337
composite concrete/bitumen, 368, 97 shotcrete [sprayed concrete]
369 Samuely, Felix discovery, 80
concrete colleges, Hatfield, 34, 34 dry-mix, 80, 321
nineteenth century, 361–362 flats, Kensington, 28, 29 shells, 37
inter-war, 362–363, 362, 363 Penguin Pool, London Zoo, 28, 90, tunnels, 315, 320–321
concrete mix, 361 90, 91 wet-mix, 80, 321
embankments, crib wall roofs, folded plate, 36–37, 36 shuttering
stabilization, 151–152 schools, CLASP system, 95, 107 reusable, 95, 96
exposed-aggregate, 364 Screwcrete piling system, 144, Tall’s system, 14
inter-war schemes, 221 286–287 Simplex piles, 127, 129, 135, 135
jointed sewers slope stabilization, crib walls, 151–152
reinforced, 360, 360 concrete, chemical action on, 324 Smeaton, John, 13, 46
unreinforced, 359–360, 360 pipes Smirke, Sir Robert
laying Bonna system, 324–325, 325, 326 Custom House, 46, 48, 48, 118
inter-war, 362–363, 363 Monier system, 324 lime concrete, 48, 48, 118
post-war, 364, 365, 366–367, 366, tanks, 328, 330 spalling, repairing, 6
369 Shaw, Norman, 62 sprayed concrete see shotcrete
M25, 367–368, 369 shear stresses stadia, Wembley, 85, 86
Milan-Lake Maggiore autostrada, recognition, 84 Stahlton prestressed floors, 195, 203
364 reinforcing bars, 67, 89 staircases
noise levels, 363–364 shells reverse spiral beam, 31, 31
Padiham bypass, 366–367 barrel vaulted, 178, 179, 180–181, spiral, 16–17
pavements, 359–360, 359 180, 187, 187 stress, units defined, 84
Roman, 361, 361 bending effects, 166 structural engineering, definition, 305
St Albans bypass, 365, 366, 366 concrete, specifications, 185 structure, and architects, 28, 34
subgrade, 359, 359 construction constraints, 168–170 surfaces
surface map cracking, 367, 368 Continental development, 170–172, Coignet patent, 13–14
unjointed, continuously reinforced, 171, 172 exposed aggregate, 5
360, 360, 367–368 ctesiphon, 37, 38, 173 textures, 4–5, 5, 354–355, 355t
roller compacted concrete [RCC], 349, cylindrial, edge-supported, 172, swimming baths
355–356, 355t, 356t 172, 173, 174 composite construction, 333
Roman cement design ferro-concrete, 335
brick beams, 52 constraints, 183 mass concrete, 333–334
discovery, 13, 46 detailing, 184–185 municipal, 332–333
hydraulic properties, 13, 117–118, ultimate load, 37 reinforced concrete, 334–335, 335
120, 344 development, 35
manufacture, 344 finishing coats, 185, 185 Tall, Joseph, 14, 59

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:34:05
24_index.qxd 16/11/2001 16:22 Page 437

Index 437

Tanner, Sir Henry, 73, 74, 305 sealants, 320 Truscon, 331
Taylor, George Ledwell, 119–120, 119 shotcrete, 315, 320–321 water-cement ratio
Telford, Thomas, 117–118 Thames, 46 concrete durability, 109
tensile stresses, reinforcing bars, 67, 83 water flow, 316 concrete strength, 77, 109, 348
terras, hydraulic cements, 46, 118 Twisteel Reinforcements, 177–178 vibrated concrete, 95
theatres, fireproof floors, 14 waterproofing
tile creasing, flat slab floors, 54–55, 55, vaulting additives, 324
56 see also barrel vaulting, bridges, slab aggregates, 324
Toroja, Eduardo vault clay puddle, 323
aqueducts, Alloz, 30, 30 conoid, 22, 22 linings, 323, 324
domes, Ageciras, 29, 30 continuous, 22, 22 Wells system, 69t
shells, Madrid, 172, 172 reservoirs, 326, 328 whole-life costings, 114
three-dimensional concepts, 29, 30, vibration, mechanical, 77, 89, 94, 95 Wilkinson, William Boutland, 15, 55,
43 Vibro piles, 142 56, 67
torsional stresses, reinforcement, 90 Williams, Sir Owen
Trussed Concrete Steel Co [Truscon], warehouses, CWS Newcastle, 70, 71, bridges, Scotland, 27, 27, 224,
68 74, 130 225
bridges, 218, 219 water retaining structures counterbalanced roofs, 35, 35
Kahn system, 72, 218, 219 see also gasholder tanks; reservoirs; flat slab construction, 86, 87, 88
water towers, 331 sewers; swimming baths; water Wembley Stadium, 85, 86
tunnels towers Williams system, specification, 69t
concrete additives, 318 culverts, 325–326, 327 wind loads, tall offices, 39
in situ linings, 315 ship test tanks, 337–338, 337 woodwool cement, 79
early use, 316, 317 water towers worker training, concrete construction,
shields, 316–317, 318 BRC, 331, 333 94
steel formwork, 317–318, 317 designs, 330, 331 Wright, Frank Lloyd, 24, 26, 26, 43
tunnel borers, 318 Indented Bar, 331
linings, purposes, 315–316 joint sealing, 323, 330 X-joists, prestressed concrete, 206,
precast units, 315, 319 Kahn bar system, 34, 331, 332 206
Don-Seg, 319–320, 319 Mouchel, 330, 330, 331, 331
reinforcement, 320 reinforcement, 330, 331 yield stresses, reinforcement, 94

Delivered by ICEVirtualLibrary.com to:


IP: 130.194.20.173
On: Tue, 17 May 2011 06:34:05

You might also like