Foseco Ferrous Foundryman’s Handbook

Foseco Ferrous
Foundryman’s Handbook

Edited by
John R. Brown

OXFORD AUCKLAND BOSTON JOHANNESBURG MELBOURNE NEW DELHI

Butterworth-Heinemann
Linacre House, Jordan Hill, Oxford OX2 8DP
225 Wildwood Avenue, Woburn, MA 01801-2041
A division of Reed Educational and Professional Publishing Ltd

A member of the Reed Elsevier plc group

First published 2000

© Foseco International Ltd 2000

All rights reserved. No part of this publication

may be reproduced in any material form (including
photocopying or storing in any medium by electronic
means and whether or not transiently or incidentally
to some other use of this publication) without the
written permission of the copyright holder except
in accordance with the provisions of the Copyright,
Designs and Patents Act 1988 or under the terms of a
licence issued by the Copyright Licensing Agency Ltd,
90 Tottenham Court Road, London, England W1P 9HE.
Applications for the copyright holder’s written permission
to reproduce any part of this publication should be
addressed to the publishers

British Library Cataloguing in Publication Data

Brown, John R.
Foseco ferrous foundryman’s handbook.
1. Founding – Handbooks, manuals, etc.
I Title II Foseco International III Foundryman’s handbook
671.2

Library of Congress Cataloguing in Publication Data

Foseco ferrous foundryman’s handbook/revised and edited by John R. Brown.
p. cm.
Includes index.
ISBN 0 7506 4284 X
1 Founding – Handbooks, manuals, etc. 2 Iron founding – Handbooks,
manuals, etc. 3 Cast-iron – Handbooks, manuals, etc. I Title: Ferrous
foundryman’s handbook II Brown John R.

TS235 F35 2000

672.2–dc21 00-033679

ISBN 0 7506 4284 X

Typeset at Replika Press Pvt Ltd, Delhi 110 040, India

Printed and bound in Great Britain
Contents

Preface xi

Acknowledgements xii

Chapter 1 Tables and general data 1

SI units and their relation to other units 1

SI, metric, non-SI and non-metric conversions 2
Conversion table of stress values 5
Areas and volumes of circles, spheres, cylinders etc. 6
The physical properties of metals 7
Densities of casting alloys 9
Approximate bulk densities of common materials 10
Patternmakers’ contraction allowances 11
Volume shrinkage of principal casting alloys 13
Comparison of sieve sizes 14
Calculation of average grain size 15
Calculation of AFS grain fineness number 16
Recommended standard colours for patterns 17
Dust control in foundries 18
Buoyancy forces on cores 18
Opening forces on moulds 19
Dimensional tolerances and consistency achieved in castings 21

Chapter 2 Types of cast iron 23

Introduction 23
Physical properties of cast irons 27
Iron casting processes 28

Chapter 3 Grey cast iron 30

Specifications 30
Relationship between composition, strength and structure of
grey cast iron 32
Applications of grey iron castings 37
The production of grey irons 37
vi Contents

Chapter 4 Melting cast iron 40

Introduction 40
Cupola melting 40
Electric melting 53
Shop floor control of metal composition 60

Chapter 5 Inoculation of grey cast iron 62

Introduction 62
Ladle inoculation 64
Late stream inoculation 66
Mould inoculation 69

Chapter 6 Ductile iron 70

Production of ductile iron 70

Melting ductile iron base 74
Cupola melting and duplexing 74
Induction furnace melting 75
Use of the tundish cover ladle 75
Sandwich treatment 77
NODULANT 77
Pure magnesium converter process 77
Cored wire treatment 78
In-the-mould treatment 78
Inhibiting elements 78
Inoculation and fading 79
Specifications for ductile cast iron 79
Heat treatment of ductile iron 82
Casting ductile iron 84
Compacted graphite irons 84

Chapter 7 Malleable cast iron 90

Introduction 90
Whiteheart malleable 90
Blackheart malleable iron 92

Chapter 8 Special purpose cast irons 95

Heat resisting alloys 95

Corrosion resistant cast irons 101
Wear resistant cast irons 102
 Contents vii

Chapter 9 Types of steel castings 108

Introduction 108
Specifications for steel castings 109
Types of steel castings 111
Carbon steels 112
Low alloy and medium alloy steels 112
Austenitic manganese steel 114
High alloy steels 114
Duplex steels 117
Physical properties of steels 117
Selection of suitable steel for casting 117

Chapter 10 Melting and treatment of steel for casting 121

Arc furnace melting 121

Induction furnace melting 123
AOD refining 125
Melting and casting quality 125

Chapter 11 Molten metal handling 130

Iron foundries 130

Metal handling systems 130
Ladle lining 130
KALTEK ladle lining system for iron foundries 132
KALTEK insulating lining system for automatic pouring boxes 136
Casting defects due to poor ladle maintenance 137
Molten metal handling in steel foundries 138
Ladle practice 138
The KALTEK ladle lining system 142
Pouring temperature for steels 144

Chapter 12 Sands and green sand 146

Silica sand 146

Measurement of sand properties 149
Non-silica sands 151
Green sand 152
Parting agents 163
Special moulding materials, LUTRON 163
Green sand moulding machines 164
viii Contents

Chapter 13 Resin bonded sand 167

Chemical binders 167

Self-hardening process 167
Sand reclamation 173
Self-hardening resin binder systems 180
Triggered hardening systems 184
Heat triggered processes 186
Gas triggered processes 192
Review of resin coremaking processes 198

Chapter 14 Sodium silicate bonded sand 204

CO2 silicate process (basic process) 204

Improvements to the CO2 silicate process 207
Self-setting sodium silicate processes 209
Adhesives and sealants 213

Chapter 15 Lost foam casting 216

Principle of the process 216

Patternmaking 217
Assembling clusters 218
Coating the patterns 218
Investing in sand 218
The mechanism of casting into foam patterns 218
Advantages of lost foam casting 221
Disadvantages 222
Applications 222
The future 224

Chapter 16 Coatings for moulds and cores 226

The need for a coating 226

Choice of coating and form of supply 227
Components of a coating 228
Application methods for coatings 230
Coatings for iron and steel foundries 234
Coatings for high production foundries 234
Coatings for jobbing moulds and cores 237
Spirit based coatings 237
The TRIBONOL process 240
Miscellaneous coatings 241
Coatings for foundry tools 243
 Contents ix

Chapter 17 Filtration and the running and gating of iron castings 245

Introduction 245
Conventional running systems without filters 245
Filtration of iron castings 250
SEDEX ceramic foam filters 256
Cellular ceramic filters 259
Combined filter, feeder and pouring cup, the KALPUR direct
pouring system 266

Chapter 18 Filtration and the running and gating of steel castings 272

Introduction 272
Controlling the flow of metal 272
Conventional running systems without filters 274
The use of ceramic foam filters 277
Inclusions in steel castings 277
STELEX ZR ceramic foam filters 279
KALPUR ST direct pour unit 286
Cost savings through the use of STELEX and KALPUR 294

Chapter 19 Feeding of castings 296

Introduction 296
Natural feeders 296
Aided feeders – feeding systems 297
The calculation of feeder dimensions 301
Determination of feeding requirements 305
Steel, malleable iron, white irons, light alloys and copper based
alloy castings 305
Grey and ductile irons 307
Foseco feeding systems 310
KALPUR filter feeder units 322
Breaker cores 323
Application of feeder sleeves 325
FERRUX anti-piping compounds for iron and steel castings 331
Aids to the calculation of feeder requirements 334
FEEDERCALC 335

Chapter 20 Computer simulation of casting processes 344

Introduction 344
Solidification modelling 344
x Contents

Mould filling simulation 346

The SOLSTAR solidification program 346
Cost benefits of solidification simulation 351
Conclusions 352

Index 353
Preface

The last edition of the Foseco Foundryman’s Handbook was published in 1994
and like all the earlier editions, it aimed to provide a practical reference
book for all those involved in making castings in any of the commonly used
alloys by any of the usual moulding methods. In order to keep the Handbook
to a reasonable size, it was not possible to deal with all the common casting
alloys in detail. Since 1994 the technology of casting has continued to develop
and has become more specialised so that it has been decided to publish the
new edition of the Handbook in two volumes:

Ferrous dealing with grey, ductile and special purpose cast irons
together with carbon, low alloy and high alloy steels
Non-ferrous dealing with aluminium, copper and magnesium casting
alloys

Certain chapters (with slight modifications) are common to both volumes:

these chapters include tables and general data, sands and sand bonding
systems, resin bonded sand, sodium silicate bonded sand and feeding systems.
The remaining chapters have been written specifically for each volume.
The Handbook refers to many Foseco products. Not all of the products are
available in every country and in a few cases, product names may vary.
Users should always contact their local Foseco company to check whether
a particular product or its equivalent is available.
The Foseco logo and all product names appearing in capital letters are
trademarks of the Foseco group of companies, used under licence.

John R. Brown
Acknowledgements

The following Organisations have generously permitted the use of their

material in this book:

The American Foundryman’s Society Inc., 505 State Street, Des Plaines,
Illinois 60016-8399, USA
British Standards Institution (BSI), Extracts from British Standards are
reproduced with the permission of BSI. Complete copies can be obtained
by post from BSI Customer Services, 389 Chiswick High Road, London
W4 4AL
Butterworth-Heinemann, Linacre House, Jordan Hill, Oxford OX2 8DP
The Castings Development Centre (incorporating BCIRA), Bordesley Hall,
The Holloway, Alvechurch, Birmingham, B48 7QB
The Castings Development Centre (incorporating Steel Castings Research
& Trade Association), 7 East Bank Road, Sheffield, S2 3PT
The Department of the Environment, Transport and the Regions
Foundry Management & Technology, Penton Media Inc., 1100 Superior
Avenue, Cleveland OH 44114-2543, USA
Foundry & Technical Liaison Ltd., 6–11 Riley Street, Willenhall, West Midlands,
WV13 1RV
Institute of British Foundrymen, Bordesley Hall, The Holloway, Alvechurch,
Birmingham, B48 7QA
Rio Tinto Iron & Titanium GmbH, Eschborn, Germany
SinterCast Ltd, Regal House, 70 London Road, Twickenham, Middlesex

The author gratefully acknowledges the help received from colleagues at

Foseco International Limited, Foseco UK Limited and other Foseco Companies.
Chapter 1
Tables and general data

SI units and their relation to other units

The International System of Units (SI System) is based on six primary
units:

Quantity Unit Symbol

length metre m
mass kilogram kg
time second s
electric current ampere A
temperature degree Kelvin K
luminous intensity candela cd

Multiples

SI prefixes are used to indicate multiples and submultiples such as 106

or 10–3

Prefix Symbol Prefix Symbol

–1
10 deca da 10 deci d
102 hecto h 10–2 centi c
103 kilo k 10–3 milli m
106 mega M 10–6 micro µ
109 giga G 10–9 nano n
1012 tera T 10–12 pico p

Example: One millionth of a metre is expressed as one micrometre, 1 µm.

2 Foseco Ferrous Foundryman’s Handbook

Derived units

The most important derived units for the foundryman are:

Quantity Unit Symbol

Force newton N (kg m/s2)

Pressure, stress newton per square metre or pascal N/m2 (Pa)
Work, energy joule J (Nm)
Power, heat flow rate watt, joule per second W (J/s)
Temperature degree Celsius °C
Heat flow rate watt per square metre W/m2
Thermal conductivity watt per metre degree W/m K
Specific heat capacity joule per kilogram degree J/kg K
Specific latent heat joule per kilogram J/kg

SI, metric, non-SI and non-metric conversions

Length:
1 in = 25.4 mm
1 ft = 0.3048 m
1m = 1.09361 yd
1 km = 1093.61 yd = 0.621371 miles
1 mile = 1.60934 km = 1760 yd
1 yd = 0.9144 m

Area:
1 in2 = 654.16 mm2
1 ft2 = 0.092903 m2
1 m2 = 1.19599 yd2 = 10.76391 ft2
1 mm2 = 0.00155 in2
1 yd2 = 0.836127 m2
1 acre = 4840 yd2 = 4046.86 m2 = 0.404686 m2 hectare
1 hectare = 2.47105 acre = 10 000 m2

Volume:
1 cm3 = 0.061024 in3
1 dm3 = 11 (litre) = 0.035315 ft3
1 ft3 = 0.028317 m3 = 6.22883 gal (imp)
1 gal (imp) = 4.54609 l (litre)
1 in3 = 16.3871 cm3
1 l (litre) = 1 dm3 = 0.001 m3 = 0.21997 gal (imp)
1 m3 = 1.30795 yd3 = 35.31467 ft3
1 pt (pint) = 0.568261 l
1 US gal = 3.78541 l = 0.832674 gal (imp)
 Tables and general data 3

1 ft3/min (cfm) = 1.699 m3/h

1 ft3/sec = 28.31681/s

Mass:
1 lb (pound) = 0.453592 kg
1 cwt = 50.802 kg
1 kg = 2.20462 lb
1 oz = 28.349 gm
1 ton = 2240 lb = 1.01605 t (tonne) = 1016.05 kg
1 ton (US) = 2000 lb = 907.185 kg

Force:
1 kgf = 9.80665 N = 2.20462 lbf = 1 kp (kilopond)
1 lbf = 4.44822 N
1 pdl (poundal) = 0.138255 N

Density:
1 kgf/m3 = 0.062428 lb/ft3
1 lb/ft3 = 16.0185 kg/m3
1 g/cm3 = 1000 kg/m3

Pressure, stress:
1 kgf/cm2 = 98.0665 kPa (kN/m2)
1 kgf/mm2 = 9.80665 N/mm2 = 1422.33 lbf/in2 = 0.63497 tonf/in2
1 lbf/in2 (psi) = 6.89476 kPa (kN/m2)
1 Pa (N/m2) = 0.000145038 lbf/in2
1 in w.g. (in H2O) = 249.089 Pa
1 N/mm2 = 1 MPa = 145.038 lbf/in2 = 0.06475 tonf/in2
= 0.10197 kgf/cm2

Power:
1 kW = 3412 Btu/hr
1 hp (horsepower) = 0.745700 kW

Energy, heat, work:

1 Btu = 1.05506 kJ
1 cal = 4.1868 J
1 kWh = 3.6 MJ = 3412 Btu
1 therm = 100 000 Btu = 105.506 MJ
1 kJ = 0.277778 W.h

Specific heat capacity, heat transfer:

1 cal/g°C = 1 kcal/kg°C = 4186.8 J/kg.K
1 Btu/lb°F = 4186.8 J/kg.K
1 Btu/h = 0.293071 W
1 cal/cm.s°C = 418.68 W/m.K (thermal conductivity)
4 Foseco Ferrous Foundryman’s Handbook

1 [Link]/ft2h°F = 0.144228 W/m.K (thermal conductivity)

1 Btu/ft2h°F = 5.67826 W/m2.K (heat transfer coeff.)

Miscellaneous:
1 [Link]. = 101.325 kPa = 760 mm Hg = 1.01325 bar
1 bar = 100 kPa = 14.5038 lbf/in2
1 cP (centipoise) = 1 mPa.s
1 cSt (centistoke) = 1 mm2/s
1 cycle/s = 1 Hz (Hertz)
1 hp = 745.7 W

Useful approximations:
1 Btu = 1 kJ 1 kg = 21/4 lb
1 ft = 30 cm 1 kgf = 10 N
1 gal = 41 / 2 l 1 std atmos. = 1 bar
1 ha = 21/2 acre 1 km = 5
/8 mile
1 hp = 3/4 kW 1 litre = 13/4 pint
1 in = 25 mm 1 lbf = 41/2 N
1 therm = 100 MJ 1 yd = 0.9 m
1 tonf/in2 = 15 N/mm2
1 psi (lbf/in2) = 7 kPa
1 N (newton) = the weight of a small apple!

Temperature:
°F = 1.8 × °C + 32
°C = (°F – 32)/1.8
0°C (Celsius) = 273.15 K (Kelvin)
 Tables and general data 5

Conversion table of stress values

Equivalent stresses
American British Metric SI
(lb/in2) (ton/in2) (kgf/mm2) (N/mm2)

250 0.112 0.176 1.724

500 0.223 0.352 3.447
1000 0.446 0.703 6.895
2000 0.893 1.406 13.789
3000 1.339 2.109 20.684
4000 1.788 2.812 27.579
5000 2.232 3.515 34.474
10 000 4.464 7.031 68.947
15 000 6.696 10.546 103.421
20 000 8.929 14.062 137.894
25 000 11.161 17.577 172.368
30 000 13.393 21.092 206.841
35 000 15.652 24.608 241.315
40 000 17.875 28.123 275.788
45 000 20.089 31.639 310.262
50 000 22.321 35.154 344.735
55 000 24.554 38.670 379.209
60 000 26.786 42.185 413.682
65 000 29.018 45.700 448.156
70 000 31.250 49.216 482.629
75 000 33.482 52.731 517.103
80 000 35.714 56.247 551.576
85 000 37.946 59.762 586.050
90 000 40.179 63.277 620.523
95 000 42.411 66.793 654.997
100 000 44.643 70.308 689.470

Conversions
10 000 4.464 7.031 68.947
22 399 10 15.749 154.438
14 223 6.349 10 98.066
14 504 6.475 10.197 100
6 Foseco Ferrous Foundryman’s Handbook

Areas and volumes of circles, spheres, cylinders etc.

π = 3.14159 (approximation: 22/7)
1 radian = 57.296 degrees
Circle; radius r, diameter d:
circumference = 2πr = πd
area = πr2 = π/4 × d2
Sphere; radius r:
surface area = 4 πr2
volume = 4/3 πr3
Cylinder; radius of base r, height h:
area of curved surface = 2πrh
volume = πr2h
Cone; radius of base r, height h:
volume = 1/2 area of base × height
= 1/2 πr2h
Triangle; base b, height h:
area = 1/2bh
 Tables and general data 7

The physical properties of metals

Element Symbol Atomic Melting Boiling Latent heat of Mean specific heat
weight point point fusion 0–100°C
(°C) (°C)
(kJ/kg) (cal/g) (kJ/kg.K) (cal/g°C)

Aluminium Al 26.97 660.4 2520 386.8 92.4 0.917 0.219

Antimony Sb 121.76 630.7 1590 101.7 24.3 0.209 0.050
Arsenic As 74.93 volat. 616 – – 0.331 0.079
Barium Ba 137.37 729 2130 – – 0.285 0.068
Beryllium Be 9.02 1287 2470 133.5 31.9 2.052 0.490
Bismuth Bi 209.0 217.4 1564 54.4 13.0 0.125 0.030
Cadmium Cd 112.41 321.1 767 58.6 14.0 0.233 0.056
Calcium Ca 40.08 839 1484 328.6 78.5 0.624 0.149
Carbon C 12.01 – – – – 0.703 0.168
Cerium Ce 140.13 798 3430 – – 0.188 0.045
Chromium Cr 52.01 1860 2680 132.7 31.7 0.461 0.110
Cobalt Co 58.94 1494 2930 244.5 58.4 0.427 0.102
Copper Cu 63.57 1085 2560 180.0 43.0 0.386 0.092
Gallium Ga 69.74 29.7 2205 80.2 19.2 0.377 0.090
Gold Au 197.2 1064.4 2860 67.4 16.1 0.130 0.031
Indium In 114.8 156 2070 – – 0.243 0.058
Iridium Ir 193.1 2447 4390 – – 0.131 0.031
Iron Fe 55.84 1536 2860 200.5 47.9 0.456 0.109
Lead Pb 207.22 327.5 1750 20.9 5.0 0.130 0.031
Lithium Li 6.94 181 1342 137.4 32.8 3.517 0.840
Magnesium Mg 24.32 649 1090 194.7 46.5 1.038 0.248
Manganese Mn 54.93 1244 2060 152.8 36.5 0.486 0.116
Mercury Hg 200.61 –38.9 357 12.6 3.0 0.138 0.033
Molybdenum Mo 96.0 2615 4610 – – 0.251 0.060
Nickel Ni 58.69 1455 2915 305.6 73.0 0.452 0.108
Niobium Nb 92.91 2467 4740 – – 0.268 0.064
Osmium Os 190.9 3030 5000 – – 0.130 0.031
Palladium Pd 106.7 1554 2960 150.7 36.0 0.247 0.059
Phosphorus P 31.04 44.1 279 20.9 5.0 0.791 0.189
Platinum Pt 195.23 1770 3830 113.0 27.0 0.134 0.032
Potassium K 39.1 63.2 759 67.0 16.0 0.754 0.180
Rhodium Rh 102.91 1966 3700 – – 0.243 0.058
Silicon Si 28.3 1412 3270 502.4 120.0 0.729 0.174
Silver Ag 107.88 961.9 2163 92.1 22.0 0.234 0.055
Sodium Na 23.00 97.8 883 115.1 27.5 1.227 0.293
Strontium Sr 87.63 770 1375 – – 0.737 0.176
Sulphur S 32.0 115 444.5 32.7 9.0 0.068 0.016
Tantalum Ta 180.8 2980 5370 154.9 37.0 0.142 0.034
Tellurium Te 127.6 450 988 31.0 7.4 0.134 0.032
Thallium Tl 204 304 1473 – – 0.130 0.031
Tin Sn 118.7 232 2625 61.1 14.6 0.226 0.054
Titanium Ti 47.9 1667 3285 376.8 90.0 0.528 0.126
Tungsten W 184.0 3387 5555 167.5 40.0 0.138 0.033
Uranium U 238.2 1132 4400 – – 0.117 0.029
Vanadium V 50.95 1902 3410 334.9 80.0 0.498 0.119
Zinc Zn 65.38 419.6 911 110.1 26.3 0.394 0.094
Zirconium Zr 90.6 1852 4400 – – 0.289 0.069
8 Foseco Ferrous Foundryman’s Handbook

The physical properties of metals (continued)

Element Thermal Resistivity Vol. change Density Coeff. of Brinell

conductivity (µ[Link] on melting (g/cm3) expansion hardness
(W/m.K) at 20°C) (%) (× 10–6/K) no.

Al 238 2.67 6.6 2.70 23.5 17

Sb 23.8 40.1 1.4 6.68 11 30
As – 33.3 – 5.73 5.6 –
Ba – 60 – 3.5 18 –
Be 194 3.3 – 1.85 12 –
Bi 9 117 –3.3 9.80 13.4 9
Cd 103 7.3 4.7 8.64 31 20
Ca 125 3.7 – 1.54 22 13
C 16.3 – – 2.30 7.9 –
Ce 11.9 85.4 – 6.75 8 –
Cr 91.3 13.2 – 7.10 6.5 350
Co 96 6.3 – 8.90 12.5 125
Cu 397 1.69 4.1 8.96 17 48
Ga 41 – – 5.91 18.3 –
Au 316 2.2 5.2 19.3 14.1 18.5
In 80 8.8 – 7.3 24.8 1
Ir 147 5.1 – 22.4 6.8 172
Fe 78 10.1 5.5 7.87 12.1 66
Pb 35 20.6 3.4 11.68 29 5.5
Li 76 9.3 1.5 0.53 56 –
Mg 156 4.2 4.2 1.74 26 25
Mn 7.8 160 – 7.4 23 –
Hg 8.7 96 3.75 13.55 61 –
Mo 137 5.7 – 10.2 5.1 147
Ni 89 6.9 – 8.9 13.3 80
Nb 54 16 – 8.6 7.2 –
Os 87 8.8 – 22.5 4.6 –
Pd 75 10.8 – 12.0 11.0 50
P – – – 1.83 6.2 –
Pt 73 10.6 – 21.45 9.0 52
K 104 6.8 2.8 0.86 83 0.04
Rh 148 4.7 – 12.4 8.5 156
Si 139 103–106 – 2.34 7.6 –
Ag 425 1.6 4.5 10.5 19.1 25
Na 128 4.7 2.5 0.97 71 0.1
Sr – 23 – 2.6 100 –
S 272 – – 2.07 70 –
Ta 58 13.5 – 16.6 6.5 40
Te 3.8 1.6 × 105 – 6.24 – –
Tl 45.5 16.6 – 11.85 30 –
Sn 73.2 12.6 2.8 7.3 23.5 –
Ti 21.6 54 – 4.5 8.9 –
W 174 5.4 – 19.3 4.5 –
U 28 27 – 19.0 – –
V 31.6 19.6 – 6.1 8.3 –
Zn 120 6.0 6.5 7.14 31 35
Zr 22.6 44 – 6.49 5.9 –
 Tables and general data 9

Densities of casting alloys

Alloy BS1490 g/ml Alloy BS1400 g/ml

Aluminium alloys Copper alloys

Pure Al 2.70 HC copper HCC1 8.9
Al–Si5Cu3 LM4 2.75 Brass CuZn38Al DCB1 8.5
Al–Si7Mg LM25 2.68 CuZn33Pb2Si HTB1 8.5
Al–Si8Cu3Fe LM24 2.79 CuZn33Pb2 SCB3 8.5
AlSi12 LM6 2.65 Phosphor bronze
CuSn11P PB1 8.8
Cast steels CuSn12 PB2 8.7
Low carbon <0.20 7.86 Lead bronze
Med. carbon 0.40 7.86 CuSn5Pb20 LB5 9.3
High carbon >0.40 7.84 Al bronze
CuAl10Fe2 AB1 7.5
Low alloy 7.86 Gunmetal
Med. alloy 7.78 CuSnPb5Zn5 LG2 8.8
Med./high alloy 7.67 Copper nickel
CuNi30Cr2FeMnSi CN1 8.8
Stainless
13Cr 7.61 Cast irons
18Cr8Ni 7.75 Grey iron 150 MPa 6.8–7.1
200 7.0–7.2
Other alloys 250 7.2–7.4
Zinc base 300 7.3–7.4
ZnAl4Cu1 6.70 Whiteheart malleable 7.45
Blackheart malleable 7.27
Lead base White iron 7.70
PbSb6 10.88 Ductile iron (s.g.) 7.2–7.3
Tin base (Babbit) 7.34 Ni-hard 7.6–7.7
Inconel Ni76Cr18 8.50 High silicon (15%) 6.8
10 Foseco Ferrous Foundryman’s Handbook

Approximate bulk densities of common materials

Material kg/m3 lb/ft3 Material kg/m3 lb/ft3

Aluminium, cast 2560 160 Lead 11370 710

wrought 2675 167 Limestone 2530–2700 158–168
Aluminium bronze 7610 475
Ashes 590 37 Magnesite 2530 158
Mercury 13 560 847
Brass, rolled 8390 524 Monel 8870 554
swarf 2500 175
Babbit metal 7270 454 Nickel, cast 8270 516
Brick, common 1360–1890 85–118 Nickel silver 8270 516
fireclay 1840 115
Bronze 8550 534 Phosphor bronze 8580 536
Pig iron, mean 4800 300
Cast iron, solid 7210 450 Pig iron and scrap
turnings 2240 140 (cupola charge) 5400 336
Cement, loose 1360 85
Chalk 2240 140 Sand, moulding 1200–1440 75–90
Charcoal, lump 290 18 silica 1360–1440 85–90
Clay 1900–2200 120–135 Silver, cast 10 500 656
Coal 960–1280 60–80 Steel 7850 490
Coal dust 850 53
Coke 450 28 Tin 7260 453
Concrete 2240 140
Copper, cast 8780 548 Water, ice 940 58.7
Cupola slag 2400 150 liquid 0°C 1000 62.4
100°C 955 59.6
Dolomite 2680 167 Wood, balsa 100–130 7–8
oak 830 52
Fire clay 1440 90 pine 480 30
French chalk 2600 162 teak 640 40
Wrought iron 7700 480
Glass 2230 139
Gold, pure 19 200 1200 Zinc, cast 6860 428
22 carat 17 500 1090 rolled 7180 448
Graphite, powder 480 30
solid 2200 138
 Tables and general data 11

Patternmakers’ contraction allowances

Castings are always smaller in dimensions than the pattern from which
they are made, because as the metal cools from its solidification temperature
to room temperature, thermal contraction occurs. Patternmakers allow for
this contraction by making patterns larger in dimensions than the required
castings by an amount known as the “contraction allowance”. Originally
this was done by making use of specially engraved rules, known as
“contraction rules”, the dimensions of which incorporated a contraction
allowance such as 1 in 75 for aluminium alloys, or 1 in 96 for iron castings.
Nowadays, most patterns and coreboxes are made using computer-controlled
machine tools and it is more convenient to express the contraction as a
percentage allowance.
Predicting casting contraction can never be precise, since many factors
are involved in determining the exact amount of contraction that occurs.
For example, when iron castings are made in greensand moulds, the mould
walls may move under the pressure of the liquid metal, causing expansion
of the mould cavity, thus compensating for some of the metal contraction.
Cored castings may not contract as much as expected, because the presence
of a strong core may restrict movement of the casting as it is cooling. Some
core binders expand with the heat of the cast metal causing the casting to be
larger than otherwise expected. For these reasons, and others, it is only
possible to predict contractions approximately, but if a patternmaker works
with a particular foundry for a long period, he will gain experience with the
foundry’s design of castings and with the casting methods used in the
foundry. Based on such experience, more precise contraction allowances
can be built into the patterns.
12 Foseco Ferrous Foundryman’s Handbook

The usually accepted contraction allowances for different alloys are given
in the following table.

Alloy Contraction allowance (%)

Aluminium alloys
Al–Si5Cu3 LM4
Al–Si7Mg LM25 1.3
Al–Si8Cu3Fe LM24
Al–Si12 LM6

Beryllium copper 1.6

Bismuth 1.3
Brass 1.56
Bronze, aluminium 2.32
manganese 0.83–1.56
phosphor 1.0–1.6
silicon 1.3–1.6

Cast iron, grey 0.9–1.04

white 2.0
ductile (s.g.) 0.6–0.8
malleable 1.0–1.4

Copper 1.6
Gunmetal 1.0–1.6
Lead 2.6
Magnesium alloys 1.30–1.43
Monel 2.0

Nickel alloys 2.0

Steel, carbon 1.6–2.0
chromium 2.0
manganese 1.6–2.6
Tin 2.0

White metal 0.6

Zinc alloys 1.18
 Tables and general data 13

Volume shrinkage of principal casting alloys

Most alloys shrink in volume when they solidify, the shrinkage can cause
voids in castings unless steps are taken to “feed” the shrinkage by the use
of feeders.

Casting alloy Volume shrinkage (%)

Carbon steel 6.0

Alloyed steel 9.0
High alloys steel 10.0

Malleable iron 5.0

Al 8.0
Al–Cu4Ni2Mg 5.3
Al–Si12 3.5
Al–Si5Cu2Mg 4.2
Al–Si9Mg 3.4
Al–Si5Cu1 4.9
Al–Si5Cu2 5.2
Al–Cu4 8.8
Al–Si10 5.0
Al–Si7NiMg 4.5
Al–Mg5Si 6.7
Al–Si7Cu2Mg 6.5
Al–Cu5 6.0
Al–Mg1Si 4.7
Al–Zn5Mg 4.7

Cu (pure) 4.0

Brass 6.5
Bronze 7.5

Al bronze 4.0
Sn bronze 4.5
14 Foseco Ferrous Foundryman’s Handbook

Comparison of sieve sizes

Sieves used for sand grading are of 200 mm diameter and are now usually
metric sizes, designated by their aperture size in micrometres (µm). The
table lists sieve sizes in the British Standard Metric series (BS410 : 1976)
together with other sieve types.

Sieve aperture, micrometres and sieve numbers

ISO/R.565 series BSS ASTM

(BS410:1976)
(µm) No. µm No. µm

(1000) 16 1003 18 1000

710 22 699 22 710
500 30 500 30 500
355 44 353 45 350
250 60 251 60 250
(212) 72 211 70 210
180
(150) 100 152 100 149
125 120 125
90 150 104 150 105
63 200 76 200 74
(45) 300 53 325 44

Notes: The 1000 and 45 sieves are optional.

The 212 and 150 sieves are also optional, but may be included to give better
separation between the 250 and 125 sieves.
 Tables and general data 15

Calculation of average grain size

The adoption of the ISO metric sieves means that the old AFS grain fineness
number can no longer be calculated. Instead, the average grain size, expressed
as micrometres (µm) is now used. This is determined as follows:

1 Weigh a 100 g sample of dry sand.

2 Place the sample into the top sieve of a nest of ISO sieves on a vibrator.
Vibrate for 15 minutes.
3 Remove the sieves and, beginning with the top sieve, weigh the quantity
of sand remaining on each sieve.
4 Calculate the percentage of the sample weight retained on each sieve,
and arrange in a column as shown in the example.
5 Multiply the percentage retained by the appropriate multiplier and add
the products.
6 Divide by the total of the percentages retained to give the average grain
size.

Example

ISO aperture Percentage Multiplier Product

(µm) retained

≥ 710 trace 1180 0

500 0.3 600 180
355 1.9 425 808
250 17.2 300 5160
212 25.3 212 5364
180 16.7 212 3540
150 19.2 150 2880
125 10.6 150 1590
90 6.5 106 689
63 1.4 75 105
≤63 0.5 38 19
Total 99.6 – 20 335

Average grain size = 20 335/99.6

= 204 µm
16 Foseco Ferrous Foundryman’s Handbook

Calculation of AFS grain fineness number

Using either the old BS sieves or AFS sieves, follow, steps 1–4 above.
5 Arrange the results as shown in the example below.
6 Multiply each percentage weight by the preceding sieve mesh number.
7 Divide by the total of the percentages to give the AFS grain fineness
number.

Example

BS sieve % sand retained Multiplied by Product

number on sieve previous sieve no.

10 nil – –
16 nil – –
22 0.2 16 3.2
30 0.8 22 17.6
44 6.7 30 201.0
60 22.6 44 1104.4
100 48.3 60 2898.0
150 15.6 100 1560.0
200 1.8 150 270.0
pan 4.0 200 800.0
Total 100.0 – 6854.2

AFS grain fineness number = 6854.2/100

= 68.5 or 68 AFS

Foundry sands usually fall into the range 150–400 µm, with 220–250 µm
being the most commonly used. Direct conversion between average grain
size and AFS grain fineness number is not possible, but an approximate
relation is shown below:

AFS grain
fineness no. 35 40 45 50 55 60 65 70 80 90
Average
grain size (µm) 390 340 300 280 240 220 210 195 170 150

While average grain size and AFS grain fineness number are useful
parameters, choice of sand should be based on particle size distribution.
 Tables and general data 17

Recommended standard colours for patterns

Part of pattern Colour

As-cast surfaces which are to be left unmachined Red or orange

Surfaces which are to be machined Yellow
Core prints for unmachined openings and end prints
Periphery Black
Ends Black
Core prints for machined openings Periphery Yellow stripes
on black
Ends Black
Pattern joint (split patterns) Cored section Black
Metal section Clear varnish
Touch core Cored shape Black
Legend “Touch”
Seats of and for loose pieces Green
and loose core prints
Stop offs Diagonal black
stripes with
clear varnish
Chilled surfaces Outlined in Black
Legend “Chill”
18 Foseco Ferrous Foundryman’s Handbook

Dust control in foundries

Air extraction is used in foundries to remove silica dust from areas occupied
by operators. The following table indicates the approximate air velocities
needed to entrain sand particles.

Terminal velocities of spherical particles of density 2.5 g/cm3 (approx.)

BS sieve Particle Terminal velocity

size dia. (µm) m/sec ft/sec ft/min

16 1003 7.0 23 1400

30 500 4.0 13 780
44 353 2.7 9 540
60 251 1.8 6 360
100 152 1.1 3.5 210
150 104 0.5 1.7 100
200 76 0.4 1.3 80

For the comfort and safety of operators, air flows of around 0.5 m/sec are
needed to carry away silica dust. If air flow rate is too high, around the
shake-out for example, there is a danger that the grading of the returned
sand will be altered.

Buoyancy forces on cores

When liquid metal fills a mould containing sand cores, the cores tend to
float and must be held in position by the core prints or by chaplets. The
following table lists the buoyancy forces experienced by silica sand
cores in various liquid metals, expressed as a proportion of the weight of
the core:

Liquid metal Ratio of buoyant force to core weight

Aluminium 0.66
Brass 4.25
Copper 4.50
Cast iron 3.50
Steel 3.90

Core print support

Moulding sand (green sand) in a core print will support about 150 kN/m2
(21 psi). So the core print can support the following load:
 Tables and general data 19

Support (kN) = Core print area (m2) × 150

1 kN = 100 kgf (approx.)
Support (kgf) = Core print area (m2) × 15 000

Example: A core weighing 50 kg has a core print area of 10 × 10 cm (the area

of the upper, support surface), i.e. 0.1 × 0.1 = 0.01 m2. The print support is
150 × 0.01 = 1.5 kN = 150 kgf.
If the mould is cast in iron, the buoyancy force is 50 × 3.5 = 175 kgf so
chaplets may be necessary to support the core unless the print area can be
increased.

Opening forces on moulds

Unless a mould is adequately clamped or weighted, the force exerted by the
molten metal will open the boxes and cause run-outs. If there are insufficient
box bars in a cope box, this same force can cause other problems like distortion
and sand lift. It is important therefore to be able to calculate the opening
force so that correct weighting or clamping systems can be used.
The major force lifting the cope of the mould is due to the metallostatic
pressure of the molten metal. This pressure is due to the height, or head, of
metal in the sprue above the top of the mould (H in Fig. 1.1). Additional
forces exist from the momentum of the metal as it fills the mould and from
forces transmitted to the cope via the core prints as the cores in cored castings
try to float.

Ladle

Bush
H Cope
F
Area A

Drag

Figure 1.1 Opening forces of moulds.

The momentum force is difficult to calculate but can be taken into account
by adding a 50% safety factor to the metallostatic force.
20 Foseco Ferrous Foundryman’s Handbook

The opening metallostatic force is calculated from the total upward-facing

area of the cope mould in contact with the metal (this includes the area of
all the mould cavities in the box). The force is:

F (kgf) = A × H × d × 1.5
1000
A is the upward facing area in cm2
H (cm) is the height of the top of the sprue above the average height of the
upward facing surface
d is the density of the molten metal (g/cm3)
1.5 is the “safety factor”

For ferrous metals, d is about 7.5, so:

F (kgf) = 11 × A × H
1000
For aluminium alloys, d is about 2.7, so:

F (kgf) = 4 × A × H
1000

Forces on cores

The core tends to float in the liquid metal and exerts a further upward force
(see page 18)
In the case of ferrous castings, this force is

3.5 × W (kgf)

where W is the weight of the cores in the mould (in kg).

In aluminium, the floating force can be neglected.
The total resultant force on the cope is (for ferrous metals)
(11 × A × H)/1000 + 3.5 W kgf

Example: Consider a furane resin mould for a large steel valve body casting
having an upward facing area of 2500 cm2 and a sprue height (H) of 30 cm
with a core weighing 40 kg. The opening force is

11 × 2500 × 30/1000 + 3.5 × 40 = 825 + 140

= 965 kgf
It is easy to see why such large weights are needed to support moulds in
jobbing foundries.
 Tables and general data 21

Dimensional tolerances and consistency

achieved in castings
Errors in dimensions of castings are of two kinds:
Accuracy: the variation of the mean dimension of the casting from
the design dimension given on the drawing
Consistency: statistical errors, comprising the dimensional variability
round the mean dimension

Dimensional accuracy

The major causes of deviations of the mean dimension from the target value
are contraction uncertainty and errors in pattern dimensions. It is usually
possible to improve accuracy considerably by alternations to pattern
equipment after the first sample castings have been made.

100

Steel
AI + malleable
10
White cast iron
Tolerance (2.5σ) (mm)

Grey cast
iron
Sand casting

1 AI
Gravity die and
low-pressure die
AI
AI Precision sand
Pressure die core assembly

0.1
Investment
Zn AI and steel
Pressure die

0.01
10 100 1000 10 000
Casting dimension (mm)

Figure 1.2 The average tolerance (taken as 2.5σ) exhibited by various casting
processes. (From Campbell, J. (1991) Castings, Butterworth-Heinemann, reproduced
by permission of the publishers.)
22 Foseco Ferrous Foundryman’s Handbook

Dimensional consistency

Changes in process variables during casting give rise to a statistical spread

of measurements about a mean value. If the mean can be made to coincide
with the nominal dimension by pattern modification, the characteristics of
this statistical distribution determine the tolerances feasible during a
production run.
The consistency of casting dimensions is dependent on the casting process
used and the degree of process control achieved in the foundry. Figure 1.2
illustrates the average tolerance exhibited by various casting processes. The
tolerance is expressed as 2.5σ (2.5 standard deviations), meaning that only
1 casting in 80 can be expected to have dimensions outside the tolerance.
There is an International Standard, ISO 8062–1984(E) Castings – System of
dimensional tolerances, which is applicable to the dimensions of cast metals
and their alloys produced by sand moulding, gravity diecasting, low pressure
diecasting, high pressure diecasting and investment casting. The Standard
defines 16 tolerance grades, designated CT1 to CT16, listing the total casting
tolerance for each grade on raw casting dimensions from 10 to 10 000 mm.
The Standard also indicates the tolerance grades which can be expected for
both long and short series production castings made by various processes
from investment casting to hand-moulded sand cast.
Reference should be made to ISO 8062 or the equivalent British Standard
BS6615:1985 for details.
Chapter 2
Types of cast iron

Introduction
The first iron castings to be made were cast directly from the blast furnace.
Liquid iron from a blast furnace contains around 4%C and up to 2%Si,
together with other chemical elements derived from the ore and other
constituents of the furnace charge. The presence of so much dissolved carbon
etc. lowers the melt point of the iron from 1536°C (pure iron) to a eutectic
temperature of about 1150°C (Fig. 2.1) so that blast furnace iron is fully
liquid and highly fluid at temperatures around 1200°C. When the iron
solidifies, most of the carbon is thrown out of solution in the form either of
graphite or of iron carbide, Fe3C, depending on the composition of the iron,
the rate of cooling from liquid to solid and the presence of nucleants.
If the carbon is precipitated as flake graphite, the casting is called ‘grey
iron’, because the fractured surface has a dull grey appearance due to the
presence of about 12% by volume of graphite. If the carbon precipitates as
carbide, the casting is said to be ‘white iron’ because the fracture has a shiny
white appearance. In the early days of cast iron technology, white iron was
of little value, being extremely brittle and so hard that it was unmachinable.
Grey iron, on the other hand, was soft and readily machined and although
it had little ductility, it was less brittle than white iron.
Iron castings were made as long ago as 500 BC (in China) and from the
15th century in Europe, when the blast furnace was developed. The great
merits of grey iron as a casting alloy, which still remain true today, are its
low cost, its high fluidity at modest temperatures and the fact that it freezes
with little volume change, since the volume expansion of the carbon
precipitating as graphite compensates for the shrinkage of the liquid iron.
This means that complex shapes can be cast without shrinkage defects.
These factors, together with its free-machining properties, account for the
continuing popularity of grey cast iron, which dominates world tonnages of
casting production (Table 2.1).
Greater understanding of the effect of chemical composition and of
nucleation of suitable forms of graphite through inoculation of liquid iron,
has vastly improved the reliability of grey iron as an engineering material.
Even so, the inherent lack of ductility due to the presence of so much graphite
precipitated in flake form (Fig. 2.2) limits the applications to which grey
iron can be put.
A malleable, or ductile form of cast iron was first made by casting ‘white
24 Foseco Ferrous Foundryman’s Handbook

δ + liquid

1600 δ

G + liquid
δ+γ γ+ liquid Fe3C + liquid
liquid
C
1200 A
Temperature (°C)

γ γ + Fe3C
α+γ or
γ+G
800

α D

α + Fe3C
or
400 α+G

0 2 4 6
wt. %C
Fe – G system Fe – Fe3C system
A B C D
(––) Fe–G 2.09 4.25 0.68 %C
1154 1154 739 °C
(---) Fe–Fe3C 2.12 4.31 6.68 0.76 %C
1148 1148 1226 727 °C

Figure 2.1 The iron–carbon phase diagram. (From Elliott, R., Cast Iron Technology,
1988, Butterworth-Heinemann, reproduced by permission of the publishers.)

Table 2.1 Breakdown of iron casting tonnages 1996 (1000s tonnes)

Total iron Grey iron Ductile iron Malleable iron

Germany 
France  6127 3669 (59.9%) 2368 (38.6%) 84 (1.37%)
UK 
USA 10 314 6048 (58.6%) 4034 (39.1%) 232 (2.25%)

Data from CAEF report The European Foundry Industry 1996

US data from Modern Castings

iron’ and then by a long heat treatment process, converting the iron carbide
to graphite. Under the right conditions the graphite developed in discrete,
roughly spherical aggregates (Fig. 2.3) so that the casting became ductile
with elongation of 10% or more. The first malleable iron, ‘whiteheart iron’
 Types of cast iron 25

Figure 2.2 Random flake graphite, 4% picral, ×100. (From BCIRA Broadsheet
138, reproduced by courtesy of CDC.)

was made by Réaumur in France in 1720. The more useful ‘blackheart’

malleable iron was developed in the USA by Boyden around 1830. Malleable
cast iron became a widely used casting alloy wherever resistance to shock
loading was required. It was particularly suitable for transmission components
for railways and automotive applications.
A major new development occurred in the late 1940s with the discovery
that iron having a nodular form of graphite could be cast directly from the

Figure 2.3 Malleable cast iron, 4% picral, ×100. (From BCIRA Broadsheet 138,
reproduced by courtesy of CDC.)
26 Foseco Ferrous Foundryman’s Handbook

melt after treatment of liquid iron of suitable composition with magnesium.

(Fig. 2.4). The use of ‘spheroidal graphite’ or ‘nodular’ iron castings has
since grown rapidly as the technology became understood and ‘ductile
iron’, as it is now generally known, has gained a large and still growing,
sector of total cast iron production (Table 2.1).

Figure 2.4 Nodular graphite, 4% picral ×100. (From BCIRA Broadsheet 138,
reproduced by courtesy of CDC.)

The great hardness and abrasion resistance of white iron has also been
exploited. The strength of white iron has been improved through alloying
and heat treatment, and white iron castings are widely used in applications
such as mineral processing, shot blasting etc. where the excellent wear
resistance can be fully used.
Finally there are a number of special cast irons designed to have properties
of heat resistance, or acid resistance etc. In the following chapters, each type
of iron will be considered separately and its method of production described.
The mechanical properties of cast iron are derived mainly from the matrix
and irons are frequently described in terms of their matrix structure, that is,
ferritic or pearlitic:

Ferrite is a Fe–C solid solution in which some Si, Mn, Cu etc. may also
dissolve. It is soft and has relatively low strength. Ferritic irons can be
produced as-cast or by annealing.
Pearlite is a mixture of lamellae of ferrite and Fe3C formed from austenite
by a eutectoid reaction. It is relatively hard and the mechanical properties
of a pearlitic iron are affected by the spacing of the pearlite lamellae,
which is affected by the rate of cooling of the iron from the eutectoid
temperature of around 730°C.
 Types of cast iron 27

Ferrite–Pearlite mixed structures are often present in iron castings.

Bainite is usually formed by an austempering heat treatment (normally
on spheroidal graphite irons) and produces high tensile strength with
toughness and good fatigue resistance.
Austenite is retained when iron of high alloy (nickel and chromium) content
cools. Heat and corrosion resistance are characteristics of austenitic irons.

Physical properties of cast irons

The physical properties of cast irons are affected by the amount and form of
the graphite and the microstructure of the matrix. Tables 2.2, 2.3, 2.4, 2.5 and
2.6 show, respectively, the density, electrical resistivity, thermal expansion,
specific heat capacity and thermal conductivity of cast irons. The figures in
the tables should be regarded as approximate.

Table 2.2 Density of cast irons

Grey iron
Tensile strength 150 180 220 260 300 350 400
(N/mm2)
Density at 20°C 7.05 7.10 7.15 7.20 7.25 7.30 7.30
(g/cm3)

Ductile iron
Grade 350/22 400/12 500/7 600/3 700/2
Density at 20°C 7.10 7.10 7.10–7.17 7.17–7.20 7.20
(g/cm3)

Malleable iron
Grade 350/10 450/6 550/4 600/3 700/2
Density at 20°C 7.35 7.30 7.30 7.30 7.30
(g/cm3)

Other cast irons

Type White cast irons Austenitic Grey
Unalloyed 15–30%Cr Ni-Cr (Ni-hard) high-Si (6%)
Density at 20°C 7.6–7.8 7.3–7.5 7.6–7.8 7.4–7.6 6.9–7.2
(g/cm3)
28 Foseco Ferrous Foundryman’s Handbook

Table 2.3 Electrical resistivity of cast irons

Grey iron
Tensile strength 150 180 220 260 300 350 400
(N/mm2)
Resistivity at 20°C 0.80 0.78 0.76 0.73 0.70 0.67 0.64
(micro-ohms.m2/m)

Ductile iron
Grade 350/22 400/12 500/7 600/3 700/2
Resistivity at 20°C 0.50 0.50 0.51 0.53 0.54
(micro-ohms.m2/m)

Malleable iron
Grade 350/10 450/6 550/4 600/3 700/2
Resistivity at 20°C 0.37 0.40 0.40 0.41 0.41
(micro-ohms.m2/m)

Table 2.4 Coefficient of linear thermal expansion for cast irons

Type of iron Typical coefficient of linear expansion for temperature ranges

(10–6 per °C)

20–100°C 20–200°C 20–300°C 20–400°C 20-500°C

Ferritic flake or nodular 11.2 11.9 12.5 13.0 13.4

Pearlitic flake or nodular 11.1 11.7 12.3 12.8 13.2
Ferritic malleable 12.0 12.5 12.9 13.3 13.7
Pearlitic malleable 11.7 12.2 12.7 13.1 13.5
White iron 8.1 9.5 10.6 11.6 12.5
14–22% Ni austenitic 16.1 17.3 18.3 19.1 19.6
36% Ni austenitic 4.7 7.0 9.2 10.9 12.1

Table 2.5 Specific heat capacity of cast irons

Typical mean values for grey, nodular and malleable irons, from room temperature to 1000°C
Mean value for each temperature range (J/kg.K)

20–100°C 20–200°C 20–300°C 20–400°C 20–500°C 20–600°C 20–700°C 20–800°C 20–900°C 20–1000°C
515 530 550 570 595 625 655 695 705 720

Typical mean values for grey, nodular and malleable irons, for 100°C ranges
Mean value for each temperature range, (J/kg.K)

100–200°C 200–300°C 300–400°C 400–500°C 500–600°C 600–700°C 700–800°C 800–900°C 900–1000°C

540 585 635 690 765 820 995 750 850

Iron casting processes

The majority of repetition iron castings are made in green sand moulds with
resin-bonded cores. The Croning resin shell moulding process is used where
 Types of cast iron 29

Table 2.6 Thermal conductivity of cast irons

Grey iron
Tensile strength 150 180 220 260 300 350 400
(N/mm2)
Thermal conductivity
(W/m.K)
100°C 65.6 59.5 53.6 50.2 47.7 45.3 45.3
500°C 40.9 40.0 38.9 38.0 37.4 36.7 36.0

Ductile iron
Grade 350/22 400/12 500/7 600/3 700/2
Thermal conductivity
(W/m.K)
100°C 40.2 38.5 36.0 32.9 29.8
500°C 36.0 35.0 33.5 31.6 29.8

Malleable iron
Grade 350/10 450/6 550/4 600/3 700/2
Thermal conductivity
(W/m.K)
100°C 40.4 38.1 35.2 34.3 30.8
500°C 34.6 34.1 32.0 31.4 28.9

high precision and good surface finish are needed. The Lost Foam Process
is also used for repetition castings. Castings made in smaller numbers are
made in chemically bonded sand moulds.
Special sand processes such as Vacuum Moulding and Full-Mould are
used for certain iron castings and there are a few permanent mould (diecasting)
foundries making iron castings, but the short die-life of only a few thousand
components has restricted the use of ferrous diecasting.
Chapter 3
Grey cast iron

Specifications
The properties of grey iron castings are affected by their chemical composition
and the cooling rate of the casting, which is influenced by the section thickness
and shape of the casting. With the widespread acceptance of SI units, there
has been a convergence of national specifications for grey cast iron based on
minimum tensile strength measured in N/mm2 (MPa) on a test piece machined
from a separately cast, 30 mm diameter bar, corresponding to a relevant
wall thickness of 15 mm. There is no requirement in terms of composition
and the foundryman is free to make his own choice based on the requirements
of the particular casting.
In 1997 a European Standard EN 1561:1997 was approved by CEN
(European Committee for Standardization). This standard has been given
the status of a national standard in countries which have CEN members.
CEN members are the national standards bodies of Austria, Belgium, Czech
Republic, Denmark, Finland, France, Germany, Greece, Iceland, Ireland,
Italy, Luxembourg, Netherlands, Norway, Portugal, Spain, Sweden,
Switzerland and United Kingdom.
In the UK, for example, the previously used standard BS 1452:1990 has
been withdrawn and replaced by BS EN 1561:1997.
The six ISO grades are generally recognised in most countries. The USA
continues to use grades based on tensile strengths measured in 1000s psi
but now has corresponding metric standards. The data in Table 3.1 is intended
to help foundrymen who are required to supply castings to other countries,
however, it must be understood that while there are similarities between
the various national standards, the grades are not necessarily identical. It is
advisable to consult the original standards for details of methods of testing
etc.
Many specifications also define grades of grey iron specified by their
hardness, related to section thickness. In the European Standard EN 1561:97
Brinell Hardness HB 30 is used, the mandatory values apply to Relevant
Wall Thickness of 40–80 mm (Table 3.2).
An approximate guide to variations in tensile strength with section
thickness is given in Fig. 3.1.
Table 3.1 National standards

Country Specification Designation Minimum tensile strength (N/mm2 )

100 150 180 200 250 275 300 350 400

Europe EN 1561:1997 EN-GJL- 100 150 200 250 300 350

Japan JIS G5501 1995 FC 100 150 200 250 300 350
Class 1 2 3 4 5 6

Russia GOST 1412 1979 Sch 10 15 18 20 25 30 35 40

USA ASTM A48-94a Grade 20 25 30 35 40 45 50 60

ASTM A48M-94 Grade 150 175 200 225 250 275 300 325 350 400

Inter- ISO 185-1988 Grade 100 150 200 250 300 350
national

Equivalent tonf/in2 6.5 9.7 12.9 16.2 19.4 22.7

Note: Consult National Specifications for full details.

32 Foseco Ferrous Foundryman’s Handbook

Table 3.2 HB30 hardness grades of grey iron (EN 1561:1997)

Grade EN-GJL- HB155 HB175 HB195 HB215 HB235 HB255

HB30 min – 100 120 145 165 185

max 155 175 195 215 235 255

Note: Consult EN 1561:1997 for details.

340

300
Tensile strength (N/mm2)

Grade 300
260

Grade 260
220
Grade 220
180
Grade 180
140
Grade 150
100
0 20 40 60 80 100 120 140 160 180
Equivalent to 30 mm diameter bar
Cross-section thickness (mm)

Figure 3.1 Variation of tensile strength with section thickness for several grades of
iron. (Data supplied by CDC.)

Relationship between composition, strength and

structure of grey cast iron
The properties of grey irons depend on the size, amount and distribution of
the graphite flakes and on the structure of the metal matrix. These, in turn,
depend on the chemical composition of the iron, in particular its carbon and
silicon content; and also on processing variables such as method of melting,
inoculation practice and the cooling rate of the casting.

Carbon equivalent

The three constituents of cast iron which most affect strength and hardness
are total carbon, silicon and phosphorus. An index known as the ‘carbon
equivalent value’ (CEV) combines the effects of these elements. The grey
iron eutectic occurs at a carbon content of 4.3% in the binary Fe–C system.
If silicon and phosphorus are present, the carbon content of the eutectic is
 Grey cast iron 33

lowered. The effect of Si and P contents on the carbon content of the eutectic
are given by the expression

carbon equivalent value (CEV) = T.C.% + Si% + P%

3
Cast irons with carbon equivalent values of less than 4.3% are called hypo-
eutectic irons, those having carbon equivalent values higher than 4.3% are
called hyper-eutectic irons. Since the structure (and hence the strength) of
flake irons is a function of composition, a knowledge of the CEV of an iron
can give an approximate indication of the strength to be expected in any
sound section. This is conveniently expressed in graphical form (Fig. 3.2).
From Fig. 3.2 it is possible to construct a series of curves showing the
reduction of strength with increasing section thickness, this is shown in Fig.
3.1 for the common grades of grey iron. It is important to realise that each
of these curves represents an average figure and a band of uncertainty
exists on either side. Nevertheless, these figures represent those likely to be
achieved on test bars cast in green sand moulds.

350
15 mm
20 mm 30 mm

300
Tensile strength (N/mm2)

50 mm
75 mm
250
100 mm

200 150 mm

150

100
5.0 4.5 4.0 3.5
CE value

Figure 3.2 Relationship between tensile strength and carbon equivalent value for
various bar diameters. (Data supplied by CDC.)

Note that the ‘carbon equivalent liquidus value’ (CEL) is frequently used
as a shop floor method of quality control since it can be directly and readily
measured. When unalloyed molten irons cool, a temperature arrest occurs
when solidification commences, the liquidus arrest. The temperature of this
arrest is related to the C, Si and P content of the iron by the expression
carbon equivalent liquidus (CEL) = T.C.% + Si%/4 + P%/2
The CEL value is a guide to the tensile strength and chilling characteristics
of the iron.
34 Foseco Ferrous Foundryman’s Handbook

Machinability

In general, the higher the hardness, the poorer the machinability and castings
with hardnesses above 250–260 are usually regarded as unsatisfactory. It is
necessary to avoid mottled, chilled or white irons which contain free carbide,
making them hard and unmachinable. Most grey iron castings are required
to be strong and readily machinable, this is achieved with a pearlitic structure
having no free ferrite or free carbide. Soft irons, freely machinable containing
appreciable amounts of free ferrite as well as pearlite, are suitable for certain
applications, particularly for heavy section castings (Fig. 3.3). (Figures 3.1,
3.2, 3.3 are reproduced by kind permission of CDC, Alvechurch, Birmingham.)
0 0

5 Mottled or white 2.5

(hard and unmachinable)
10 5.0
Bar diameter (mm)

Plate thickness (mm)

15 7.5

20 10.0
Pearlite
25 Ferrite and pearlite (strong and 12.5
(soft and machinable) machinable)
30 15.0

35 17.5

40 20.0
5.0 4.5 4.0 3.5 3.0
CE value

Note : The general microstructure of a wide plate is similar to that of a bar with a
diameter equal to twice the thickness of the plate. At a surface however the chilling
tendency of a plate casting will be greater and the structure can then be assumed
to be similar to that of a bar with a diameter equal to the thickness of the plate. The
left-hand scale is therefore used to assess surface chilling tendency and the right-
hand scale for assessing the general structure for a plate casting of a given thickness.

Figure 3.3 Relationship between section size, CEV, and structure. (Data supplied
by CDC.)

Classification of graphite flake size and shape

The properties of grey iron castings are influenced by the shape and
distribution of the graphite flakes. The standard method of defining graphite
forms is based on the system proposed by the American Society for the
Testing of Metals, ASTM Specification A247 which classifies the form,
distribution and size of the graphite. Certain requirements must be met
 Grey cast iron 35

before a sample is evaluated. Attention must be paid to the location of the

microspecimen in relation to the rest of the casting, to the wall thickness
and the distance from the as-cast surface. Care is also needed in grinding
and polishing so that as much graphite as possible is retained in a
representative cross-section. The specimen is normally examined in the
polished, unetched condition at a magnification of 100× and with a field of
view of about 80 mm.
Flake morphologies are divided into five classes (Fig. 3.4a).

Type A is a random distribution of flakes of uniform size, and is the

preferred type for engineering applications. This type of graphite structure
forms when a high degree of nucleation exists in the liquid iron, promoting
solidification close to the equilibrium graphite eutectic.
Type B graphite forms in a rosette pattern. The eutectic cell size is large
because of the low degree of nucleation. Fine flakes form at the centre of
the rosette because of undercooling, these coarsen as the structure grows.

Form: I II

III IV

V VI

Figure 3.4 (a) Reference diagrams for the graphite form (Distribution A). The
diagrams show only the outlines and not the structure of the graphite.
36 Foseco Ferrous Foundryman’s Handbook

Dimensions of the graphite particles forms I to VI

Reference Dimensions of the particles True dimensions

number observed at ×100
(mm) (mm)

1 >100 >1
2 50– 100 0.5–1
3 25– 50 0.25–0.5
4 12– 25 0.12–0.25
5 6– 12 0.06–0.12
6 3– 6 0.03–0.06
7 1.5– 3 0.015–0.03
8 <1.5 <0.015

Distribution: A

B C

D E

Figure 3.4 (b) Reference diagrams for the distribution of graphite (Form 1). The
diagrams show only the outlines and not the structure of the graphite.

Type C structures occur in hyper-eutectic irons, where the first graphite

to form is primary kish graphite. It may reduce tensile properties and
cause pitting on machined surfaces.
Type D and Type E are fine, undercooled graphites which form in rapidly
cooled irons having insufficient graphite nuclei. Although the fine flakes
 Grey cast iron 37

increase the strength of the eutectic, this morphology is undesirable because

it prevents the formation of a fully pearlitic matrix.

The ASTM Specification also provides standards for measuring flake size.
This is done by comparing a polished microspecimen of the iron at a standard
magnification of 100× with a series of standard diagrams (Fig. 3.4b).

Applications of grey iron castings

Grades of grey iron suitable for various types of casting

Unalloyed grey iron is used for a wide variety of castings. The following
table indicates typical grades of iron used for making certain types of casting.
Table 3.3 is only intended as a guide, casting buyers may specify different
grades from those in the table. Some buyers may have their own specifications
which do not coincide exactly with international or national specifications.

The production of grey irons

In order to produce castings having the required properties, the following
factors must be controlled:

the charge to be melted;

the method of melting;
treatment of the liquid metal before casting.

When deciding on a chemical composition for any casting, the effects of the
main constituents have to be considered. These can be summarised briefly
as follows:

Carbon. The effect of carbon must be considered together with that of

silicon and phosphorus and the concept of carbon equivalent value has
already been discussed. CEV has a major effect on the strength and hardness
of the casting (Figs 3.2, 3.3). CEV also affects the casting properties of the
iron. Irons having CEV below about 3.6% become difficult to cast sound,
because liquid shrinkage occurs and feeding is necessary to achieve complete
soundness.

Silicon. Next in importance to carbon, with regard to the properties of

iron, is silicon. Silicon is a graphitising element, high silicon irons, over
about 1.6%Si, tend to be graphitic, while low silicon irons are mottled or
white. Silicon contents of grey iron are generally around 2.0%.

Manganese. Manganese is necessary to neutralise the effect of sulphur in

38 Foseco Ferrous Foundryman’s Handbook

Table 3.3 Typical grades of grey iron and their use

Class of casting Grade of grey iron used

(tensile strength N/mm2 )

Air-cooled cylinders 200–250

Agricultural machinery 200–250
Bearing caps 200
Bearing housings 200
Boilers (central heating) 150–200
Brake cylinders 200
Brake discs 200
Brake drums 200
Clutch housings 200
Clutch plates 200
Cooking utensils 150–200
Counterweights 100–150
Cylinder blocks 200–250
Cylinder heads 200–250
Diesel 250
Electric motor stator cases 200
Drain covers, gulleys, etc. 150
Flywheels 200
Gear boxes 200–250
Ingot moulds 100–150
Lathe beds 100
Machine tool bases 200
Manifolds 200–250
Paper mill drier rolls 150–200
Piano frames 150–200
Rainwater pipes, gutters 150
Refrigerator compressors 250
Tractor axles 150–200
gear boxes 200
Valves, low pressure (gas) 150–200
water 200
hydraulic 250
Water pumps 200

iron. Without sufficient Mn, iron sulphide forms during solidification and
deposits around grain boundaries where it renders the metal hot-short and
likely to produce cracked castings. If sufficient Mn is present, MnS forms as
the molten metal cools and floats out of the metal into the slag layer.
The formula: Mn% = 1.7 × S% + 0.3% represents the amount of Mn
needed to neutralise the sulphur, although the sulphur should not be allowed
to exceed about 0.12%. In addition to combining with sulphur, Mn is a
pearlite stabiliser and it increases the hardness of the iron. However, it is
not primarily used for strengthening because it can affect nucleation adversely.
Mn levels in most irons are typically 0.5–0.8%.
 Grey cast iron 39

Sulphur. Sulphur is generally harmful in grey iron, and should be kept to

below 0.12%. Even if the sulphur is adequately neutralised with Mn, if the
high S slag is trapped in the casting, it can cause blowhole defects beneath
the casting skin.

Phosphorus. P has limited solubility in austenite and so segregates during

solidification forming phosphides in the last areas to solidify. While P increases
the fluidity of all cast irons, it makes the production of sound castings more
difficult and the shock resistance of the iron is reduced. For most engineering
castings, phosphorus should be kept below 0.12%, but up to 1.0% may be
allowed to improve the running of thin section castings where high strength
is not required.

Chromium. Added in small amounts, Cr suppresses the formation of free

ferrite and ensures a fully pearlitic structure, so increasing hardness and
tensile strength. Too much Cr causes chill at the edges of the casting, reducing
machinability seriously. Cr up to about 1% may be added to grey irons used
for special purposes, such as camshafts, where chills are often used to create
wear resistant white iron on the cam noses.

In addition to the above elements, the effect of certain impurities in the

iron must also be considered:

Copper. Cu increases tensile strength and hardness by promoting a pearlitic

structure and reducing free ferrite. It reduces the risk of chill in thin sections.
Up to about 0.5% Cu may arise from the presence of Cu as a tramp element
in steel scrap.

Tin. Sn has a similar effect to that of copper, though smaller amounts are
effective. Up to 0.1%Sn ensures a fully pearlitic matrix and reduces free
ferrite. Like copper, small amounts of tin may arise from steel scrap.

Lead. The presence of lead in grey iron, in amounts as low as 0.0004%, can
cause serious loss of strength through its harmful effect on the structure of
flake graphite. Lead contamination in cast iron comes usually through the
inclusion of free-cutting steel in the scrap steel melting charges, though it
can also arise from certain copper alloys such as gun-metal.

Aluminium. Contamination of automotive steel scrap by light alloy

components is the usual source of Al in iron. Levels of 0.1% Al may occur.
Al promotes hydrogen pick-up from sand moulds and may cause pinhole
defects in castings.

Nitrogen. Nitrogen above about 0.01% causes blowholes and fissures in

iron castings, heavy section castings are most seriously affected. Nitrogen
usually arises during cupola melting, particularly if high steel charges are
used, but the use of moulds and cores containing high nitrogen resins can
also cause problems. Addition of 0.02–0.03% titanium neutralises the effect
of nitrogen.
Chapter 4
Melting cast irons

Introduction
Iron foundries require metal of controlled composition and temperature,
supplied at a rate sufficient to match the varying demands of the moulding
line. The metallic charge to be melted consists usually of foundry returns,
iron scrap, steel scrap and pig iron with alloying additions such as ferrosilicon.
The charge is usually melted in a cupola or in an electric induction furnace.
Gas-fired or oil-fired rotary furnaces can also be used, but their use is less
common.

Cupola melting
The cupola (Fig. 4.1) is the classical iron melting unit and is still the most
widely used primary melting unit for iron production due to its simplicity,
reliability and the flexibility in the quality of charge materials that can be
used because some refining of undesirable elements such as zinc and lead
can be achieved. While the cupola is an efficient primary melting unit, it
does not adapt easily to varying demands, nor is it an efficient furnace for
superheating iron. For this reason it is often used in conjunction with an
electric duplexing furnace.
The simplest form is the cold blast cupola which uses ambient temperature
air to burn the coke fuel. The metal temperature that can be achieved is
normally from 1350 to 1450°C but higher temperatures can be achieved
through the use of divided blast (as in Fig. 4.1) or oxygen enrichment. The
refractory linings of cold blast cupolas have a short life of less than 24
hours, so cupolas are operated in pairs, each used alternately while the
other is re-lined.
In hot blast cupolas (Fig. 4.2), the exhaust gases are used to preheat the
blast to 400–600°C, reducing coke consumption and increasing the iron
temperature to more than 1500°C. They may be liningless or use long life
refractories giving an operating campaign life of several weeks.
‘Cokeless’ cupolas (Fig. 4.3), have been developed in which the fuel is
gas or oil with the charge supported on a bed of semi-permanent refractory
spheres. They have advantages of reduced fume emission.
 Melting cast irons 41

Charging
door

Metal and
coke charge

Melting
Blast air zone

Blast air Tuyéres

Slag

Iron
Sand bed

Figure 4.1 Section through a cupola (From ETSU Good Practice Case Study 161;
courtesy of the Department of the Environment, Transport and the Regions.)

Cold blast cupola operation

The cupola is charged with:

1. coke, the fuel to melt the iron;

2. limestone, to flux the ash in the coke etc.;
3. metallics, foundry scrap, pig iron, steel and ferroalloys;
4. other additions to improve the operation
42 Foseco Ferrous Foundryman’s Handbook

CUPOLA HEAT EXCHANGER DUST COLLECTOR CHIMNEY

Figure 4.2 Hot blast cupola. (From ETSU Good Practice Case Study 366; courtesy
of the Department of the Environment, Transport and the Regions.)

The cupola is blown with air to combust the coke and the air flow controls
the melting rate and metal temperature. The output of a cupola depends
primarily on the diameter of the shaft of the furnace and on the metal/coke
ratio used in the charge. Table 4.1 summarises the operating data for typical
cold blast cupolas.
A useful measure of the efficiency of operation of a cupola is the ‘Specific
Coke Consumption’ (SSC) which is
Annual tonnage of coke × 1000
= SSC (kg/tonne)
Annual tonnage of metallics charged
This takes into account both charge coke and bed coke. When the cupola is
operated for long enough campaigns, the amount of coke used to form the
bed initially can be ignored. However, as the melting period decreases, the
role of the cupola bed becomes more important. Table 4.2 summarises data
from 36 cupola installations in the UK in 1989. This table provides a useful
reference against which the operation of any cold blast cupola can be
compared.

Coke

The performance of the cupola is highly dependent on the quality of the

coke used. Typical foundry coke has the following properties:

Moisture 5% max.
Ash 10% max.
 Melting cast irons 43

Charge opening

Air pipe

Blast inlet
Charge

Shell cooling

Siphon with slag separator Ceramic

bedding
Carburization
Water-cooled grate
Temperature measuring instrument Burner
Superheater

Deslagging
opening

Tap hole Feeder

Inductor Tilting cylinder

Figure 4.3 Schematic diagram of a cokeless cupola in a duplex system.

(From R.F. Taft, The Foundryman, 86, July 1993 p. 241.)

Volatiles 1% max.
Sulphur 1% max.
Mean size 100 mm
Undersize <5% below 50 mm

The coke size directly affects coke consumption per tonne of iron melted
and also the melting rate. Optimum cupola performance is achieved with
coke in the size range 75–150 mm, if smaller coke is used, metal temperature
is reduced and a higher blast pressure is needed to deliver the required
amount of air to the cupola. Increasing the size of coke above about 100 mm
has no beneficial effect, probably because large pieces of coke tend to be
fissured and break easily during charging and inside the cupola.
Coke usage in the cold blast cupola is typically 140 kg per tonne of iron
melted (this is an overall figure including bed coke), it is usual to charge
coke at the rate of about 10–12% of the metal charged, but the exact amount
used depends on many factors such as tapping temperature required, melting
rate and the design of the cupola, see Table 4.2.
Table 4.1 Cupola operation data
Metric units

Diameter Melting rate (tonnes/h) Blast Typical charge (kg) Bed height Shaft height (m)
of melting metal : coke ratio rate pressure at 10:1 coke rate above tuyeres tuyeres to
zone (cm) 10 : 1 8:1 m3/h cm H2O kPa coke iron limestone (cm) charge door sill

50 1.97 1.57 1340 104 10.2 20 200 7 100 2.5

60 2.84 2.46 1940 107 10.5 28 284 9 100 3.0
80 5.11 4.36 3450 114 11.2 51 510 17 105 3.0
100 7.99 6.83 5380 119 11.7 80 800 26 105 3.5
120 11.50 9.79 7750 130 12.7 115 1150 38 110 4.0
140 15.60 13.33 10 600 137 13.4 157 1570 52 110 4.0
160 20.44 17.41 13 800 147 14.4 200 2040 67 110 4.5
180 25.88 22.05 17 450 157 15.4 260 2590 85 115 5.0
200 31.95 27.22 21 550 175 17.2 320 3200 106 115 5.0
Table 4.1 (Continued)
Imperial units

Diameter Melting rate (ton/h) Blast Typical charge (lbs) Bed height Shaft height (feet)
of melting metal :coke ratio rate pressure at 10:1 coke rate above tuyeres tuyeres to
zone (inches) 10:0 8:1 cfm in. w.g. coke iron limestone (inches) charge door sill

18 1.6 1.3 665 40 36 360 12 38 16

24 2.9 2.5 1180 42 65 650 21 39 16
30 4.5 3.9 1840 44 100 1000 33 41 16
36 6.6 5.6 2650 46 150 1500 50 41 19
42 9.0 7.6 3620 48 200 2000 66 42 19
48 11.7 10.0 4720 51 260 2600 86 42 22
54 14.8 12.6 5950 53 330 3300 109 43 22
60 18.3 15.6 7360 57 410 4100 135 43 22
66 22.1 18.8 8900 60 500 5000 165 44 22
72 26.5 22.5 10 650 63 594 5940 196 44 22
78 31.1 26.4 12 500 69 697 6970 230 44 22
84 36.1 30.6 14 500 75 809 8090 267 45 22

The above figures represent good average practice and are intended to act as a rough guide only.
Table 4.2 Data for cupolas in the UK (1989)

Melt rate Cupola Water Tapping Melt Bed coke Restore SCC Type of % Coke Divided
(t/hr) dia. (inches) cooled? temp. (°C) period (hrs) (kg) coke (kg) (kg/tonne) metal charge blast

1.5 22.0 no 1400 1.5 100 407 grey 12.00

2.0 36.0 1500 3.5 500 252 grey 14.00
3.0 30.0 no 1300 3.0 500 260 grey 4.00
3.0 32.0 1340 2.0 600 150 216 grey 8.9
3.0 32.0 no 1450 7.0 500 60 133 malleable 11.00
3.0 36.0 yes 1350 3.0 400 150 313 grey 16.00
3.0 22.0 2.0 100 267 grey 20.00
3.0 30.0 yes 1450 9.0 210 100 grey
3.0 26.0 1550 2.0 242 grey
3.0 30.0 no 1475 7.5 490 0 214 malleable 16.00
3.0 30.0 no 1430 2.0 400 150 grey 7.20
3.0 30.0 no 1480 7.0 700 140 208 grey 15.00
3.5 33.0 yes 1300 3.0 380 212 grey 18.00
4.0 33.0 no 1450 6.0 1750 grey 9.00
4.0 35.0 no 1450 3.0 600 217 grey 12.00 x
4.0 36.0 no 1470 3.5 650 150 105 grey 7.00
4.0 30.0 no 1500 5.0 840 500 211 grey 12.00
4.0 34.0 no 1460 3.0 375 127 grey 12.4

(Contd)
Table 4.2 (Continued)

Melt rate Cupola Water Tapping Melt Bed coke Restore SCC Type of % Coke Divided
(t/hr) dia. (inches) cooled? temp. (°C) period (hrs) (kg) coke (kg) (kg/tonne) metal charge blast

4.5 31.0 1550 10.0 1000 86 164 grey 13.50

5.0 32.0 no 1530 8.0 600 129 grey 12.5
5.0 33.0 no 1550 4.0 750 243 grey 16.0 x
5.0 36.0 no 1470 8.0 1750 250 178 grey 17.80 x
5.0 36.0 no 1500 2.0 420 144 grey 100.00
5.0 39.0 no 1460 8.0 450 150 171 grey 11.00 x
8.0 42.0 no 1500 4.0 1500 153 grey 11.00
8.0 53.9
8.0 48.0 yes 1550 9.0 900 363 140
8.0 38.0 no 1490 7.0 1100 115 grey 8.64 x
8.0 48.0 no 1440 8.0 1500 138 grey wide range
9.0 42.0 yes 1500 33.0 2000 180 121 grey 10.60 x
9.5 48.0 yes 1460 8.0 1000 500 129 grey 10.0 x
10.0 52.0 no 1450 8.0 1800 400 154 grey 13.10
10.0 48.0 yes 1500 15.0 1800 300 138 grey 13.70 x
12.0 48.0 1520 8.5 1800 300 112 grey 9.50 x
12.0 43.0 yes 1530 20.0 1200 200 104 malleable 9.50 x
20.0 72.0 yes 1550 336.0 4000 169 duct 16.50

From: Coke consumption in iron foundry cupolas, Energy of Consumption Guide 7, November 1990, reproduced by permission of the Energy Efficiency Office
of the Department of the Environment.
48 Foseco Ferrous Foundryman’s Handbook

Fluxes

Fluxes are added to the cupola charge to form a fluid slag which may easily
be tapped from the cupola. The slag is made up of coke ash, eroded refractory,
sand adhering to scrap metal and products of oxidation of the metallic charge.
Limestone is normally added to the cupola charge, it calcines to CaO in the
cupola and reacts with the other constituents to form a fluid slag. Dolomite,
calcium–magnesium carbonate, may also be used instead of limestone.
The limestone (or dolomite) should contain a minimum of 96% of CaCO3
(and MgCO3) and should be in the size range 25–75 mm.
The amount of the addition is dependent on the coke quality, the cleanliness
of the charge and the extent of the lining erosion. Normally 3–4% of the
metallic charge weight is used. Too low an addition gives rise to a viscous
slag which is difficult to tap from the furnace. Too high an addition will
cause excessive attack on the refractory lining. When the coke bed is charged,
it is necessary to add around four times the usual charge addition of limestone
to flux the ash from the bed coke.
Other fluxes may also be added such as fluorspar, sodium carbonate or
calcium carbide. Pre-weighed fluxing briquettes, such as BRIX, may also, be
used. BRIX comprises a balanced mixture of fluxing agents which activates
the slag, reduces its viscosity and produces hotter, cleaner reactions in the
cupola. This raises carbon content, reduces sulphur and raises metal
temperature.
Correct additions of flux are essential for the consistent operation of the
cupola and care should be taken to weigh the additions accurately.

The metallic charge

Table 4.3 gives the approximate metal compositions needed for the most
frequently used grades of grey iron. (Data supplied by CDC.)

Table 4.3 Metal composition needed to produce the required grade of grey iron

Grade 150 200 250 300 350

Total carbon (%) 3.1–3.4 3.2–3.4 3.0–3.2 2.9–3.1 3.1 max

Silicon (%) 2.5–2.8 2.0–2.5 1.6–1.9 1.8–2.0 1.4–1.6
Manganese (%) 0.5–0.7 0.6–0.8 0.5–0.7 0.5–0.7 0.6–0.75
Sulphur (%) 0.15 0.15 0.15 max 0.12 max 0.12 max
Phosphorus (%) 0.9–1.2 0.1–0.5 0.3 max 0.01 max 0.10 max
Molybdenum (%) 0.4–0.6 0.3–0.5
Cu or Ni (%) 1.0–1.5
Note: Copper may partially replace nickel as an alloying addition

Metallic charge materials

The usual metallic charge materials are:

Return scrap: runners, risers, scrap castings etc. arising from the foundry
 Melting cast irons 49

operation. Care must be taken to segregate each grade of returns if the

foundry makes more than one grade of iron.
Pig iron: being expensive, the minimum amount of pig iron should be
used. Use of pig iron is a convenient way of increasing carbon and silicon
content. Special grades of pig iron having very low levels of residual
elements are available and they are particularly useful for the production
of ductile iron.
Steel scrap: is normally the lowest cost charge metal, it is used for lowering
the total carbon and silicon contents.
Bought scrap iron: care must be taken to ensure that scrap of the correct
quality is used, particularly for the production of the higher strength
grades of iron.

Harmful materials
Care must be taken to ensure that contaminants are not introduced into the
iron. The most common harmful elements are:

Lead, usually from leaded free-cutting steel scrap.

Chromium, from stainless steel.
Aluminium, from aluminium parts in automotive scrap.

Size of metallic charge materials

Thin section steel scrap (below about 5 mm) oxidises rapidly and increases
melting losses. On the other hand, very thick section steel, over 75 mm, may
not be completely melted in the cupola. Metal pieces should be no longer
than one-third of the diameter of the cupola, to avoid ‘scaffolding’ of the
charges.

Ferroalloys
Silicon, manganese, chromium, phosphorus and molybdenum may all be
added in the form of ferroalloys. In some countries, Foseco supplies briquetted
products called CUPOLLOY designed to deliver a specific weight of the
element they introduce, so that weighing is unnecessary.
Ferrosilicon in lump form, containing either 75–80% or 45–50% Si may be
used. Ferromanganese in lump form contains 75–80% Mn. Both must be
accurately weighed before adding to the charge.

Pig irons
Typical pig iron compositions are given in Table 4.4. Refined irons for foundry
50 Foseco Ferrous Foundryman’s Handbook

use are normally made in a hot blast cupola from selected scrap, they may
contain copper, tin, chromium and other alloy elements. Base irons for ductile
(s.g., nodular) iron production are made from specially pure ores, and have
very low residual element contents. They are available in a range of
specifications.

Table 4.4 Foundry pig iron

Grade Typical composition

TC(%) Si(%) Mn(%) S(%) P(%)

Blast furnace irons 3.4–4.5 0.5–4.0 0.7–1.0 0.05 max 0.05 max
Refined irons 3.4–3.6 0.75–3.5 0.3–1.2 0.05 max 0.1 max
Ductile base irons 3.8 0.05–3.0 0.01–0.20 0.02 max 0.04 max

Purchased cast iron scrap is available in a number of grades, typical

compositions are shown in Table 4.5.

Table 4.5 Cast iron scrap

Type Typical composition

TC(%) Si(%) Mn(%) S(%) P(%)

Ingot mould scrap 3.5–3.8 1.4–1.8 0.5–1.0 0.08 0.1
Heavy cast iron scrap 3.1–3.5 2.2–2.8 0.5–0.8 0.15 0.5–1.2
Medium cast iron scrap 3.1–3.5 2.2–2.8 0.5–0.8 0.15 0.5–1.2
Automobile scrap 3.0–3.4 1.8–2.5 0.5–0.8 0.15 0.3 max

Typical charges needed to produce the most frequently used grades of iron
are given in Table 4.6.

Table 4.6 Typical furnace charges

Grade 150 Grade 200 Grade 250

25% pig iron 30% low P pig iron 25% low P pig iron
40% foundry returns 35% foundry return 35% foundry returns
30% bought cast iron 20% low P cast iron scrap 15% low P scrap
5% steel scrap 15% steel scrap 25% steel scrap

Cupola charge calculation

In a normally operated, acid cold blast cupola, the composition of the metal
tapped can be predicted with reasonable accuracy from the composition of
the furnace charge. The tendency is for the total carbon to attain the eutectic
equivalent. If the quantity charged is above this value, a loss may be expected.
On the other hand, where the charge contains less than the eutectic value,
the trend is towards a carbon pick-up. The exact amount of carbon change
must be established by experience for a particular cupola operation, but the
following ‘Levi equation’ is a good starting guide.
 Melting cast irons 51

TC% in the charge Si% + P% at spout

TC% at spout = 2.4 + –
2 4
Silicon is always lost in the cupola, generally a loss of 15% of that charged
may be assumed, but higher losses may occur if high steel charges are used.
Manganese losses are usually about 25%. Phosphorus changes little. Sulphur
always increases due to pick-up from the coke, but the precise amount
cannot be predicted and must be based on experience.
Based on these guide lines, a calculation may be made as follows:
To make a Grade 250 iron with the composition:

TC Si Mn P
3.2 1.7 0.7 0.1

Material Amount Composition Contribution to charge (%)

charged TC Si Mn P TC Si Mn P
(%)

Low-P pig iron 25 3.0 3.0 1.0 0.1 ×0.25 0.75 0.75 0.25 0.03
Grade 250 returns 35 3.2 1.7 0.7 0.1 ×0.35 1.12 0.60 0.25 0.04
Low-P scrap iron 15 3.2 2.2 0.8 0.15 ×0.15 0.48 0.33 0.12 0.02
Steel scrap 25 0.1 0.1 0.3 0.03 ×0.25 0.03 0.03 0.08 –
Ferromanganese 0.3 75 ×0.003 0.23
Total 2.38 1.71 0.93 0.09
Changes during melting Si loss 15% –0.26
Mn loss 25% –0.23
Addition at spout 70% ferrosilicon +0.25
Expected composition T.C. = 2.4 + 2.38/2 – (1.45 + 0.09)/4
= 3.2
Si = 1.70
Mn = 0.70
P= 0.09

Calculations such as the above example, should only be used as a guide.

The precise carbon pick-up and silicon losses achieved depend on factors
such as coke quality, metal temperature, melting rate etc. Experience will
enable more accurate predictions to be made. A number of computer programs
are available which carry out the calculations rapidly and enable ‘least cost’
charges to be selected.

Cupola output
The maximum output from a cupola is determined primarily by the shaft
diameter, Table 4.1. In normal use, it is necessary to be able to vary the output
to match the requirements of the moulding line. This is done by varying the
blast rate. Increasing or reducing the air supplied to the cupola burns more
or less coke and increases or reduces the melting rate. Unfortunately, changing
the blast rate also changes the temperature of the metal, and to some extent,
52 Foseco Ferrous Foundryman’s Handbook

its composition. This is one of the main drawbacks of the cupola as a melting
furnace. Another problem is that the only way of changing the composition
of the liquid iron is to change the make-up of the charge, and it usually takes
around an hour before the change is seen at the tap-hole. To overcome these
difficuties, it is common practice to tap the cupola into an electric holding
furnace where the temperature and composition can be accurately controlled,
and variations in metal demand can be accommodated.

Emissions from cupolas

Exhaust gases from cupolas are hot and contain dust, grit and SO2 gas. For
many years, the emissions permitted from cold blast cupolas were readily
achieved by the use of simple wet arresters. The cupola gases pass through
a curtain of water which removes the grit particles, absorbs up to half of the
SO2 but does not remove dust. Present day environmental regulations in
most countries impose increasingly strict limitations on the dust emissions
permitted from cupolas, requiring additional dust-arresting plant to be fitted.
Wet scrubbers, bag filters and electrostatic precipitators can be used.
There are two types of wet scrubbers: venturi scrubbers and disintegrators.
Venturi scrubbers rely on the pressure drop across a restricted throat and
disintegrators on the wetting and agglomeration of dust particles by the
action of water carried by a rapidly spinning rotor. Capital cost and running
costs, power and maintenance, are high.
Dry bag filters are capable of achieving lower emission levels than wet
scrubbers. The gases must be cooled before filtration making capital costs
higher than wet scrubbers but running costs may be lower.
Electrostatic precipitators are efficient but are expensive and require
specialised maintenance, they are uncommon on foundry cupolas.

The long campaign hot blast cupola

Hot blast cupolas were, until recently, only considered economical for
foundries with large continuous requirements for molten iron. Hot blast
cupolas are operated on long campaigns, many with unlined, water-cooled
steel shells. Independently fired blast systems have been used but they
have high fuel costs and have now been largely abandoned. There is now
renewed interest in the long campaign hot blast cupola, with recuperative
systems using the heat from combusting the cupola offtake gases to heat the
blast (Fig. 4.2). In part, the change has come about because of environmental
concerns. Cold blast cupolas, in the UK and elsewhere, have in the past
been allowed to operate with simple, low cost emission control. Environmental
controls are now becoming more stringent, requiring high efficiency filtration
of cupola offtake gases. This generally demands combustion then cooling of
the gases prior to filtration. Rather than waste this heat, more foundries are
turning to the hot blast cupola.
 Melting cast irons 53

The upper section of the cupola is lined with refractory, while the melting
zone may be liningless (having a water-cooled shell) or it may use a high
quality backing lining with a replaceable inner lining. The liningless cupola
can be operated for several weeks without dropping the bottom. The refractory
lined cupola is usually operated for a week without replacing refractories.
Combustion of the offtake gases is maintained by introducing air into the
shaft below the charging door together with a gas-fired afterburner which
automatically ignites if the temperature of the gases falls too low. The offtake
gases are drawn from the cupola through a recuperator which preheats the
incoming blast air to around 500°C. The blast air is enriched with 1.5–2.0%
oxygen. The waste gases are cooled to 175°C before passing through a dry
bag filter prior to discharge to atmosphere.
Tapping temperatures of 1530°C are achieved. The main savings over
conventional cold blast cupola practice is found in the reduced coke
consumption. Savings of up to 30% of coke usage are claimed. Long campaign
cupolas can be designed for economical operation from 10 tonnes/hr upwards.
The long campaign hot blast cupola is considered by many to be the most
economical method of melting grey iron for foundries.

The cokeless cupola

This is a continuously melting tower furnace in which the metallic charge is
supported on a water-cooled grate on which is a bed of carbonaceous
refractory spheres. Heat for melting is provided by gas (or oil) burners (Fig.
4.3). Superheating of the liquid iron is performed by the heated refractory
spheres and carbon can be added by injecting a suitable recarburiser into
the well of the cupola. Eliminating the coke eliminates sulphur pick-up,
making the cokeless cupola suitable for the production of base iron for
ductile iron production. It also eliminates the main source of atmospheric
pollution. The cokeless cupola retains the advantages of cupola melting:
continuous operation, ability to accept a wide range of raw materials including
wet, oily and contaminated scrap and some refining which removes harmful
elements such as lead and zinc.
The cokeless cupola is particularly attractive in countries where good
quality foundry coke is not available. The most efficient way of using the
cokeless cupola is to tap at around 1350–1400°C into an electric duplexing
furnace where temperature and composition are controlled. This practice
reduces gas consumption to about 55 m3/tonne of metal melted and greatly
reduces the consumption of the refractory spheres of the bed to around
1 kg/tonne of metal melted. Cokeless cupolas with capacity from 5–15
tonnes/h are in use.

Electric melting
Electric melting in the form of arc, induction and resistance furnaces is used
54 Foseco Ferrous Foundryman’s Handbook

increasingly, both for primary melting and holding of liquid iron. Induction
furnaces are the most popular, there are two basic types, the channel furnace
and the coreless induction furnace. Compared with cupola melting, induction
furnaces offer the following advantages:

The ability to produce iron of closely controlled composition.

The ability to control tapping temperature precisely.
A wide range of charge materials can be used.
Environmental pollution is much reduced.
The stirring effect of induction power rapidly incorporates additions.

The channel furnace

The channel furnace consists of an upper vessel, holding the bulk of the
charge material, with an inductor bolted on the underside (Fig. 4.4).

Cover hot-face lining may be

Note: High-alumina hot-face different grade of refractory
lining may be brick, rammed from body
or castable refractory

Joint High-alumina
hot-face lining Slag port

Rammed or cast
inductor-lining
(high alumina or MgO)

Furnace shell

Alumina-silicate fibre Joint

moist felt next to shell
Insulating blocks or
bricks Throat may be lined in
different grade of
refractory from body

Figure 4.4 A bath channel induction fumace, showing typical lining arrangement.
(Reproduced by courtesy of CDC.)

All the power is induced within the inductor, heating a loop of molten
metal which transfers its heat to the main body of the charge by convection
and induction forces. The channel furnace will only work providing this
loop is maintained 24 hours per day. The temperature of the metal in the
 Melting cast irons 55

loop is higher than that of the metal in the main vessel, which limits the
operating temperature of a channel furnace, since a high loop temperature
shortens the life of the refractory of the loop. This means that their use for
melting is restricted to low temperature metals, such as brass or aluminium.
Channel furnaces are frequently used to duplex iron from a cupola, that is,
the cupola is used as the primary melting unit and continuously delivers
liquid iron to a large (say, 20 tonnes) channel furnace where compositional
variations are corrected and the temperature is maintained at the required
value. This means that the varying demand from the foundry can be met
with metal of precisely the required specification, which cannot be achieved
with a cupola alone.
The disadvantage of the channel furnace is the necessity to operate it for
24 hours per day, and the problems which arise if the refractories of the
induction loop fail.

The coreless induction furnace

In a coreless furnace, the coil surrounds the entire charge (Fig. 4.5). The
mass of refractory is much less than in the channel furnace while the shape
is a simple hollow cylinder. Hence coreless furnaces are much simpler and

Primary or furnace coil

Charge or secondary coil

Lining

Furnace frame

Figure 4.5 Section through a coreless induction furnace. (From Jackson, W.W. et
al. (1979) Steelmaking for Steelfounders, SCRATA; reproduced by courtesy of CDC.)
56 Foseco Ferrous Foundryman’s Handbook

less costly to reline, although they require more frequent relines than the
channel furnace.
The coreless furnace can be designed to operate at any frequency from
50 Hz upwards. Induction heating of liquid metal causes a stirring effect in
the metal. The lower the frequency of the primary current, the more intense
is the stirring. Therefore in a mains frequency furnace operating at 50 or
60 Hz, the turbulence is greater than in one operating at a higher frequency.
Because of the high turbulence, the power input to a mains frequency furnace
is restricted to around 250 kW per tonne of capacity. With higher frequencies,
the power density can be increased to three or four times this level.
The frequency of operation also affects current penetration. The current
induced in the metal charge is a maximum at the surface, reducing at greater
depths. The higher the frequency, the less the effective penetration depth.
This means that at 50 Hz, the smallest ferrous charge piece that can be
efficiently heated is around 450 mm diameter. The smallest practicable furnace
that can hold such a piece is 750 kg in capacity. Mains frequency furnaces
are not effective in sizes smaller than this. On the other hand, at 10 kHz
charge pieces less than 10 mm diameter can be heated, so furnaces as small
as 5 kg capacity can be used.
For most foundry applications, furnaces operating in the range 250–
3000 Hz are used, such furnaces:

allow high power densities giving high melting rates without excessive
stirring;
can be emptied completely after each melt and restarted with virtually
any size of scrap;
have short start-up time and offer flexibility for alloy changes;
allow high production rates from small furnaces;

The development of the solid state inverter in the early 1970s allowed the
possibility of a cheap and reliable form of frequency converter. Silicon
controlled rectifiers (SRCs) convert AC to DC and back to AC at the desired
frequency. Once these had been fully developed, the medium frequency
induction furnace replaced the earlier mains frequency and triple frequency
furnaces. So whereas in the past, choice of frequency was invariably a
compromise because of the limited range possible, now the correct unit for
the application can be supplied. One supplier (Inductotherm) for example,
has supplied:

70 Hz for brass swarf

100 Hz for aluminium scalpings
100–150 Hz for cast iron borings
250 Hz for Al extrusion scrap
250 Hz for iron from foundry returns and steel scrap
500 and 1000 Hz for steel melting and for melting wet cast iron borings
1000 and 3000 Hz for melting a wide variety of copper alloys
3000 Hz for investment casting
10000 Hz for the jewellery trade
 Melting cast irons 57

For iron melting furnaces having a melt capacity of around 2–20 tonnes,
frequencies of 200–1000 Hz are usually used, being powered at up to
750 kW per tonne of capacity. Such furnaces normally have a charge-to-tap
time of about 1–2 hours and produce metal at the rate of 2–10 tonnes per
hour.

Energy consumption

A figure of about 500 kWh/tonne is attainable when using a high-powered

medium frequency furnace in a melting role, i.e. when holding periods are
minimised. The energy used can be broken down approximately for a furnace
having the following characteristics (from C.F. Wilford, The Foundryman,
1981 p. 153):

Furnace capacity 4 tonnes

Installed power 3000 kW
Frequency 500 Hz
Standing heat losses (lid on)
averaged over complete melt cycle 47 kW
Power supply efficiency (to furnace coil) 96%
Coil efficiency 80%
Bath diameter 930 mm

Calculated energy consumption

Theoretical energy to melt 4 tonnes of charge to
1500°C 1480 kWh
Energy consumed (allowing for efficiency factors) 1927
Furnace heat losses over one cycle (lid on) 40
Charging time (lid off) 30 min
Extra energy loss incurred with lid off 10
Time for deslagging, temperature measurement
and analysis (lid off) 5 min
Extra energy incurred with lid off 16
Energy consumed to replace lid off losses 34
Total energy consumed (for 4 tonnes) 2001
kWh/t 500

The above figures are theoretical, surveys of actual foundry installations

show that figures of 520–800 kWh/tonne are common, the variation being
due to individual melting practice such as the rate at which the pouring line
will accept molten metal and whether furnace lids are used effectively.
Attention to energy saving measures should allow figures of 550–650 kWh
to be achieved.
58 Foseco Ferrous Foundryman’s Handbook

Charge materials

The maximum dimension of a piece of charge is around one-third of the

crucible diameter. If larger, there is a danger of bridging. The value may be
exceeded when a long piece of charge is fed in a controlled manner into the
crucible. Charge bridging can also be a problem when melting charges of
cast iron borings. It is difficult to generate circulating electric currents in a
cold charge having high electrical resistance between individual pieces (as
found with oxidised borings), so a large sintered mass may form which
does not easily sink down as melting occurs.
An advantage of the medium frequency furnace is that wet charge
components can safely be charged into an empty crucible, eliminating the
necessity of using a charge pre-drying stage. However, care must be taken
not to charge wet material into a fully molten bath.

Alloy recovery

Addition Recovery (%)

Carbon (graphite, petroleum coke) 80–88

Silicon (ferrosilicon 75%) 90
Manganese (ferromanganese 75%) 90–95
Chromium (ferrochromium 60–75%) 90
Molybdenum (ferromolybdenum 60–70%) 90
Nickel 95
Copper 95
Sulphur (iron sulphide) 100

Slag removal

Slag is formed in electric melting furnaces from the products of oxidation of

the elements in the charge, particularly the iron, silicon and manganese;
from refractory erosion and from dirt, sand or rust on the charge. Slag floats
on the surface of the metal and must be removed before tapping. Since slag
removal is an unpleasant task for the furnace operator, it is advisable to
avoid the use of dirty or rusty charge materials. Many slags are liquid and
of low viscosity and difficult to collect, they can be coagulated by adding
SLAX. At molten metal temperatures the SLAX granules expand and form
a low density, high volume crust which mops up the slag, which can then be
lifted off the metal with ease, leaving the surface clean. SLAX is based on
siliceous minerals and reacts with the slag to increase the silica content, and
hence the viscosity. Between 0.07 and 0.2% by weight (0.7–2.0 kg/tonne of
metal) of SLAX 10 or SLAX 30 is scattered over the slag on top of the
furnace. This is rabbled to form a dry, expanded crust and skimmed off.
 Melting cast irons 59

Refractories for coreless induction furnaces

Coreless furnaces in iron foundries are usually lined with a silica refractory
bonded with boric acid or, preferably boric oxide. The quality of the silica is
important, it is mined as quartzite and should have the following approximate
composition:

SiO2 Al2O3 Fe2O3 CaO MgO Alkali

98.9% 0.6% 0.2% 0.1% 0.04% 0.2% max

Correct particle sizing is essential so that the lining can be compacted to as

high a density as possible. Typical gradings are:

20% > 1 mm
20% 0.5–1.0 mm
30% 0.1–0.5 mm
30% < 0.1 mm

Boric oxide is usually used as the bonding agent, being mixed by the refractory
supplier. Around 0.7–0.8% B2O3 is used. During the fritting cycle, as the
temperature at the hot face increases, the boric oxide dissolves the silica
fines, producing a borosilicate glass which fills the interstices between the
silica grains and cements them together.
The usual practice with medium frequency furnaces is to coat the copper
coil of the furnace with a layer of ‘mudding’ about 6 mm thick, of a medium
to high alumina cement. This remains in place when the hot face is knocked
out. Between this and the hot face is a layer of ceramic fibre insulation. The
working face is formed by compacting the silica refractory behind a steel
former concentrically placed within the coil. Formers are normally constructed
from mild steel sheet according to the furnace manufacturer’s design.
Refractory is poured between the former and the coil and compacted using
vibratory ramming tools or manual compaction. Inhalation of silica dust
presents a hazard and respirators should be used during installation or
wrecking of the lining.
The lining is fritted by slow inductive heating of a metallic charge placed
inside the steel former. The heating rate depends on the size of the furnace
and the manufacturer’s recommendations should be followed. In general, a
low heating rate of 50–100°C/h should be used until the temperature reaches
700°C, after which the rate can be increased to 100–200°C/h (the faster rate
being possible with furnaces of size 10 tonnes or less). The temperature
should be raised to 30–50°C above the normal operating temperature of the
furnace, and held for about one hour to complete the fritting operation.
The lining life is very dependent on the particular practice used in the
foundry and the type of iron being melted. For example, the high carbon
and low silicon contents of most ductile base irons, together with the higher
temperatures involved, tends to result in lower lining lives than furnaces
60 Foseco Ferrous Foundryman’s Handbook

used for grey or malleable irons. A 5-tonne furnace melting grey iron should
be capable of melting 600–800 tonnes of iron from one lining, but less than
half this amount if a ductile base iron is being melted.
If a range of irons is melted, requiring higher temperature than normal,
or producing more aggressive slags, it may be necessary to use an alumina
lining. These can be used up to 1750°C and have greater corrosion resistance,
but the cost is much higher. The installed lining cost is around 2.5 times
greater than a silica lining, although the life may be three times longer.

Operating systems

Most iron foundries use two furnace bodies, identical in size, fed from a
single power supply with some means of switching the power supply between
the two furnaces. This allows a continuous supply of molten metal with one
furnace dispensing molten metal while the other is melting the next batch.
Switching techniques have been developed to enable a single power supply
to provide melting power to one furnace while simultaneously providing
holding power to the second so that temperature control can be maintained.
Mechanised charging systems, vibratory conveyors or drop bottom
charging buckets are frequently used to ensure maximum furnace utilisation.
De-slagging is the most arduous and time consuming operation, back tilting
the furnace aids the process.
While most of the ferrosilicon and carburiser are added during furnace
charging, some carbon and silicon losses will occur at high molten metal
temperatures. Trimming additions of 0.2 to 0.3%C and 0.2%Si are typical
during the final stages of the melting process, the stirring action of the
medium frequency power allowing rapid solution and consistent metal
composition.

Fume extraction

Electric melting plant produces less fume than cupola melting, but the ever
increasing stringency of environmental regulations requires fume extraction
plant to be fitted. The charging and pouring operations generate the majority
of the dust and fume emissions within the melt cycle. Close capture fume
hoods or high velocity lip extraction units are specified for fume capture
and dry bag filter systems are sufficient to handle the levels of fume that
arise.

Shop floor control of metal composition

The carbon and silicon content of unalloyed cast irons can be quickly
determined on the shop floor by thermal analysis. A sample of molten iron
 Melting cast irons 61

is poured into a small, expendable test mould about 25 mm diameter and 65

mm deep made from resin bonded sand and coated with tellurium to ensure
that the sample freezes white. The test mould also contains a thermocouple
connected to a temperature recorder. As the sample solidifies, the temperature
recorder plots a cooling curve which displays the ‘liquidus arrest’ when the
sample first starts to freeze, then the ‘eutectic arrest’ when freezing is complete.
The liquidus arrest measures the carbon equivalent liquidus value (CEL)
given by:

CEL = %C + %Si + %P
4 2
Note that CEL is not the same as CEV, %C + (%Si + %P)/3.
The eutectic arrest temperature, on an unalloyed iron of low P%, is a
measure of the silicon content of the iron. Thus from the two arrest
temperatures, the carbon and the silicon content can be calculated. Simple
calculators are available to enable the C% and Si% to be read, or digital
meters are available which display the C% and Si% directly. The carbon
content can be determined with an accuracy of ±0.05%. The silicon
measurement has an accuracy of about ±0.15%
To achieve good results:

the sample must be poured at a high enough temperature to give a well-

defined liquidus arrest. This means that the sampling spoon must be
preheated;
the iron should not be inoculated before testing;
the sample must solidify white.

Special expendable sample moulds are available for the purpose.

Chapter 5
Inoculation of grey cast iron

Introduction
In order to achieve the desired mechanical properties in iron castings, the
liquid iron must have the correct composition and it must also contain
suitable nuclei to induce the correct graphite structure to form on solidification.
The liquid iron must have a suitable ‘graphitisation potential’, this is
determined mainly by its carbon equivalent value, and in particular by the
silicon content. It is normal practice to adjust the graphitisation potential by
controlling the silicon content. However, the effect of other elements must
also be considered. Table 5.1 shows the effect of common alloying elements
relative to silicon for concentrations normally found in practice.
Table 5.1 The graphitising and carbide stabilising effect of
elements relative to Si

Graphitisers Carbide stabilisers

C +3.0 Mn –0.25
Ni +0.3 Mo –0.35
P +1.0 Cr –1.20
Cu +0.3 V –1.0 to 3.0
Al +0.5

From Cast Iron Technology, Elliott, R. (1988), Butterworth-Heinemann,

reproduced by permission of the publishers.

Example: the effect of 1%Al is approximately equivalent to the graphitising

power of 0.5%Si. 1%Cr will neutralise the effect of about 1.2%Si.
Even if iron of the correct chemical analysis is made in the melting furnace,
castings having the desired graphite structure will not be produced without
the addition of inoculants. Inoculants are alloys added in small amounts to
induce eutectic graphite nucleation. Without the presence of suitable nuclei,
liquid iron will ‘undercool’ below the eutectic temperature (Fig. 5.1).
Uninoculated grey iron castings will contain:

undercooled forms of graphite, associated with this will be ferrite;

cementite in thin sections or close to edges and corners.

Such iron is unlikely to meet tensile and hardness specifications and will be
difficult to machine.
 Inoculation of grey cast iron 63

(a)

(b)

Figure 5.1 Graphite structure of (a) uninoculated grey cast iron (×100) and
(b) inoculated grey cast iron (×100). (From BCIRA Broadsheet 161-4; reproduced by
courtesy of CDC.)

There are two main methods of inoculation, ladle and late inoculation. In
the former, the inoculant is added either as the liquid iron enters the ladle
or just afterwards. Late inoculation refers to treatment after the metal has
left the ladle, for example, as it enters the mould (stream inoculation) or by
using an insert in the mould (in-mould inoculation). Inoculants reach
maximum effectiveness immediately after treatment and fade quickly over
a period of 10–20 minutes. It is therefore desirable to inoculate as late as
possible before casting.
64 Foseco Ferrous Foundryman’s Handbook

Inoculants are mostly based on graphite, ferrosilicon or calcium silicide,

with ferrosilicon being the most commonly used. Pure ferrosilicon is not
effective as an inoculant, it is the presence of minor elements that determine
the effectiveness of the product. Graphite itself is a powerful inoculant but
it is not effective on low sulphur irons.

Table 5.2 INOCULIN products for inoculation of grey, ductile and compacted
graphite irons

INOCULIN Active Use

product constituents

10 Graphite, alloys containing General grey iron, especially where

Si, Ca, Al, Zr S exceeds 0.08%. Recommended for
max. chill reduction and best graphite
structures in cupola iron. Does not affect
composition of the iron.
25 Ferroalloy containing All types of grey, ductile and CG irons.
65% Si, Ca, Al, Zr and Mn High solubility even in low temperature
metal.
80 Ferroalloy containing 75% Powerful inoculant for grey, ductile
Si, Ca, Zr, Al and CG irons. Good fade resistance.
90 Specially graded Fine powder inoculant for use in
INOCULIN 25 MSI 90 Stream Inoculator.
98 Specially graded Graded for late stream inoculation in
INOCULIN 80 MSI 90 Stream Inoculator.

In addition to the above range of inoculants, Foseco supplies certain

special grades in some countries for particular applications such as low
sulphur irons and for ductile pipe manufacture.

Ladle inoculation
The selected grade of INOCULIN for ladle inoculation should always be
added to the metal stream when tapping from furnace to ladle, or ladle to
ladle. Additions should begin when the ladle is one-quarter full and be
completed when the ladle is three-quarters full, so that the last metal merely
mixes.
Never put INOCULIN into the bottom of the ladle and tap onto it.
The amount of inoculant needed is governed by several factors. The
following rules guide the use of inoculation:

Low carbon equivalent irons require greater amounts of inoculant.

Grey cast irons with less than 0.06% sulphur are difficult to inoculate,
specially formulated products may be required.
For a given iron, the thinner the section of casting, the greater the inoculation
required.
 Inoculation of grey cast iron 65

Electric melted irons require more inoculation than cupola melted irons.
Electric melting will also produce low sulphur contents.
High steel scrap charges will require more inoculation.
Where inoculated iron is held for more than a few minutes after inoculation,
there is a need of a higher level of treatment.

It is therefore difficult to give an accurate estimate of the amount of INOCULIN

which is required for every situation. In general, INOCULIN additions of
0.1–0.5% by weight of metal will be satisfactory for grey cast irons, higher
additions are needed for ductile (SG) irons (see p. 79). Care must be taken
not to over-inoculate grey irons, otherwise problems will arise with shrinkage
porosity due to too high a nucleation level. Many grades of INOCULIN
contain high Si content, so that by adding 0.5% of inoculant, the silicon
content of the iron will be raised by as much as 0.3%, this must be allowed
for by adjusting the Si analysis of the furnace metal.

Control methods

The wedge chill test is a simple and rapid method of assessing the degree of
chill reduction obtained by the use of INOCULIN in grey cast irons. Carried
out on the foundry floor, the wedge test is frequently used as a routine
check even when full laboratory facilities are available. The most common
dimensions for the wedge are illustrated in Fig. 5.2.

t
h

b I

Base (b) Height (h) Length (I)

mm in mm in mm in
1 7
6 /4 11 /16 57 2 1 /4
1 7
13 /2 22 /8 100 4
19 3
/4 38 11/2 127 5
25 1 57 21/4 127 5

Figure 5.2 The wedge chill test.

The wedge is made in a mould prepared from silicate or resin bonded

sand. After pouring, it must be allowed to cool in the mould to a dull red
heat (c. 600°C), after which it can be quenched in water and fractured. The
width at the point where clear chill ceases, t, is measured and this gives a
good indication of the need for inoculation and of the effectiveness of an
66 Foseco Ferrous Foundryman’s Handbook

inoculation process. In general, casting sections should be not less than

three times the wedge reading if chill at the edges and in thin sections is to
be avoided.
After ladle inoculation, the metal must be cast quickly to avoid inoculant
fade.
For certain applications such as continuous casting of iron bar or automatic
pouring of castings, inoculant can be added in the form of filled steel wire
containing INOCULIN 25 which can be fed into a ladle or the pouring basin
of an automatic pouring machine at a computer-controlled rate using the
IMPREX Station (see pp. 73, 78). IMPREX wire is available in a range of
diameters from 6 mm upwards.

Late stream inoculation

With the increasing number of foundries where castings are made on highly
mechanised moulding and pouring lines, the requirements for inoculation
are becoming more difficult to meet. Particular difficulties arise with the
use of automatic pouring furnaces where conventional ladle inoculation is
not possible. A method of carrying out inoculation at the casting stage is
needed and this must be consistent and automatic in operation. The MSI 90
Metal Stream Inoculator is intended for use in these conditions.
It is designed to add controlled amounts of inoculant to the liquid cast
iron just before it enters the mould. The use of late stream inoculation
techniques leads to the virtual elimination of fading. This permits a substantial
reduction in the amount of inoculant used. The inoculant addition thereby
produces a smaller change in iron composition leading to improved
metallurgical consistency. The cost of inoculation is also lower.
The MSI 90 Stream Inoculator consists of two units, Fig. 5.3, a control unit
and a dispensing unit linked together by a special cable and air line assembly.
The inoculant dispensing cabinet is located in a fixed position over the
mould being poured. A storage hopper for the inoculant is mounted above
the dispensing cabinet. In the latest version, MSI SYSTEM 90-68E, Fig. 5.4,
the flow of inoculant can be regulated either by optical detection of the start
and end of iron flow via an optical module and fibre optic system or by
connecting the system to the pouring furnace electrical signal used to regulate
the flow of liquid iron. The monitoring system checks INOCULIN 90 level,
dispensing tube status, inoculant flow, gate status, compressed air and
dispensing unit temperature. The monitor can automatically interrupt
pouring in the event of malfunction. The control unit is fitted with a printer
port allowing records to be kept. The control cabinet is positioned in a
secure, easily accessible place and may be some distance from the point of
inoculation.
The MSI 90 Stream Inoculator can be operated in conjunction with a
variety of types of pouring equipment:
 Inoculation of grey cast iron 67

Storage hopper for INOCULIN 90

Low-voltage electricity
cable and air line
Sensor to detect
metal stream
Control
unit Ladle

Dispensing unit
controlling flow of inoculant
Metal stream

Delivery tube for INOCULIN 90

to be added to metal stream

Mould

Figure 5.3 The principles involved in the MSI System 90.

Figure 5.4 MSI System 90 Type 68E.

pouring furnaces
ladle transporters
automatic ladle pouring devices
conventional ladles with fixed or variable pouring positions (provided
the latter is within a limited radius).
68 Foseco Ferrous Foundryman’s Handbook

The inoculant used in late stream inoculators must have a number of important
features:

It must be a powerful inoculant.

It must be finely divided to ensure free-flowing properties and rapid
solution.
It must be very accurately graded, without superfine material which
would blow away, or large particles which jam the gate mechanism.
It must dissolve rapidly and cleanly to avoid the presence of undissolved
inoculant particles in the castings.

Filter Ingate
Sprue Runner

Ratio of cross-sectional areas:

INOTAB Sprue : Filter : Runner : Ingate
cast mould 1 : : 1.1 : 1.2
Inoculant

INOTAB and filter application gating

system deslgn

4 : 8 : 3
Conventional gating with INOTAB
cast mould Inoculant

INOTAB cast mould inoculant set in pouring basin

Figure 5.5 Application of INOTAB cast mould inoculant.

 Inoculation of grey cast iron 69

These requirements are met by INOCULIN 90, specially developed for this
purpose. INOCULIN 90 is an inoculating grade of ferroalloy containing
balanced proportions of Si, Mn, Al, Ca and Zr, and is an excellent inoculant
for grey and ductile irons. INOCULIN 90 should not be used for normal
ladle inoculation because of its very fine size grading.
Stream inoculation is very efficient since fading is eliminated. The normal
addition rate for grey iron is from 0.03–0.20%, typically 0.1%, much less
than would be used for ladle inoculation. For ductile iron, addition rates
range from 0.06–0.3%, typically 0.2%.

Mould inoculation
There are several ways in which mould inoculation can be performed:
powdered inoculant can be placed in the pouring bush; or it can be placed
at the bottom of the sprue. A more reliable method is to use sachets or
precast slugs of inoculant in the pouring bush or in the running system
(Fig. 5.5).
INOPAK sachets are sealed paper packets containing 5, 10 or 20 g of
graded, fast-dissolving inoculant which can be placed in the runner bush, at
the top of the sprue or in some other situation where there is a reasonable
degree of movement in the metal stream. For most purposes, the addition
rate should be 0.1%, i.e. 5 g of INOPAK for each 5 kg of iron poured.
INOTAB cast mould inoculant tablets are designed to be placed in the
runner where they gradually dissolve in the metal stream as the casting is
poured, giving uniform dissolution. This ensures that inoculation takes place
just before solidification of the iron. Application is simple using core prints
to locate the INOTAB tablet.
INOTAB tablets are normally applied at 0.07–0.15% of the poured weight
of iron. The metal temperature and pouring time of the casting must be
considered when selecting the tablet weight. A minimum pouring temperature
of 1370°C (2500°F) is recommended. It is important that the INOTAB tablet
is located where there is continual metal flow during pouring to ensure
uniform dissolution and the typical application methods are shown in
Fig. 5.5.
Chapter 6
Ductile iron

Production of ductile iron

Ductile iron, also known as spheroidal graphite (s.g.) iron or nodular iron,
is made by treating liquid iron of suitable composition with magnesium
before casting. This promotes the precipitation of graphite in the form of
discrete nodules instead of interconnected flakes (Fig. 2.4). The nodular iron
so formed has high ductility, allowing castings to be used in critical
applications such as:

Crankshafts, steering knuckles, differential carriers, brake callipers, hubs,

brackets, valves, water pipes, pipe fittings and many others.

Ductile iron production now accounts for about 40% of all iron castings and
is still growing.
While a number of elements, such as cerium, calcium and lithium are
known to develop nodular graphite structures in cast iron; magnesium
treatment is always used in practice. The base iron is typically:

TC Si Mn S P
3.7 2.5 0.3 0.01 0.01

having high carbon equivalent value (CEV) and very low sulphur. Sufficient
magnesium is added to the liquid iron to give a residual magnesium content
of about 0.04%, the iron is inoculated and cast. The graphite then precipitates
in the form of spheroids. It is not easy to add magnesium to liquid iron.
Magnesium boils at a low temperature (1090°C), so there is a violent reaction
due to the high vapour pressure of Mg at the treatment temperature causing
violent agitation of the liquid iron and considerable loss of Mg in vapour
form. This gives rise to the familiar brilliant ‘magnesium flare’ during
treatment accompanied by clouds of white magnesium oxide fume. During
Mg treatment, oxides and sulphides are formed in the iron, resulting in
dross formation on the metal surface, this dross must be removed as
completely as possible before casting. It is important to remember that the
residual magnesium in the liquid iron after treatment oxidises continuously
at the metal surface, causing loss of magnesium which may affect the structure
of the graphite spheroids, moreover the dross formed may result in harmful
inclusions in the castings.
 Ductile iron 71

Several different methods of adding magnesium have been developed,

with the aim of giving predictable, high yields. Magnesium reacts with
sulphur present in the liquid iron until the residual sulphur is about 0.01%.
Until the sulphur is reduced to near this figure, the magnesium has little
effect on the graphite formation. In the formation of MgS, 0.1%S requires
0.076%Mg. A measure of the true Mg recovery of the treatment process can
be expressed as:
0.76 × (S% in base metal – S% residual) + residual Mg%
Mg recovery % =
Mg% added
Mg recovery is lower at high treatment temperatures and is dependent on
the particular treatment process used. Magnesium may be added as pure
Mg, or as an alloy, usually Mg–ferrosilicon or nickel–magnesium. Other
materials include briquettes, called NODULANT, formed from granular
mixtures of iron and magnesium and hollow mild steel wire filled with Mg
and other materials.

Magnesium content of treatment materials

Mg–Fe–Si alloy 3–20%

Ni–Mg alloy 5–15%
Mg ingot or wire >99%
Mg–Fe briquettes 5–15%
Cored wire 40–95%

MgFeSi alloys usually also contain 0.3–1.0% cerium accompanied by other

rare earth elements. 0.5–1.0%Ca is also a common addition to the treatment
alloy.

Typical analysis of magnesium ferrosilicon nodulariser

Element 5% MgFeSi 10% MgFeSi

Si % 44–48 44 –48
Mg % 5.5–6.6 9.0 –10.0
Ca % 0.2–0.6 0.5 –1.0
RE % 0.4–0.8 0.4 –1.0
Al % 1.2 max 1.2 max

RE (rare earths) contain approximately 50%Ce

Treatment methods include:

Sandwich ladle: the treatment alloy is contained in a recess in the bottom

of a rather tall ladle and covered with steel scrap. The method is suitable
for use only with treatment alloys containing less than 10% Mg (Fig. 6.1a).
Tundish cover: this is a development of the treatment ladle in which a
specially designed cover for the ladle improves Mg recovery and almost
eliminates glare and fume (Fig. 6.1c).
72 Foseco Ferrous Foundryman’s Handbook

Plunger: the alloy is plunged into the ladle using a refractory plunger bell
usually combined with a ladle cover and fume extraction (Fig. 6.1d).
Porous plug: a porous-plug ladle is used to desulphurise the metal with
calcium carbide and the treatment alloy is added later while still agitating
the metal with the porous plug.
Converter: a special converter-ladle is used, containing Mg metal in a

Molten iron Molten iron

Ladle
Ladle

Treatment
Cover alloy
Alloy

(a) (b)

(b)

Raising and lowering device

Metal level
Cover

Molten metal

Ladle

Plunging bell

Treatment
alloy

(c) (d)
 Ductile iron 73

Salamander
Stopper plate

Magnesium
chamber

Metal

(e)

(f)

Joint

Ingate to
Cope
casting
or riser
Down-sprue (cope or drag)
Runner bar

Reaction chamber
Inlet
Joint Drag
(g)
Figure 6.1 Treatment methods for making ductile iron. (a) Sandwich treatment.
(b) Pour-over treatment. (c) Tundish cover ladle. (d) Plunging treatment. (e) GF
Fischer converter. (f) IMPREX cored-wire treatment station (g) In-mould system.
74 Foseco Ferrous Foundryman’s Handbook

pocket. The ladle is filled with liquid iron, sealed and rotated so that the
Mg metal is submerged under the iron (Fig. 6.1e).
Cored wire treatment: wire containing Mg, FeSi, Ca is fed mechanically
into liquid metal in a covered treatment ladle at a special station (Fig.
6.1f).
Treatment in the mould (Inmold): MgFeSi alloy is placed in a chamber moulded
into the running system, the iron is continuously treated as it flows over
the alloy (Fig. 6.1g).

All the methods have advantages and disadvantages; simple treatment

methods can only be used with the more costly low-Mg alloys, generally
containing high silicon levels which can be a restriction since a low Si base
iron must be used. In order to use high Mg alloys and pure Mg, expensive
special purpose equipment is needed so the method tends to be used only
by large foundries.
A survey on ductile iron practice in nearly 80 US foundries in 1988 (AFS
Trans. 97, 1989, p. 79), showed that the biggest change in the previous 10
years was the increase in the use of the tundish ladle, used by over half of
the foundries in the survey. The growth had come at the expense of open-
ladle, plunging, porous plug and sandwich processes. More recently, cored-
wire treatment has been developed and its use is growing.

Melting ductile iron base

While the cupola can be used for the production of ductile iron, the need for
high liquid iron temperatures and close composition control has encouraged
the use either of duplexing with an induction furnace, or using a coreless
induction furnace as prime melter.
In the US survey referred to above, coreless induction furnaces were
used by 84% of the smaller foundries (producing less than 200 t/week).
Almost all larger foundries duplexed iron from an acid cupola to an induction
furnace, with channel furnaces being favourite.

Cupola melting and duplexing

If magnesium treatment with MgFeSi alloy is used, a low Si base iron is
needed. The process may be summarised as follows:

Melt in acid cupola, charge foundry returns and steel scrap plus low
sulphur pig iron if necessary.

Tap at around 2.8–3.2%C

0.6–1.0%Si
0.08–0.12%S
 Ductile iron 75

Desulphurise, using porous plug treatment with calcium carbide, to about

0.10%S, carburise to 3.6–3.8%C.
Transfer to induction furnace, adjust C and Si and temperature to required
levels.
Treat with MgFeSi and inoculate.
Cast.

Induction furnace melting

Charge foundry returns, steel scrap, ferrosilicon and carburiser to achieve
the desired composition.
If sulphur is below 0.025%, desulphurisation is not necessary, but the
higher the sulphur content, the more magnesium must be used so the
cost of treatment increases.
Treat with Mg and inoculate.
Cast.

If a converter or cored-wire Mg treatment is used, high silicon base irons are

satisfactory. Separate desulphurisation is not necessary since, with these
processes it is economical to use pure magnesium as a desulphuriser.

Use of the tundish cover ladle

The most commonly used treatment method, particularly in smaller foundries
is the tundish covered treatment ladle. The principle is shown in Fig. 6.1c.
The use of a refractory dividing wall to form an alloy pocket in the bottom
of the ladle gives improved Mg recovery compared to a pocket recessed in
the bottom of the ladle. Treatment batches are usually in the range, 450–
1000 kg. Figure 6.2 shows the design of a ladle suitable for the treatment of
about 450–500 kg of iron. The diameter of the filling hole is chosen to minimise
the generation of fume while allowing the ladle to be filled quickly without
excessive temperature loss. It is essential that the MgFeSi alloy is not exposed
to the liquid iron until quite late in the filling procedure, so the filling hole
is positioned to introduce liquid iron away from the alloy pocket in the
ladle bottom. The Mg alloy in the alloy pocket is covered with steel turnings
or FeSi pieces of size 25 × 6 mm, then when the level in the ladle reaches the
dividing wall, iron flows over and forms a semi-solid mass with the cover
material allowing the ladle to be almost filled before the reaction starts, thus
ensuring good recovery of Mg.
In order to minimise temperature losses during treatment, the ladle and
cover should be separately heated with gas burners before assembly.
Immediately before use, the ladle should be filled with base iron from the
melting furnace and allowed to soak for a few minutes before returning the
iron to the furnace. The prescribed weight of MgFeSi alloy is charged through
76 Foseco Ferrous Foundryman’s Handbook

Alloy feed pipe φ = 75 mm

(removable cap)

203 mm
φ 38 mm

584
241
100

Liquid iron level at 1000 lbs

(455 kg)

φ = 444
610

Cover material
203

MgFeSi alloy

All dimensions in mm

Figure 6.2 Plan and cross-section of tundish/cover Iadle. (From Anderson, J.V.
and Benn, D. (1982) AFS Trans, 90, 159–162.)

the alloy charging tube which is plugged after removal of the charging
funnel. The treatment alloy may be any of the MgFeSi alloys with Mg in the
range 3–6%, additions of 1.5–3.0% are made giving Mg additions of 0.08–
0.15%. Tapping time is usually around 40 seconds.
The temperature loss during treatment is around 50°C, so the tapping
temperature must be adjusted accordingly, treatment temperatures of around
1530°C are commonly used. After treatment, the tundish cover is removed,
the metal transferred to a pouring ladle where inoculation may take place,
then it is cast.
 Ductile iron 77

Sandwich treatment
A popular method of treatment, frequently used in smaller foundries, is the
sandwich method (Fig. 6.1a). This is essentially the same as the tundish-
cover method but carried out in an open ladle. The magnesium alloy is
placed in a pocket in the bottom of the ladle and covered with steel scrap
(2–3% of the metal weight) or a steel plate. The molten metal stream is
directed away from the pocket. The pocket must be deep enough to contain
all the alloy and the steel scrap which should be of small size to produce a
high packing density. The treatment ladle is usually deep, with a height-to-
diameter ratio of 1.5–2.0 : 1, the extra metal depth increases the recovery of
the magnesium, which can be as high as 50% when a 5%Mg alloy is used.

NODULANT
NODULANT briquettes are essentially formed of pure magnesium and
sponge iron. They contain 10% Mg and minor quantities of calcium, cerium,
silicon and carbon. The briquettes weigh between 16 and 20 g and have a
density of 4.3–4.5 g/ml, so they are suitable for use in the sandwich or
tundish cover technique in just the same way that MgFeSi is used. Magnesium
yields are around 40%. The major advantage of using NODULANT is that
a negligible amount of silicon is added during the treatment. This permits
the use of higher silicon in the base iron enabling all the available ductile
iron returns to be used in the charge. By increasing the base silicon, the
lining life of the induction furnace is increased by as much as 40%.

Pure magnesium converter process

The Georg Fischer converter (Fig. 6.1e), has a reaction chamber formed by
a graphite-clay plate of semi-circular section set into the lining of the converter.
Molten iron is charged with the converter in the horizontal position. The
reaction chamber is charged with pure Mg lumps and other additives (if
required) and sealed with a locking stopper. The converter spout is closed
by a pneumatically operated lid. The vessel is then tilted to vertical allowing
a limited amount of metal to enter through holes in the chamber and react
with the magnesium which starts to vaporise. The vapour pressure in the
reaction chamber rises slowing the further entry of liquid iron and allowing
controlled treatment of the contents of the converter. Treatment takes 60–90
seconds. The melt is first deoxidised then desulphurised. When the sulphur
content has dropped to less than 0.002%, the melt starts to absorb magnesium.
Magnesium recovery can be as high as 70%. Base irons with high sulphur
contents of 0.2–0.3%S can be used because of the high efficiency of Mg use.
The standard converter allows up to 2.5 tonnes of iron to be treated 6–8
78 Foseco Ferrous Foundryman’s Handbook

times per hour. Larger units with capacity up to 10 tonnes are available. The
reaction chamber wall has a limited life of 200–800 treatments. Temperature
loss is 22–33°C in a 1-tonne converter but less in larger converters. Since no
Si is added during treatment, an unlimited amount of ductile iron returns
can be used.
The process is operated under licence from Disa Georg Fischer.

Cored-wire treatment
The equipment and principle of the method is shown in Fig. 6.1f which
shows the Foseco IMPREX-Station. The treatment station consists of a coil
of hollow mild steel tube filled with Mg metal, a feeding machine, a guide
tube and a ladle with a close fitting lid. On entering the molten metal, the
sheathing of the wire dissolves releasing the core material below the metal
surface. Wires vary in size from 4 mm to 13 mm. The amount of wire needed
is dependent on the sulphur content of the base iron, the temperature of the
iron and the reactivity of the wire used. Once the treatment parameters
have been established, it is a simple matter to calculate the amount of
wire required and the treatment time. Feed details can then be programmed
into a computer. Typical feed rates of 9 mm wire are 30–50 metres per
minute. 1500 kg of metal can be treated in about 2 minutes. Treatment
temperature starts at around 1450°C, dropping to around 1410°C at the
finish.

In-the-mould treatment
It is possible to carry out the nodularising treatment in the mould by
incorporating a specially designed magnesium treatment chamber in the
gating system into which the treatment alloy is placed (Fig. 6.1g). The gating
method must fill with metal as quickly as possible and must maintain constant
flow conditions so that all the metal, from first to last, is equally treated.
Since treatment of iron with Mg always produces some MgO and MgS
dross, care must also be taken to avoid dross entering the casting. This is not
easy to achieve and requires a good deal of experimentation so the method
is generally only used by large repetition foundries, which are able to devote
considerable time to solving the process problems.

Inhibiting elements
Certain elements which may be present in the base iron have an inhibit-
ing effect on nodule formation, the following elements are known to be
harmful:
 Ductile iron 79

Aluminium above 0.13%

Arsenic above 0.09%
Bismuth above 0.002%
Lead above 0.005%
Tin above 0.04%
Titanium above 0.04%

Antimony, tellurium and selenium are also harmful. The combined effect of
two or more of these elements may be even more harmful. The addition of
cerium and other rare earth elements, together with calcium will neutralise
many of the harmful effects of inhibiting elements and most MgFeSi
nodularising alloys contain 0.3–1.0% Ce and other rare earths. 0.5–1.0% Ca
is also commonly present.

Inoculation and fading

Immediately after treatment, the iron must be inoculated. Larger additions
of inoculant are needed compared with grey iron and from 0.5–0.75% of a
graphitising inoculant such as INOCULIN 25 should be used. Inoculation
treatment is not permanent, the effect begins to fade from the time the
inoculant is added. As the inoculating effect fades, the number of nodules
formed decreases and the tendency to produce chill and mottle increases. In
addition the quality of the graphite nodules deteriorates and quasi-flake
nodules may occur.
When inoculating ductile iron, the inoculant must be added after the
magnesium flare has subsided. A common practice is to tap about half the
metal onto the magnesium alloy and wait for the flare to finish before adding
the inoculant to the tapping stream as the rest of the metal is tapped. If the
metal is transferred from the treatment ladle to a casting ladle, an effective
practice is to make a further small addition of inoculant as the metal is
poured into the casting ladle. About 0.1–0.2% of inoculant is adequate.
Significant fading occurs within five minutes of inoculation. Because of
this problem, late stream or mould inoculation is commonly used in ductile
iron production, see p. 68.

Specifications for ductile cast iron

Table 6.1 lists a number of national and international specifications for ductile
iron, it is necessary to consult the original specifications for details of the
methods of testing and the mandatory values that must be achieved. In
recent years, specifications in different countries have been converging so
they are now all quite similar. Table 6.2 lists the suggested chemical
compositions required to produce castings that meet the specifications in
the as-cast state.
Table 6.1 Specifications for ductile (nodular) cast irons

Country Minimum tensile strength/elongation

Specification (N/mm2/%)

Europe EN-GJS- 350-22 400-18 400-15 450-10 500-7 600-3 700-2 800-2 900-2
CEN 1563:1997
UK* BS2789 350/22 400/18 420/12 450/10 500/7 600/3 700/2 800/2 900/2
1985
USA ASTM A536 60-40-18 60-42-10 65-45-12 70-50-05 80-55-06 80-60-03 100-70-03 120-90-02
1993
Japan JIS FCD 350-22 400-18 400-15 450-10 500-7 600-3 700-2 800-2
G5502
1995
Inter- ISO 1083 350-22 400-18 400-15 450-10 500-7 600-3 700-2 800-2 900-2
national 1987
Hardness
Typical HB <160 130-175 135-180 160-210 170-230 190-270 225-305 245-335 270-360
Typical F F F F&P F&P F&P P P or T TM
structures

Notes: The European CEN 1563 Standard also specifies 350-22-LT and 400-18-LT for low temperatures.
350-22-RT and 400-18-RT for room temperature.
*BS2789 has been withdrawn and replaced by EN1563:1997.
The designations of US Standards e.g. 100-70-03, refers to min. tensile strength(lbf/in2)-min. proof stress-elongation %.
The structures are: TM, tempered martensite; P or T, pearlite or tempered structure; P & F, pearlite and ferrite; F, ferrite.
This Table is intended only as a guide, refer to the National Standards for details.
Table 6.2 Suggested analyses for as-cast production of ductile iron

Average Grades Grade Grade Grade

casting 800/2,700/2,600/3 500/7 400/12 400/18
section TC Si Mn(max) TC Si Mn(max) TC Si Mn(max) TC Si Mn(max)

<13 mm 3.6–3.8 2.6–2.8 0.5 3.6–3.8 2.6–2.8 0.3 3.6–3.8 2.6–2.8 0.2 3.6–3.8 2.6–2.8 0.1
13–25 3.5–3.6 2.2–2.5 0.6 3.5–3.6 2.2–2.5 0.35 3.5–3.6 2.2–2.4 0.25 3.5–3.6 2.2–2.4 0.15
25–50 3.5–3.6 2.1–2.3 0.7 3.5–3.6 2.1–2.4 0.4 3.5–3.6 2.2–2.4 0.3 3.5–3.6 2.2–2.4 0.20
50–100 3.4–3.5 1.9–2.1 0.8 3.4–3.5 2.0–2.2 0.5 3.4–3.5 2.0–2.2 0.35 3.4–3.5 1.8–2.0 0.2
>100 3.4–3.5 1.8–2.0 0.8 3.4–3.5 1.8–2.0 0.6 3.4–3.5 1.8–2.0 0.40 3.4–3.5 1.8–2.0 0.25

Notes: For the higher strength grades, 800,700,600, additions of 0.5% Cu or 0.1% Sn may be made to encourage pearlite formation.
In all grades: Phosphorus should be less than 0.05%
Chromium should be less than 0.05%
Residual Mg should be 0.03–0.06%
82 Foseco Ferrous Foundryman’s Handbook

Heat treatment of ductile iron

It is obviously desirable to achieve the required properties in the as-cast
form, but this is not always possible because of variations of section thickness
etc. Heat treatment of the castings will eliminate carbides in thin sections,
produce more consistent matrix structures and for a given structure, the
mechanical properties are often improved by heat treatment, especially by
normalising. Where tempered martensite structures are needed, heat treatment
is essential.

Stress relief

Heat at 50–100°C/h to 600°C (taking care not to exceed 610°C), soak for one
hour plus an hour for every 25 mm of section thickness in the thickest
section. Cool at 50–100°C/h to 200°C or less. Ensure that the castings are
adequately supported in the furnace so that they are not subjected to stress.

Breakdown of carbides

Thin section castings may contain carbides in the as-cast structure, these
can be eliminated by soaking the castings at 900–925°C for 3 to 5 hours.

Annealing to produce a ferritic matrix

Castings should be soaked at 900–925°C for 3–5 hours, followed by slow

cooling at around 20–35°C/h through the critical temperature (about 800–
710°C), then furnace cooled at, say 50–100°C/h to 200°C.

Normalising to produce a pearlitic matrix

Soak the castings above the critical temperature then air cool. Again a soaking
temperature of 900–925°C is usually used, to ensure that carbides are broken
down, then use forced air cooling to form pearlite. The type of heat treatment
furnace available and the size of the load determines the cycle that is possible.
It may be necessary to adjust the metal composition with tin or copper to
help the formation of fully pearlitic structures.

Hardened and tempered structures

Austenitise at 900–920°C then oil quench. Tempering is usually carried out
at 600–650°C.
 Ductile iron 83

Austempered ductile iron (ADI)

Austempering is an isothermal heat treatment for producing ‘bainitic’

structures. It can double the strength of ductile iron while retaining good
ductility and toughness. Wear resistance and fatigue properties are excellent
so that ADI is comparable with wrought steel.
The ADI heat treatment is a two-stage process, shown in Fig. 6.3.
Austenitising is carried out at 815–930°C to fully transform the matrix to
austenite. This is done either in a non-oxidising atmosphere furnace or in a
high temperature salt bath, temperatures and times are determined by
chemical composition, section size and grade of ADI required. 1 to 1.5 hours
is usually adequate. Slow initial heating of the casting is desirable to avoid
the danger of cracking of complex shapes. The castings are then quenched
to the required isothermal heat treatment temperature, usually between 210
and 400°C. This is usually done in a salt bath (Fig. 6.3). The castings are held
at temperature for 1–2 hours to complete the transformation of austenite to
bainite. The lower temperatures give high hardness, strength and wear
resistance, while the higher heat treatment temperatures result in higher
ductility and toughness. After the isothermal treatment, the castings are
cooled to ambient temperature.
Unalloyed ductile irons may be austempered in sections up to about
8 mm. Thicker section castings require the addition of Mo or Ni to increase
the hardenability.
Typical changes in properties due to austempering of an unalloyed iron are:

As-cast Austempered 1h Austempered 1h

at 300°C at 375°C

Tensile strength (N/mm2) 475 1465 1105

Elongation (%) 19 1 9
Hardness (HB) 160 450 320

Austempered ductile iron finds applications as a replacement for forged

steel components in the agricultural, mining, automotive and general
engineering industries; for example: plough tips, digger teeth, spring brackets,
rear axle brackets, gears etc. ADI production is growing but its use is limited
to some extent by the lack of suitable heat-treatment facilities.
The European/British Specification BS EN 1564:1997 defines four grades
of ADI, Table 6.3.
Table 6.3 European grades of ADI

Material designation Tensile 0.2%PS Elongation Hardness

strength (MPa) (MPa) (%) (HB)

EN-GJS-800-8 800 500 8 260–320

EN-GJS-1000-5 1000 700 5 300–360
EN-GJS-1200-2 1200 850 2 340–440
EN-GJS-1400-1 1400 1100 1 380–480
84 Foseco Ferrous Foundryman’s Handbook

Mechanical properties are measured on test pieces machined from

separately cast test pieces.
In North America, the ASTM has defined five standard grades of ADI,
Table 6.4.
Table 6.4 The five ASTM standard ADI grades (ASTM A897M-90)

Grade TensiIe* Yield* Elongation* Impact Typical

strength strength (%) energy* hardness
(MPa) (MPa) (Joules) (BHN)

1 850 550 10 100 269–321

2 1050 700 7 80 302–363
3 1200 850 4 60 341–444
4 1400 1100 1 35 388–477
5 1600 1300 N/A N/A 444–555

*Minimum values

Casting ductile iron

Ductile iron differs from grey iron in its casting characteristics in two important
respects. Unlike grey iron, ductile iron is a dross-forming alloy. The residual
magnesium which is needed to ensure nodular graphite formation rapidly
oxidises whenever the liquid metal is exposed to air; in the ladle, during
metal transfer and in the mould. A magnesium silicate dross is formed
which may give rise to defects at or just below the casting surface, usually
on the upper surfaces of the castings. For this reason, it is common practice
to filter ductile iron castings through ceramic filters, see Chapter 18.
The other major difference compared with grey iron, is the need to feed
ductile iron castings to ensure freedom from shrinkage defects. Ductile irons
always have a high carbon equivalent so the volume of graphite precipitated
during solidification should ensure completely sound sections. However,
the expansion resulting from the graphite precipitation results in large
pressures being exerted on the mould walls, much higher than those found
with grey iron (figures of 1000–1500 kPa or 145–217 lbf/in2 have been
measured compared with 170–200 kPa or 24–29 lbf/in2 in grey iron). Only
the strongest moulds, such as sodium silicate bonded moulds or well-vibrated
lost foam moulds, will resist such pressures so the use of feeders is common
when ductile iron castings are made.

Compacted graphite irons

Compacted graphite (CG) irons are a range of cast irons having mechanical
properties intermediate between those of grey and ductile irons. Under the
microscope the graphite appears as short, thick flakes with rounded ends,
readily distinguished from true flake graphite (Fig. 6.4). Compacted graphite
 Ductile iron 85

1000
930°C

815°C
800
Temperature °C

600
400°C
(260–320 BHN)
Grade 1
400
AD1
230 °C
21/2 hr max
(400–500 BHN)
200 Grade 4
AD1

0 1 2 3 4 5 6 7 8
Time hr
Typical austempering heat treatment stages

Figure 6.3 Typical austempering heat-treatment stages. (R.D. Forrest, 13th DISA/
GF Licensee Conference 1997. Courtesy Rio Tinto Iron and Titanium GmbH.)

100 µm CGI microstructure

Figure 6.4 Structure of compacted graphite. (Courtesy SinterCast.)

is interconnected in a branched structure, and is more like flakes than the
completely isolated nodules in ductile iron. Compacted graphite is not the
same as vermicular graphite which can occur in irons of very low sulphur.
Compacted graphite iron has good casting characteristics, due to its high
fluidity and low solidification shrinkage. The tensile, yield and fatigue
strengths of CG iron are 1.5–2 times that of grey iron, approaching ductile
iron. Thermal conductivity is comparable with grey iron and machinability
86 Foseco Ferrous Foundryman’s Handbook

is intermediate between grey and ductile irons of similar matrix structures.

This combination of properties makes CG iron suitable for upgrading of
castings traditionally produced of grey iron.

Production of compacted graphite iron

There are several methods by which compacted graphite may be produced:

Cerium additions
Magnesium additions
Nitrogen additions
Magnesium plus titanium additions

One method involves the joint addition of a nodularising and a de-

nodularising agent, usually magnesium and titanium often with a small
cerium addition as well. Special treatment alloys have been developed for
the production of CG iron.

Mg Ti Ce Ca Al Si Fe
4.5–5.5 8.0–10.0 0.3–0.4 1.0 max 1.5 max 50–54 balance

As with ductile iron, it is desirable to start with a low sulphur iron, below
0.02%. The base iron for treatment is best melted in an induction furnace
and should have composition in the range:

CE C Si Mn S P
3.7–4.5 3.1–3.9 1.7–2.9 0.1–0.6 0.035 max 0.06 max

A sandwich technique may be used with addition rates between 0.6–1.6%

depending on the foundry conditions. The treatment temperature should
be above 1350°C to avoid the formation of a fully nodular structure. Inoculation
of the treated iron is necessary and additions of 0.2–0.5% ferrosilicon are
common. The matrix structure may be made ferritic or pearlitic as with
ductile iron. Titanium-containing alloys can cause problems due to build-
up of Ti in foundry returns leading to impaired machinability.
The use of compacted graphite iron was limited for many years by the
difficulty foundries experienced in controlling the process within the narrow
range of CGI stability so it was considered an unreliable material.
A Swedish company, SinterCast, has developed a method of process
control for CGI which allows CGI to be made reliably. The process can be
licensed by foundries. Figure 6.5 represents the transition of graphite
morphology from flake to compacted and ultimately to spheroidal graphite
as a function of Mg content. This suggests that a range of 0.005–0.010% total
Mg will cause formation of compacted graphite. This curve does not fully
account for reactions between Mg and dissolved elements in the iron,
particularly oxygen and sulphur. Figure 6.6 illustrates by deep etched scanning
 Ductile iron 87

Grey iron to ductile iron transition

100
Ductile iron
80
% Nodularity

60

40

20
CGI

0
Grey

0.010 0.020 0.030 0.040 % Total Mg

Figure 6.5 The effect of total Mg content on the transformation of graphite morphology
from flake to compacted and ultimately to spheroidal graphite. (Courtesy SinterCast.)

Figure 6.6 The transition of graphite morphology from flake to compacted to

spheroidal shown by deep etched scanning electron micrographs.
(Courtesy SinterCast.)

electron micrographs, the transition of graphite morphology from flake to

compacted to spheroidal. The SinterCast Process involves taking a sample
of liquid iron from the ladle in a patented sampling cup and carrying out a
thermal analysis using two thermocouples, one in the centre of the cup and
the other adjacent to the wall. The two cooling curves allow the degree of
modification and inoculation required to be determined. Independent
additions of cored wire containing Mg and inoculant are then made to the
ladle. Ladle to ladle variations in the oxygen and sulphur content of the iron
are thus allowed for and reliable compacted graphite iron is made.
88 Foseco Ferrous Foundryman’s Handbook

Foundry properties of compacted graphite iron

The fluidity is governed by carbon equivalent (CE) and temperature and is

similar to grey or ductile irons of the same CE. Because CG irons are stronger
than grey irons, a higher CE can be used to obtain the same strength, this
allows greater fluidity and easier running of thin sections. CG iron is dross-
forming, just as ductile iron, and filtration of castings is desirable. CG irons
are more prone to chill than grey irons but less likely to chill than ductile
iron. Good inoculation is necessary. Fading occurs, but to a smaller extent
than in ductile iron, but excessive delays between treatment, inoculation
and casting should be avoided.
There is some disagreement about the level of feeding required for CG
iron. There is less tendency for mould wall movement than with ductile
iron, nevertheless some feeding appears to be desirable.

Applications of compacted graphite irons

There has been great interest in its use for automotive castings such as
diesel cylinder blocks and heads, hydraulic components, exhaust manifolds,
brake drums, brake discs, flywheels etc. The lack of consistency of properties
held back the wide scale application of CG iron but with the greater control
now possible its use is expected to develop.

Properties of compacted graphite irons

Table 6.5 Comparison of CG iron properties with grey and ductile irons

Property Grey irons CG irons Ductile irons

Tensile strength
ton/in2 11–20 20–38 26–45
lb/in2 25–45 000 45–85 000 60–100 000
kg/mm2 16–32 30–60 40–70
N/mm2 160–320 300–600 400–700
Elongation (%) nil 3–6 6–25
Modulus (lb/in2) 14–16×106 20–23×106 25–27×106
(GN/m2) 96–110 140–160 170–190
Charpy impact
Joules, 25°C nil 3–7 17
Fatigue limit
un-notched (ton/in2) 7–8 15–20 12–18
(N/mm2) 108–123 230–310 185–280
Machinability very good very good good
Corrosion moderate intermediate good
resistance
 Ductile iron 89

Table 6.6 Specifications for compacted graphite iron: ASTM A842-85 (reapproved
1991) compacted graphite iron

Grade
250a 300 350 400 450b

Tensile strength (MPa) 250 300 350 400 450

Yield strength (MPa) 175 210 245 280 315
Elongation (%) 3.0 1.5 1.0 1.0 1.0
Hardness (HB) 179 max 143–207 163–229 197–225 207–269
a
ferritic grade
b
pearlitic
Hardness not mandatory
Chapter 7
Malleable cast iron

Introduction
Malleable irons are cast white, that is, their as-cast structure consists of
metastable carbide in a pearlitic matrix. The castings must then be annealed
to convert the brittle carbide structure and develop a structure of roughly
spherical graphite aggregates in a matrix which can be either ferritic or
pearlitic, depending on composition and heat treatment.
There are two types of malleable iron, blackheart and whiteheart. Malleable
iron has a long history, whiteheart iron having been developed in 1722 by
the French metallurgist, Réaumur, while blackheart iron was developed in
the USA in 1820. Malleable iron was widely used for automotive and
agricultural components, pipe fittings, valves etc. but since the development
of spheroidal graphite ductile iron its use has declined, due to the high cost
of the annealing treatment which requires expensive furnace equipment.
Malleable iron is still widely used for small pipe fittings, electrical fittings
and builders hardware, particularly for thin section castings and castings
which are subsequently galvanised.

Whiteheart malleable
In the whiteheart process, the white, as-cast iron is decarburised during
annealing leaving a structure of iron carbide in a metallic matrix. When
fractured, the appearance is whitish, giving rise to the name, ‘whiteheart’.
Decarburisation is only possible in thin sections; in heavier sections, some
conversion of carbide to graphite nodules occurs so that the annealed casting
has a white rim with a core having different structure and mechanical
properties. This limits the applications to which it can be put. Whiteheart
can be melted in a cupola and is a low cost material which still finds
applications in small, thin section castings.

Composition of whiteheart malleable

Typical compositions are

 Malleable cast iron 91

Before annealing After annealing

Total carbon 3.0–3.7% 0.5–2.0%

Silicon 0.4–0.8 0.4–0.7
Manganese 0.1–0.4 0.1–0.4
Sulphur 0.3 max 0.3 max
Phosphorus 0.1 max 0.1 max

Annealing is a combined decarburisation and graphitisation process

performed in an oxidising atmosphere. Originally it was done by packing
castings into iron ore mixtures but now it is carried out in continuous,
atmosphere controlled furnaces at about 1070°C. Small castings may be
fully decarburised and are referred to as weldable malleable irons. Table 7.1
lists the European specifications for whiteheart malleable iron. The USA has
no equivalent standard.

Table 7.1 Specifications for whiteheart malleable cast irons

Grade Test bar Tensile Elongation 0.2% Proof HB

dia. strength strength typical
(mm) (min) (%) (min)
(N/mm2) (N/mm2)

EN-GJMW-300-4 12 350 4 – 230

EN-GJMW-360-12 12 360 12 190 200
EN-GJMW-400-5 12 400 5 220 220
EN-GJMW-450-7 12 450 7 260 260
EN-GJMW-550-4 12 550 4 340 340

International ISO 5922–1981

Whiteheart
W 35-04 9 340 5 –
12 350 4 – 230 max
15 360 3 –
W 38-12 9 320 15 170
12 380 12 200 200 max
15 400 8 210
W 40-05 9 360 8 200
12 400 5 220 220 max
15 420 4 230
W 45-07 9 400 10 230
12 450 7 260 150 max
15 480 4 280

The European Standard CEN 1562:1997 has superseded the former national standards,
for example:
France: NF A32-701(1982); Germany: DIN 1692 (1982); UK: BS6681:1986.
Notes: It is advisable to consult the original standards for details of the mandatory
values, methods of testing etc.
USA has no standard for whiteheart malleable iron.
92 Foseco Ferrous Foundryman’s Handbook

Blackheart malleable iron

The iron is typically melted in a cupola and duplexed into an electric furnace
where temperature and composition are adjusted. The cupola metal has the
composition:

C Si Mn S P
2.5–2.6 1.1 0.2–0.3 0.2 0.1

The final composition in the electric furnace is:

C Si Mn S P Cr
2.4–2.6 1.3–1.45 0.4–0.55 0.2 max 0.1 max 0.05 max

Castings are poured at around 1450°C. Sometimes small additions (around

0.01%) of bismuth are added in the ladle to ensure fully white as-cast
structures. White irons contract on solidification so to ensure freedom from
shrinkage, the castings must be fed, see Chapter 19. As-cast malleable iron
is extremely brittle, allowing feeders and running systems to be broken
easily from the castings. Castings which have abrupt changes of section
develop internal stresses on cooling which may be enough to cause cracking
of the castings after shakeout. Slow cooling in the mould may be needed to
avoid this happening.
The castings are annealed to develop the required graphite clusters (Fig.
2.3). A typical cycle is about 48 hours long (Fig. 7.1) it may be carried out in
batch-type or continuous furnaces in a controlled atmosphere to avoid
oxidation of the castings. The rate of cooling in the final section of the heat

1st stage 2nd stage

960
Temperature (°C)

790

720

500

12 24 36 48 60
Time (h)
Figure 7.1 Typical heat treatment cycle for a short cycle blackheart malleable iron.
(From Elliott, R. (1988) Cast Iron Technology, Butterworth-Heinemann, reproduced
by permission of the publisher.)
 Malleable cast iron 93

treatment determines the matrix structure of the castings which can be

ferritic or pearlitic according to the physical properties required.

Specifications for malleable cast irons

Table 7.2 lists the European and International specifications for blackheart
malleable cast irons. The European Standard CEN 1562:1997 has superseded
the former national standards. Table 7.3 lists the US ASTM specifications.
Table 7.2 European and international specifications for blackheart malleable cast
irons

Grade Test bar Tensile Elongation 0.2% Proof HB

dia. (mm) strength (min.) (%) stress (min.) typical
(N/mm2) (N/mm2)

EN-GJMB-300-6 12 or 15 300 6 – 150 max

EN-GJMB-350-10 12 or 15 350 10 200 150 max
EN-GJMB-450-6 12 or 15 450 6 270 150–200
EN-GJMB-500-5 12 or 15 500 5 300 165–215
EN-GJMB-550-4 12 or 15 550 4 340 180–230
EN-GJMB-600-3 12 or 15 600 3 390 195–245
EN-GJMB-650-2 12 or 15 650 2 430 210–260
EN-GJMB-700-2 12 or 15 700 2 530 240–290
EN-GJMB-800-1 12 or 15 800 1 600 270–320

International Standard ISO 5922-1981

Blackheart
B30-06 12 or 15 300 6
B32-12 12 or 15 320 12 190 150 max
B35-10 12 or 15 350 10 200

Pearlitic
P45-06 12 or 15 450 6 270 150–200
P50-05 12 or 15 500 5 300 160–220
P55-04 12 or 15 550 4 340 180–230
P60-03 12 or 15 600 3 390 200–250
P65-02 12 or 15 650 2 430 210–260
P70-02 12 or 15 700 2 530 240–290
P80-01 12 or 15 800 1 600 270–310

The European Standard CEN 1562:1997 has superseded the former national standards,
for example:
France: NF A32-702(1982); Germany: DIN 1692 (1982); UK: BS6681:1986.
Notes: It is advisable to consult the original standards for details of the mandatory
values, methods of testing etc.
94 Foseco Ferrous Foundryman’s Handbook

Table 7.3 US specifications for blackheart malleable irons

Specification Grade Test bar Tensile Elong. Yield stress Hardness

diameter strength (%) (N/mm2) (HB)
(mm) (N/mm2)

ASTM A 47-90 220M10 15.9 340 10 220 156 max

Ferritic 32510 50000 psi 32000 psi
ASTM A220-88 280M10 15.9 400 10 280 149–197
Pearlitic 40010 60000 psi 40000 psi
310M8 15.9 450 8 310 156–197
45008 65000 psi 45000
310M6 15.9 450 6 310 156–207
45006 65000 psi 45000
340M5 15.9 480 5 340 179–229
50005 70000 psi 50000 psi
410M4 15.9 550 4 410 197–241
60004 80000 psi 60000 psi
480M3 15.9 590 3 480 217–269
70003 85000 70000 psi
550M2 15.9 650 2 550 241–285
80002 95000 psi 80000 psi
620M1 15.9 720 1 620 269–321
90001 105000 psi 90000 psi
ASTM A602-94 M3210 (10) (32000 psi) 156 max1
Automotive M4504 (4) (45000 psi) 163–2172
malleable M5003 (3) (50000 psi) 187–2412
castings M5503 (3) (55000 psi) 187–2413
M7002 (2) (70000 psi) 229–2693
M8501 (1) (85000 psi) 269–3023

ASTM A47-90 (reapproved 1995) covers ferritic malleable irons, a metric version
was reapproved in 1996.
ASTM A220-88 (reapproved 1993 including a metric version) covers pearlitic malleable
irons.
ASTM A602-94 covers automotive malleable castings.
Note: ASTM A602-94 specifies only hardness and heat treatment. Annealed1 Air-
quenched2 and tempered3 Liquid quenched and tempered.
Chapter 8
Special purpose cast irons

Heat resisting alloys

Unalloyed cast iron shows only slight scaling and no growth at temperatures
up to 350°C for times of up to 10 years. At 450°C scaling and growth will
occur in less than one year in most grey, ductile and malleable cast irons. To
achieve greater resistance to heat, special purpose irons must be used.

Temperatures up to 600°C

Unalloyed grey irons can be used up to about 600°C. The presence of

phosphorus up to 1.0% improves the scaling resistance and can be used in
applications such as fire bars where mechanical strength is not a major
consideration.

Temperatures up to 700°C

The addition of about 0.5% Cr improves the oxidation resistance of grey

irons and is also used for fire bars. Automotive exhaust manifolds and
turbo-charger casings are subjected to increasingly high temperatures in
modern engines, unalloyed cast iron can be used at temperatures up to
about 600°C, but for higher temperature service it is necessary to alloy the
iron with Cr and Mo, e.g.

Low alloy grey iron

C Si Mn S P Cr Mo

3.0–3.4 1.6–2.8 0.4–1.0 <0.12 0.1–0.4 0.5–2.0 0.3–0.5

Temperatures up to 750°C

Ductile (spheroidal graphite) irons with silicon in the range 4–6% have
good resistance to scaling and growth and good dimensional stability up to
about 750°C. Molybdenum up to 0.5% may be added to improve the high
96 Foseco Ferrous Foundryman’s Handbook

temperature creep resistance. They are used for exhaust manifolds and turbo-
charger casings and are better able to resist thermal stress than grey iron
castings. Typical compositions are:

High silicon ductile iron

C Si Mn P Mo

3.2 4.0 0.3 max 0.05 max 0.5

2.9 4.0 0.3 max 0.05 max 0.5
2.6 6.0 0.3 max 0.05 max 0.5

Temperatures up to 850°C

For heat resistance up to 850°C, nickel alloy cast irons, known as Ni-resist (a
trade name of the International Nickel Company) are used. Nickel in
combination with Mn and Cu produces a stable austenitic matrix and Cr in
combination with Ni forms an effective oxidation resistant scale. Most
countries have similar standard specifications for austenitic cast irons, based
on the original Ni-Resist specifications. Each country designates the alloys
differently. Table 8.1 lists the designations of the flake graphite austenitic
irons. Table 8.2 lists the specifications of austenitic flake graphite irons.
Spheroidal graphite Ni-resist can also be made. Table 8.3 lists the
international designations. Table 8.4 lists the British specifications.

Temperatures up to 1000°C

High Cr white irons have excellent resistance to oxidation at temperatures

up to 1000°C and can be used in applications where they are not subjected
to impact loading. They are used for furnace parts, sinter pallets, recuperator
tubes etc. Typical analyses are:

C Si Mn S P Cr

2.0 1.4 1.0 0.1 0.1 max 14.0

2.8 1.8 1.6 17.0
1.0 1.0 0.7 30.0
1.3 1.3 1.0 33.0

Ingot moulds

Products such as ingot moulds, slag ladles, pig-moulds etc. are subjected in
use to severe thermal cycling which induces distortion and crazing. High
carbon iron is used for such applications, typically: (Continued on p. 101)
Table 8.1 Designations of austenitic cast irons with flake graphite

Trade name ISO France Germany UK USA ASTM A436

2892:1973 NF A 32-301 DIN 1694 BS 3468:1986 1984
1972 1981 (reapproved 1994)

– L-NiMn 13 7 L-NM 13 7 GGL-NiMn 13 7 – –

Ni-Resist 1 L-NiCuCr 15 6 2 L-NUC 15 6 2 GGL-NiCuCr 15 6 2 F1 Type 1
Ni-Resist 1b L-NiCuCr 15 6 3 L-NUC 15 6 3 GGL-NiCuCr 15 6 3 – Type 1b
Ni-Resist 2 L-NiCr 20 2 L-NC 20 2 GGL-NiCr 20 2 F2 Type 2
Ni-Resist 2b L-NiCr 20 3 L-NC 20 3 GGL-NiCr 20 3 – Type 2b
Nicrosilal L-NiSiCr 20 5 3 L-NSC 20 5 3 GGL-NiSiCr 20 5 3 – –
Ni-Resist 3 L-NiCr 30 3 L-NC 30 3 GGL-NiCr 30 3 F3 Type 3
Ni-Resist 4 L-NiSiCr 30 5 5 L-NSC 30 5 5 GGL-NiSiCr 30 5 5 – Type 4
– L-Ni 35 L-N 35 – – Type 5
– – – – – Type 6
Table 8.2 ISO and UK specifications for austenitic flake graphite irons

Tensile strength
ISO grade UK C max Si Mn Ni Cr Cu (kgf/mm2) (N/mm2) (ton/in2)

L-NiMn 13 7 – 3.0 1.5–3.0 6.0–7.0 12.0–14.0 0.2 max 0.5 max 14.3 140 9.1
L-NiCuCr 15 6 2 F1 3.0 1.0–2.8 0.5–1.5 13.5–17.5 1.0–2.5 5.5–7.5 17.3 170 11.0
L-NiCuCr 15 6 3 – 3.0 1.0–2.8 0.5–1.5 13.5–17.5 2.5–3.5 5.5–7.5 19.4 190 12.3
L-NiCr 20 2 F2 3.0 1.0–2.8 0.5–1.5 18.0–22.0 1.0–2.5 0.5 max 17.3 170 11.0
L-NiCr 20 3 – 3.0 1.0–2.8 0.5–1.5 18.0–22.0 2.5–3.5 0.5 max 19.4 190 12.3
L-NiSiCr 20 5 3 – 2.5 4.5–5.5 0.5–1.5 18.0–22.0 1.5–4.5 0.5 max 19.4 190 12.3
L-NiCr 30 3 F3 2.5 1.0–2.0 0.5–1.5 28.0–32.0 2.5–3.5 0.5 max 19.4 190 12.3
L-NiSiCr 30 5 5 – 2.5 5.0–6.0 0.5–1.5 29.0–32.0 4.5–5.5 0.5 max 17.3 170 11.0
L-Ni 35 – 2.4 1.0–2.0 0.5–1.5 34.0–36.0 0.2 max 0.5 max 12.2 120 7.8

Note: Refer to the standards for further details, the above figures are intended only as a guide.
Table 8.3 Designations of heat-resisting austenitic spheroidal graphite irons

International ISO 2892 UK France Germany US ASTM A439-83

trade name BS 3468:1986 NF A 32-301 DIN 1694 (reapproved 1994)

– S–NiMn 13 7 S6 S-NM 13 7 GGG-NiMn 13 7 –

Ni-Resist D-2 S-NiCr 20 2 S2 S-NC 20 2 GGG-NiCr 20 2 D-2
Ni-Resist D-2W -– S2W – GGG-NiCrNb 20 2 –
Ni-Resist D-2B S-NiCr 20 3 S2B S-NC 20 3 GGG-NiCr 20 3 D-2B
Nicrosilal
Spheronic S-NiSiCr 20 5 2 – S-NSC 20 5 2 GGG-NiSiCr 20 5 2 –
Ni-Resist D-2C S-Ni 22 S2C S-N 22 GGG-Ni 22 D-2C
Ni-Resist D-2M S-NiMn 23 4 S2M S-NM 23 4 GGG-NiMn 23 4 –
Ni-Resist D-3A S-NiCr 30 1 S3 S-NC 30 1 GGG-NiMn 30 1 D-3A
Ni-Resist D-3 S-NiCr 30 3 – S-NC 30 3 GGG-NiCr 30 3 D-3
Ni-Resist D-4A – – – GGG-NiSiCr 30 5 2 –
Ni-Resist D-4 S-NiSiCr 30 5 5 – S-NSC 30 5 5 GGG-NiSiCr 30 5 5 D-4
Ni-Resist D-5 S-Ni 35 – S-N 35 GGG-Ni 35 D-5
Ni-Resist D-5B S-NiCr 35 3 – S-NC 35 3 GGG-NiCr 35 3 D-5B
Ni-Resist D-5S – S5S – GGG-NiSiCr 35 3 2 D-5S

Note: For detailed specifications refer to the national and international standards.
Table 8.4 British Specification BS 3468(1986) Heat-resisting austenitic spheroidal graphite irons

Grade C Si Mn Ni Cu Cr Nb P Mg TS(N/mm2 ) Elong. (%) HB Impact

J, min.

S2 3.0 1.5–2.8 0.5–1.5 18–22 0.5 1.5–2.5 – 0.08 – 370–490 7–20 140–230 –
S2B 3.0 1.5–2.8 0.5–1.5 18–22 0.5 2.5–3.5 – 0.08 – 370–490 7–20 140–230 4
S2C 3.0 1.5–2.8 1.5–2.5 21–24 0.5 0.5 – 0.08 – 370 min 20 min 140–220 20
S2M 3.0 1.5–2.5 4.0–4.5 21–24 0.5 0.2 – 0.08 – 420 min 25 min 160–250 15
S2W 3.0 1.5–2.2 0.5–1.5 18–22 0.5 1.5–2.2 0.12–0.2 0.05 0.06 370–490 7–20 140–200 –
S3 2.5 1.5–2.8 0.5–1.5 28–32 0.5 2.5–3.5 – 0.08 – 370–490 7–20 140–230 –
S5S 2.2 4.8–5.4 1.0 34–36 0.5 1.5–2.5 – 0.08 – 370–470 7–20 130–180 –
S6 3.0 1.5–2.5 6–7 12–14 0.5 0.2 – 0.08 – 390 min 15 min 120–200 –

Note: Refer to British Specification BS 3468(1986) for further details, the above figures are intended only as a guide.
 Special purpose cast irons 101

C Si Mn S P

3.6–3.9 1.4–1.8 0.7–1.0 as low as possible

The iron is of eutectic composition or slightly hyper-eutectic, since this is

found to give the maximum mould life. Compacted graphite iron is also
used for this application (see p. 84).

Corrosion resistant cast irons

Unalloyed cast irons exhibit reasonable corrosion resistance, particularly to
alkali environments, but special alloys have been developed for use in
corrosive conditions. These include, high nickel (Ni-Resist), high silicon
and high chromium alloys. Table 8.5 summarises the corrosion resistance of
various alloy irons in different media.

Table 8.5 Corrosion resistance of cast irons

Unalloyed grey, Ni-Resist High silicon High chromium

ductile and See Table 8.1 10–16% Si 30% Cr, 1–2% C
malleable

Atmospheric B A A A
including coastal
Water soft B A A A
hard A A A A
sea C A A A
Dilute acid C B A A1
Concentrated acid C2 C A3 C
Dilute alkali A A C A
Concentrated alkali B A C C

Notes: A – Good resistance 1 Unalloyed iron resists concentrated sulphuric

acid quite well
B – Medium resistance 2 High silicon irons do not resist hydrofluoric
acid
C – Poor resistance 3 High chromium irons are less resistant to
hydrochloric acid

Ni-Resist alloy irons are used for the manufacture of pumps for the handling
of sulphuric acid and alkalis, for sea water pumps and for handling sour
crude oils in the petroleum industry.
High silicon cast irons, 10–16% Si, are suitable for use in chemical plant
and for cathodic protection anodes. The grades shown in Table 8.6 are used.
High silicon irons are low strength, brittle materials which are susceptible
to cracking due to stresses induced by cooling. Moulds and cores used for
their manufacture should have good breakdown properties. Electric furnace
melting is necessary because of their low carbon content. The alloys are
102 Foseco Ferrous Foundryman’s Handbook

Table 8.6 High silicon cast irons

Grade C (max) Si Mn (max) S (max) P (max) Cr

Si 10 1.2 10.00–12.00 0.5 0.1 0.25 –

Si 14 1.0 14.25–15.25 0.5 0.1 0.25 –
SiCr 14 4 1.4 14.25–15.25 0.5 0.1 0.25 4.0–5.0
Si 16 0.8 16.00–18.00 0.5 0.1 0.25 –

Grade Si 14 is used for general applications

Si 10 has higher tensile strength but lower corrosion resistance
Si 16 has high corrosion resistance but lower strength
Si Cr 14 4 is used for the manufacture of cathodic protection anodes
For further details, refer to BS 1591.

prone to unsoundness due to hydrogen gas and care must be taken during
melting to avoid contact with hydrogen, by the avoidance of moisture on
charge materials and refractories and maintaining low levels of Al in the iron.
Heat treatment is advisable to eliminate internal stresses, the castings
should be stripped from the moulds as hot as possible and charged into a
furnace at about 600°C then heated to 750–850°C for 2–8 hours, depending
on metal section. They should then be slowly cooled to below 300°C before
removing from the furnace.
High silicon irons have high hardness and are difficult to machine; grinding
is preferred.

Wear resistant cast irons

Wear resistant alloy irons are an important class of cast irons, they find wide
application for the manufacture of mineral crushing plant, e.g hammers for
rock crushers, grinding balls, liners for crushing mills etc. They are also
used in shot blasting machines for impellers, liners etc. Three types of cast
iron are used:

Unalloyed and low alloy grades of white iron having a structure of massive
carbides in a pearlitic matrix. These alloys are extremely brittle and have
been largely superseded by tougher, alloyed white irons.
Ni-hard irons containing Ni to increase hardenability by ensuring that
austenite transforms to martensite after heat treatment. They also contain
Cr to increase the hardness of the carbide.
High Cr-Mo white irons combining abrasion resistance with toughness.
The Mo increases hardenability allowing heavy section castings to be
made with a martensitic structure, either as cast or heat treated.

Low alloy cast irons

These include irons having the compositions:

 Special purpose cast irons 103

BS 4844:1986 C Si Mn Cr

1A,1B,1C 2.4–3.4 0.5–1.5 0.2–0.8 2.0 max

Grade Hardness (min) HB

Thickness < 50 mm Thickness > 50 mm

1A 400 350
1B 400 350
1C 250 200

Such irons are extremely brittle and have been largely superseded by alloyed
irons.

Ni-hard irons

The first alloyed irons to be developed were the Ni-hards (a trade name of
the International Nickel Company). They contain Ni to increase hardenability,
ensuring that austenite transforms to martensite after heat treatment. The
Cr content increases the hardness of the eutectic carbide. The choice of alloy
composition depends on casting thickness and intended use. For maximum
abrasion resistance, the C content should be at the top of the range, 3.2–
3.6%, but for impact resistance it is usual to use lower carbon contents, 2.7–
3.2%. Table 8.7 lists a number of national specifications for alloy irons of the
Ni-hard type. Note that the correspondence between specifications is not
exact and the standards themselves must be referred to for details.
Ni-hards are usually stress relieved at 200–230°C for at least 4 hours to
improve strength and impact resistance.

High Cr-Mo irons

The addition of Cr in the range 12–28%, together with Ni and Mo forms a

range of alloys which combines abrasion resistance with toughness suitable
for large size equipment in the mining, coal and mineral processing industries.
Eutectic carbides of the form, M7C3 are formed and the matrix can be austenitic,
martensitic or pearlitic according to the application. Some components may
be cast pearlitic to aid machining, then heat-treated to give an abrasion
resistant martensitic structure. Tables 8.8 and 8.9 list specifications of this
type.

Production of high chromium irons

An induction furnace is normally used to achieve the required temperature.

Table 8.7 European specifications for abrasion resistant white irons

Hardness HB min
Trade name Country Specification Grade C Si Mn Ni Cr Cu Mo Pmax Smax As–cast Hardened Annealed

Ni-hard 1 France NF A 32-401 FB Ni4 Cr2 HC 3.2–3.6 0.2–0.8 0.3–0.7 3.0–5.51 1.5–2.5 – 0.0–1.0 – 500–700
1980
Germany DIN 1695 G-X330 NiCr4 2 3.0–3.6 0.2–0.8 0.3–0.7 3.3–5.0 1.4–2.4 – 0.5 450(430)2 550(500) –
1981
UK BS 4844:1986 2B 3.2–3.6 0.3–0.8 0.2–0.8 3.0–5.5 1.5–2.5 – 0.5 max 0.3 0.15 5503
1
Ni-hard 2 France FB Ni4 Cr2 BC 2.7–3.2 0.2–0.8 0.3–0.7 3.0–5.5 1.5–2.5 – 0.0–1.0 – – 450–650
Germany G–X260 NiCr4 2 2.6–2.9 0.2–0.8 0.3–0.7 3.3–5.0 1.4–2.4 – 450(430)2 520(480) –
UK 2A 2.7–3.2 0.3–0.8 0.2–0.8 3.0–5.5 1.5–2.5 – 0.5 max 0.3 0.15 5003

Ni-hard 3 France FB A 2.7–3.9 0.4–1.5 0.2–0.8 03–3.04 0.2–2.04 0.3–2.04 0.1–1.04 – – 400–600

Germany –

UK –

Ni-hard 4 France FB Cr9 Ni5 2.5–3.6 1.5–2.2 0.3–0.7 4.0–6.0 5.0–11.0 – 0.5 max 550–750

Germany G-X 300 2.5–3.5 1.5–2.2 0.3–0.7 8.0–10.0 4.5–6.5 – 0.5 450(430) 600(535)
CrNiSi 9 5 2
UK 2C 2.4–2.8 1.5–2.2 0.2–0.8 4.0–6.0 8.0–10.0 – 0.5 max 0.3 0.15 500
2D 2.8–3.2 1.5–2.2 0.2–0.8 4.0–6.0 8.0–10.0 – 0.5 max 0.3 0.15 550
2E 3.2–3.6 1.5–2.2 0.2–0.8 4.0–6.0 8.0–10.0 – 0.5 max 0.3 0.15 600

Notes: This table is for comparison purposes only. The National Standards should be referred to for details.
1
Ni may be partly replaced by Cu.
2
Supply as-cast only when no requirements are made about impact strength.
3
In castings with section thickness greater than 125 mm, HB may be 50 less.
4
Optional or, if necessary, as a combination.
Table 8.8 European specifications for high Cr-Mo abrasion resistant alloys

Country Specification Grade C Si Mn Ni Cr Cu Mo P S

France NF A 32–401 FB Cr15MoNi 2.0–3.6 0.2–0.8 0.5–1.0 0.0–2.5 14.0–17.0 – 0.5–3.0 – –

1980
Germany DIN 1695 G-X300CrMo15 3 2.3–3.6 0.2–0.8 0.5–1.0 0.71 14.0–17.0 – 1.0–3.0 – –
1981 CrMoNi15 2 1 2.3–3.6 0.2–0.8 0.5–1.0 0.8–1.21 14.0–17.0 – 1.8–2.2 – –
UK BS 4844 3A 2.4–3.0 1.0 max 0.5–1.5 0.0–1.0 14.0–17.0 0–1.2 0.0–2.5 0.1 –
1986 3B 3.0–3.6 1.0 max 0.5–1.5 0.0–1.0 14.0–17.0 0–1.2 1.0–3.0 0.1 0.1
France FB Cr20MoNi 2.0–3.6 0.2–0.8 0.5–1.0 0.0–2.0 11.0–14.0 – 0.5–3.0 – –
Germany G-X260CrMoNi 2.3–2.9 0.2–0.8 0.5–1.0 0.8–1.21 17.0–22.0 – 1.8–2.2 – –
20 2 1
UK 3C 2.2–3.0 1.0 max 0.5–1.5 0.0–1.5 17.0–22.0 2.0 max 0.0–3.0 0.1 0.06
2
France FB Cr26MoNi 1.5–3.5 0.2–1.2 0.5–1.5 0.0–2.5 22.0–28.0 0.0–1.5 0.5–3.0 – –
Germany G-X260Cr27 2.3–2.9 0.5–1.5 0.5–1.5 1.2 24.0–28.0 – 1.0 – –
G-X300CrMo27 1 3.0–3.5 0.2–1.0 0.5–1.0 1.2 23.0–28.0 – 1.0–2.0 – –
UK 3D 2.4–2.8 1.0 max 0.5–1.5 0–1.0 22.0–28.0 2.0 max 0–1.5 0.1 0.06
3E 2.8–3.2 1.0 max 0.5–1.5 0–1.0 22.0–28.0 2.0 max 0–1.5 0.1 0.06

Notes: This table is for comparison purposes only. The National Standards should be referred to for details.
1
can also be produced as a Cu-containing alloy.
2
in order to obtain a hard austenitic structure, Mn may be 4.0% max.
Table 8.9 USA Specifications for abrasion resistant cast irons: ASTM A532/A532M-93a

Class Type Designation C Mn Si Ni Cr Mo Cu P S

1 A Ni-Cr-Hc 2.8–3.6 2.0 0.8 3.3–5.0 1.4–4.0 1.0 – 0.3 0.15

1 B Ni-Cr-Lc 2.4–3.0 2.0 0.8 3.3–5.0 1.4–4.0 1.0 – 0.3 0.15
1 C Ni-Cr-GB 2.5–3.7 2.0 0.8 4.0 1.0–2.5 1.0 – 0.3 0.15
1 D Ni-HiCr 2.5–3.6 2.0 2.0 4.5–7.0 7.0–11.0 1.5 – 0.1 0.15
11 A 12%Cr 2.0–3.3 2.0 1.5 2.5 11.0–14.0 3.0 1.2 0.1 0.06
11 B 15%Cr-Mo 2.0–3.3 2.0 1.5 2.5 14.0–18.0 3.0 1.2 0.1 0.06
11 D 20%Cr-Mo 2.0–3.3 2.0 1.0–2.2 2.5 18.0–23.0 3.0 1.2 0.1 0.06
111 A 25%Cr 2.0–3.3 2.0 1.5 2.5 23.0–30.0 3.0 1.2 0.1 0.06
 Special purpose cast irons 107

The Cr–Fe oxides formed during melting attack silica lining materials, so
alumina linings should be used. Charges are based on steel scrap and foundry
returns with additions of high and low carbon ferrochromium to obtain
iron of the required carbon content. Ferromolybdenum is added as needed.
A loss of about 5% of the Cr content will occur during melting.
A melting temperature of 1600–1650°C is needed for the low carbon, high
Cr irons, but a lower temperature of 1550–1600°C is sufficient for the higher
carbon alloys.
A viscous oxide film forms on the surface of the molten metal in the
furnace and the ladle. The iron must be carefully skimmed before pouring,
and running systems must be designed to trap oxide dross. Metal filters
should be used if possible. To avoid oxide defects on the casting surface,
gating should provide rapid filling of the mould with minimum turbulence.
Chapter 9
Types of steel castings

Introduction
Steel is the most versatile of the structural engineering materials, so that
more grades of steel are available for casting than any other alloy type. The
casting of steel presents a number of special problems due mainly to the
high casting temperature, typically 1550–1600°C, and the volume shrinkage
(6.0–10.0%) that occurs on freezing. Steel castings range in size from a few
grams to hundreds of tonnes. Steel castings are often used in critical situations
where optimum mechanical properties are essential so that freedom from
casting defects is particularly important. The advantages of cast steel over
other cast metals, such as ductile iron, are:

High modulus – minimising deflection under heavy static loading

Shock and impact resistance – to cope with severe dynamic loading
Heat-treatable – to give the required toughness, hardness and strength
Weldable – so that castings can be joined to wrought steel products
High temperature strength – special alloys are available

Steel castings compete in many areas with steel weldments. In welding, it is

difficult to avoid stress concentrations at junctions while castings have well-
radiused joints so that fatigue resistance of cast steel junctions is superior to
that of welded junctions. Also welds, especially in large structures, are not
always amenable to stress-relieving after welding so that internal stresses
are likely to be greater in weldments than in steel castings. This may lead to
distortion of weldments or even failure in highly stressed components.
Castings allow freedom of design which is not possible with weldments.
The main applications of steel castings, roughly in order of tonnage used
in the UK, are:

Valves, pumps and compressors

Construction and earth moving equipment
Crushing, grinding, quarrying and dredging
Oil and gas rigs
Mining equipment
Motor vehicles
Defence equipment
Railways
 Types of steel castings 109

Shipbuilding
Iron and steelworks
Nuclear power plant
General engineering equipment
Electrical plant
Agricultural equipment

While the overall tonnage of steel castings has been on a downward trend
for a number of years, the overall value has increased due to a growing
trend towards increased quality standards, reflecting the high standards
needed for nuclear plant, gas and oil rigs etc.
Processes used for steel castings are:

Green sand for high volume castings

Chemically bonded sand where numbers required are smaller
Ceramic block moulds for castings requiring particularly good
surface finish
Lost wax investment casting for smaller castings made in large
numbers where good surface finish and
dimensional accuracy is needed

The Lost Foam process is not normally used for steel casting because of the
danger of carbon pick-up from the foam pattern.

Specifications for steel castings

Steel casting specifications in Europe are undergoing major changes as new
European specifications are being drawn up. At the time of writing, the
only European standard to be issued in the UK is BS EN 10213:1996 covering
Steel Castings for Pressure Purposes. During the next few years, EN standards
covering steels for General Engineering, Corrosion Resisting, Heat Resisting,
Wear Resisting, Structural and Centrifugal applications will replace the
existing national standards in European countries.
Until the new standards are issued, individual national standards apply.
However, attention should be paid to the International Standard ISO 4990-
1986 Steel Castings, General Technical Delivery Requirements which specifies
the methods of testing steel castings and other matters which need to be
agreed between purchaser and supplier. The contents of this standard are
being incorporated into the forthcoming European standard.
ISO 4990 requires the purchaser to supply details with the enquiry covering:

A description of the casting(s) by pattern number or drawing

Dimensional tolerances, machining allowances and datum points for
machining
The material standard and grade of steel
110 Foseco Ferrous Foundryman’s Handbook

The type of document required for inspection and testing

Additional information such as agreed manufacturing method
Size of test lot
Procedures for marking, machining, protection, packaging, loading,
dispatching and the destination
Submission of sample castings
Methods of statistical control
Inspection procedures

The standard lists details of inspection and testing to be carried out by the
manufacturer including:

Documents required
Sampling, preparation of test pieces and mechanical and chemical test
methods
Test blocks
Heat treatment
Repair welding
Non-destructive tests
Special tests such as intergranular corrosion and pressure tests
Test reports must provide the required traceability of the castings they
represent

It is normally required that the manufacturer shall maintain records of results

of all the chemical and mechanical tests performed by the foundry for a
minimum of 5 years.
Other ISO Standards detail testing methods:

ISO 148 Steel – Charpy impact test (V-notch).

ISO/R 783 Mechanical testing of steel at elevated temperatures –
Determination of lower yield stress and proof stress and proving
test.
ISO 2605/1 Steel products for pressure purposes – Derivation and veri-
fication of elevated temperature properties – Part 1: Yield or
proof stress of carbon and low alloy steel products.
ISO 2605/2 Steel products for pressure purposes – Derivation and veri-
fication of elevated temperature properties – Part 2: Proof stress
of austenitic steel products.
ISO 3452 Non-destructive testing – Penetrant inspection – General
principles.
ISO 3651/2 Austenitic stainless steels – Determination of resistance to
intergranular corrosion – Part 2: Corrosion test in a sulphuric
acid/copper sulphate medium in the presence of copper
turnings (Monypenny Strauss test).
ISO 5579 Nondestructive testing – Radiographic examination of metallic
materials by X- and gamma rays – Basic rules.
ISO 6506 Metallic materials – Hardness test – Brinell test
 Types of steel castings 111

ISO 6892 Metallic materials – Tensile testing.

ISO 8062 Castings – System of dimensional tolerances.

British Standards for steel castings

A number of British Standard Specifications cover steel castings. There are

equivalent standards in all industrialised countries, but the number of
specifications is so great that it is not possible to list them here. The British
Standards for steel castings include:

BS 1504 Steel Castings for Pressure Purposes

BS 1965 Pt 2 Butt Welding Pipe Fittings for Pressure Purposes (Austenitic
Stainless Steel)
BS 3100 Steel Castings for General Engineering Purposes
Section 2: Carbon and Carbon–Manganese Steels
Section 3: Low Alloy Steel Castings
Section 4: Corrosion Resisting, Heat Resisting and
High Alloy Steel Castings
BS 3146 Part 1 Investment Castings in Metal
Carbon and Low Alloy Steels
Corrosion and Heat Resisting Steels, Nickel and Cobalt
Base Alloys
Vacuum Melted Alloys
BS 2772 Iron and Steel for Colliery Haulage and Winding Equipment
1.5% Mn Steel Castings for Mine Car Couplers
BS 4473 Railway Rolling Stock Material
Cast Wheel Centres in Unalloyed Steel for Tyred
Wheels
BS 4534 Weldable Cr-Ni Centrifugally Cast Steel Tubes
BS Aerospace Series Specifications

Lists of equivalent standards are given in British and Foreign Specifications for
Steel Castings, published by the Castings Development Centre (formerly the
Steel Castings Research and Trade Association) East Bank Road, Sheffield
S2 3PT, 4th Ed. 1980. The large number of specifications may be judged
from the fact that the above book lists over 180 British specifications alone.

Types of steel castings

It is convenient to classify steels into three groups

Carbon steels
Low and medium alloy steels
High alloy steels
112 Foseco Ferrous Foundryman’s Handbook

Carbon steels
These are steels in which carbon is the main alloying element, other elements
such as manganese and silicon are present for the purposes of deoxidation.
The effectiveness of heat treatment in increasing the strength of the steel
depends on carbon content and carbon steels may be further subdivided
into

Low carbon steel <0.20%C

Medium carbon steel 0.20–0.50%C
High carbon steel >0.50%C

Applications of carbon steel castings include

Ship’s structural castings

Railway rolling stock
Automotive castings
Hot metal ladles
Rolling mill equipment
Rolls and rollers
Machines and tools
Mine and quarry equipment
Oil and petroleum equipment

Typical medium carbon steel specifications are shown in Table 9.1 (from BS
3100). There are equivalent specifications in other countries, for example:

UK BS 3100 A1 A2 A3
USA ASTM A27-77 N-1 A27-77 N-2 A148-73 gr. 80-40
Germany DIN 1681 GS-45 GS-52 GS-60

Carbon steels are most frequently used in the annealed, normalised or

annealed and normalised conditions, though other heat treatments may be
used to develop special properties, for example, for magnetic permeability.

Low alloy and medium alloy steels

Castings in this class normally contain amounts of manganese or silicon
greater than is needed for deoxidation (e.g. >1% Mn) or to which other
elements have been specially added. Alloy steels will contain Ni, Cr or Mo
as main alloying constituents and smaller amounts of V, Cu and possibly B.
These elements are added primarily to alter hardenability so that the required
properties can be obtained in different section thicknesses.
Pearlitic manganese steels (1.5%Mn), Table 9.1, are used where strengths
a little higher than attainable in plain carbon steels together with good
Table 9.1 BS 3100:1976 Steel Castings for General Engineering Purposes
Carbon and Carbon–Manganese Steel Castings

Carbon Steel Castings for General Purposes 11/2% Manganese Steel Castings for General Purposes
A1* A2* A3* A4 A5 A6
Chemical composition (%) min max min max min max min max min max min max

C – 0.25 – 0.35 – 0.45 0.18 0.25 0.25 0.33 0.25 0.33

Si – 0.60 – 0.60 – 0.60 – 0.60 – 0.60 – 0.60
Mn – 0.90(1) – 1.0 (1) – 1.0 (1) 1.2 1.6 1.2 1.6 1.2 1.6
P – 0.06 – 0.06 – 0.06 – 0.05 – 0.05 – 0.05
S – 0.06 – 0.06 – 0.06 – 0.05 – 0.05 – 0.05
Cr – 0.25 (2) – – – – – – – – – –
Mo – 0.15 (2) – – – – – – – – – –
Ni – 0.40 (2) – – – – – – – – – –
Cu – 0.30 (2) – – – – – – – – – –
Mechanical properties
TS (N/mm2) 430 – 490 – 540 – 540 690 620(5) 770(5) 690(6) 850(6)
0.2%PS (N/mm2) 230 – 260 – 295 – 320 – 370(5) – 495(6) –
Elongation (%) 22 – 18 – 14 – 16 – 13(5) – 13(6) –
Angle of bend (°) 120 (3) – 90 (3) – – – – – – – – –
Radius of bend (t)# 1.5 (3) – 1.5 (3) – – – – – – – – –
Charpy impact (J) 25 (3) – 20 (3) – 18 (3) – 30 – 25(5) – 25(6) –
Hardness (HB) – – – – – – 152 207 179(5) 229(5) 201(6) 255(6)

Notes: *If attached test samples are used, mechanical properties expected only where max. section thickness of casting is < 500 mm
# t = thickness of the test piece.
(1) For each 0.01%C below the max., an increase of 0.04%Mn will be permitted up to a max. of 1.10.
(2) Residual elements: the total shall not exceed 0.80%.
(3) Either a bend test or an impact test may be specified.
(4) Impact test mandatory only if specified by purchaser.
(5) Limiting section thickness, 100 mm.
(6) Limiting section thickness, 53 mm.
This table is intended only as a guide, refer to the British Standard for details.
114 Foseco Ferrous Foundryman’s Handbook

toughness are required. Applications are excavating buckets, travelling wheels,

gears, tractor track links and pneumatic tool parts.
Frequently used alloy steels are 11/2% Mn-Mo, 11/4% Cr-Mo, 0.75% Ni-
Cr-Mo, 11/2% Ni-Cr-Mo and 3% Cr-Mo (Table 9.2). Such higher strength
steels find use for gears, shaft pinions, impellers, rolls, rollers, crusher heads,
mill liners, crankshafts, cement processing machinery, fittings and components
in the oil industry.
Steel castings with a hard skin and a very tough core may be obtained by
local hardening (flame or induction hardening or carburising or nitriding).

Austenitic manganese steel

This widely used steel contains from 10–14%Mn and 1.0–1.4%C (typically
12%Mn, 1.25%C) (Table 9.3). The as-cast structure contains carbides and
other transformation products which produce brittleness. The castings are
therefore austenitised at 1000°C or above and quenched in water, producing
a fully austenitic structure that is quite tough. On account of their unique
ability to resist pounding and severe abrasive wear, Mn-steel castings are
used for railway trackwork, dredging, excavating, pulverising and crushing
equipment for mines and quarries, cement mills etc.

High alloy steels

These are used where corrosion resistance or heat resistance is required.
The alloys may be subdivided into three main groups:

Fe-Cr
Fe-Cr-Ni
Fe-Ni-Cr

The iron-chromium steels contain little or no nickel and 10–30%Cr. The Cr

content influences their resistance to scaling, the 13%Cr steels having
reasonable oxidation resistance up to 650°C while 28%Cr steel can be used
up to around 1100°C. The 13%Cr steels are used for pumps and steam
turbine equipment. 20–30%Cr steels are used in the chemical industry for
pump parts and valves, also for furnace parts, grates, kiln parts etc.
The iron-chromium-nickel alloys include the well-known 18/8 stainless
steels to which additions of Nb or Ti may be used to stabilise the austenite
on heating. Mo may also be added and Ni raised to 10% to allow the steel
to be used at higher temperatures (up to 1100°C) and improve the corrosion
resistance. High corrosion resistance and toughness require the lowest carbon
possible and special melting techniques allow figures as low as 0.03%C to
be achieved, although achieving such low carbon contents at the surface of
the casting is not easy since most moulding materials can cause surface
Table 9.2 BS 3100:1976 Steel Castings for General Engineering Purposes
Low Alloy Steel Castings

C–1/2 Mo 11/4Cr–Mo 21/4 Cr–Mo 3%Cr–Mo 11/4%Cr– Mo

B1 B2 B3 B4 BW4
Chemical composition (%) min max min max min max min max min max

C – 0.20 – 0.20 – 0.18 – 0.25 0.55 0.65

Si 0.20 0.60 – 0.60 – 0.60 – 0.75 – 0.75
Mn 0.50 1.00 0.50 0.80 0.40 0.70 0.30 0.70 0.50 1.00
P – 0.05 – 0.05 – 0.05 – 0.04 – 0.06
S – 0.05 – 0.05 – 0.05 – 0.04 – 0.06
Cr – 0.25 1.00 1.50 2.00 2.75 2.50 3.50 0.80 1.50
Mo 0.45 0.65 0.45 0.65 0.90 1.20 0.35 0.60 0.20 0.40
Ni – 0.40 – 0.40 – 0.40 – 0.40 – –
Cu – 0.30 – 0.30 – 0.30 – 0.30 – –
Mechanical properties
TS (N/mm2) 460 – 480 – 540 – 620 – – –
2
0.2%PS (N/mm ) 260 – 280 – 325 – 370 – – –
Elongation (%) 18 – 17 – 17 – 13 – – –
Angle of bend 120 – 120 – 120 – 120 – – –
Radius of bend (t) 1.5t – 1.5t – 3t – 3t – – –
Charpy Impact (J) 20 – 30 – 25 – 26 – – –
Hardness (HB) – – 140 212 156 235 179 255 341 –

Notes:
t = thickness of the test piece
This table is intended only as a guide, refer to the British Standard for details.
116 Foseco Ferrous Foundryman’s Handbook

Table 9.3 BS 3100:1976 Steel Castings for General Engineering

Purposes
Austenitic Manganese Steel Castings
BW10
Chemical composition (%) min max
C 1.00 1.25
Si – 1.00
Mn 11.00 –
P – 0.07
S – 0.06
Cr – –
Mo 0.45 –
Ni – –
Cu – –
The mechanical properties shall be agreed between the manufacturer
and the purchaser.
For special applications, max. carbon may be increased to 1.35%.
This table is intended only as a guide, refer to the British Standard
for details.

recarburisation due to the breakdown of carbon-containing binders. These

corrosion resistant steels are widely used for marine fittings, pump parts,
valve bodies, impellers etc.
One of the most widely used cast stainless steels is BS 3100: 316C16
(Table 9.4) an 18Cr/10Ni/2.5Mo cast equivalent of 316 wrought stainless
alloy (US equivalent, CF-8M).

Table 9.4 Cast stainless steel BS3100: 316C16

C Si Mn P S Cr Mo Ni

0.08 1.5 2.0 0.04 0.04 17–21 2.0–3.0 10 min

Mechanical properties
UTS (N/mm2) 480
0.1%PS (N/mm2) 240
Elongation (%) 26
Impact (J) 34

(Refer to BS3100 for details)

Such steels can be used at temperatures up to about 550°C making them

suitable for high pressure steam valves for nuclear and chemical applications
as well as for valves and pumps for handling hot acids and also marine
fittings. Unlike the wrought alloys, cast stainless steels contain from 5–40%
ferrite which provides strength, improves weldability and maximises
resistance to corrosion in specific environments.
Alloys for heat resisting applications contain 25%Cr and 12%Ni for use
up to 1050°C and 25%Cr–20%Ni for use up to 1100°C. They are used for
exhaust manifolds, burner nozzles, furnace parts, stack flues etc.
 Types of steel castings 117

The iron-nickel-chromium alloys are used for heat resisting applications.

Widely used alloys include one containing 17%Cr, 25%Ni, and one having
37%Ni, 15%Cr. The castings are used for furnace parts, salt and cyanide
pots, annealing trays etc.
Typical steel casting compositions of the above steel types are given in
Table 9.5.

Duplex steels
The severe conditions experienced in off-shore drilling platforms and other
marine applications require high strength corrosion resistant steels. Alloys
having a duplex ferrite-austenite microstructure have been developed for
this purpose (Table 9.6). High hardness is produced by the high copper
content (3%). The steel is heat treated by austenitising at around 1100°C
when the copper dissolves. Oil quenching retains the copper in solution,
subsequent ageing at around 480°C produces a ferrite-austenite structure
with precipitated copper increasing strength and hardness.
Only a few specifications are quoted here. Details of Steel Casting
Specifications world-wide are presented in British and Foreign Specifications
for Steel Castings, published by CDC, East Bank Road, Sheffield S2 3PT. The
large number of specifications may be judged from the fact that the above
book lists over 180 British specifications alone.
Further details of the properties of all the above steels may be obtained
from Steel Castings Design Properties and Applications, edited by W.J. Jackson,
published by CDC, East Bank Road, Sheffield S2 3PT.
The Steel Castings Handbook, sixth edition, by Steel Founders’ Society of
America and ASM International, 1995 contains full details of material selection,
mechanical properties, design and manufacture of castings etc.

Physical properties of steels

Table 9.7 gives typical physical properties of commonly used cast steels.
These figures should be regarded as approximate only, since the exact
composition and state of heat treatment affects the properties. Where
properties are quoted for a particular temperature, the figure is the average
value from room temperature to the quoted temperature.

Selection of suitable steel for casting

The choice and specification of a suitable steel for a specific application is
far from easy. Not only is it necessary to specify an alloy having suitable
properties: strength, hardness, fatigue properties, low temperature strength,
Table 9.5 BS 3100:1976 Steel Castings for General Engineering Purposes
Alloy Steel Castings

13%Cr 25–30%Cr 18/10CrNi 25/20CrNi

420C29 452C12 316C16 310C40
Chemical composition (%) min max min max min max min max

C – 0.20 1.00 2.00 – 0.08 0.30 0.50

Si – 1.00 – 2.00 – 1.50 – 1.50
Mn – 1.00 – 1.00 – 2.00 – 2.00
P – 0.04 – 0.06 – 0.04 – 0.04
S – 0.04 – 0.06 – 0.04 – 0.04
Cr 11.50 13.50 25.00 30.00 17.00 21.00 24.00 27.00
Mo – – – 1.50 2.00 3.00 – 1.50
Ni – 1.00 – 4.00 10.00 – 19.00 22.00
Cu – – – – – – – –
Mechanical Properties
TS (N/mm2) 690 – 480 – 480 – 620 –
2
0.2%PS (N/mm ) 465 – 280 – 240 – 370 –
Elongation (%) 11 – 17 – 26 – 13 –
Angle of bend – – 120 – – – 120 –
Radius of bend (t) – – 1.5t – – – 3t –
Charpy Impact (J) – – 30 – 34 – 26 –
Hardness (HB) 201 255 140 212 – – 179 255

This table is intended only as a guide, refer to the National Standards for details.
 Types of steel castings 119

Table 9.6 Duplex steel GX2CrNiMoCN25-6-3-3

C Si Mn P S Cr Mo Ni Cu N

0.03 1.0 1.50 0.035 0.025 24.5– 2.5– 5.0– 2.75– 0.12–
26.5 3.5 7.0 3.5 0.22
Mechanical properties:
UTS (N/mm2) 650–850
0.1%PS (N/mm2) 480
Elongation (%) 22
Impact (J) 50

Table 9.7 Physical properties of some steels

Composition Temp.°C Density Specific Thermal Thermal Electrical

(g/cm3) heat expansion conductivity resistivity
capacity coeff. (W/m.K) (micro-
(J/kg.K) (10–6/K) ohms.m2/m)

Carbon steels RT 7.86 59.5 13.2

100 482 12.19 57.8 19.0
200 523 12.99 53.2 26.3
400 595 13.91 45.6 45.8
600 741 14.68 36.8 73.4
800 960 14.79 28.5 108.1
1000 13.49 27.6 116.5
Low alloy RT 7.85 25.4
steels 100 456 12.45 48.2 30.6
200 477 13.2 45.6 39.1
400 532 14.15 39.4 60.0
600 599 14.8 33.9 88.5
Stainless RT 7.92 15.9 69.4
18/8 100 511 14.82 16.3 77.6
200 532 16.47 17.2 85.0
400 569 17.61 20.1 97.6
600 649 18.43 23.9 107.2
800 641 19.03 26.8 114.1
1000 28.1 119.6
13%Cr 100 7.75 482 11.0 24.3 57.0

wear resistance, corrosion resistance, high temperature strength and resistance

to oxidation etc. but it is usually necessary to define the inspection procedures
to be used. These will include

Visual examination of the casting surface to determine surface roughness

and freedom from cracks
Magnetic particle surface examination
Liquid penetrant surface examination
Radiographic inspection, usually in comparison to a system of reference
radiographs
120 Foseco Ferrous Foundryman’s Handbook

Ultrasonic testing, where heavy sections prohibit the use of radiography

Pressure or leak testing
Dimensional inspection
Statistical process control
Quality Assurance usually in accordance with a standard such as ISO
9002 or ISO 4990 Steel Castings – General Technical Delivery Requirements

The cost of minimum specification requirements is included in the basic

casting price, but additional testing will involve additional costs which
must be agreed between customer and supplier before castings are ordered.
Chapter 10
Melting and treatment of
steel for casting

Large steel foundries may use electric arc furnaces but induction furnaces
are the most commonly used melting furnaces for making steel castings.
Arc furnaces are capable of using low cost scrap charges, since refining
takes place in the furnace but arc furnaces have limitations since there is
always some pick-up of carbon from the graphite electrodes, so that very
low carbon stainless steels (<0.03%C) cannot be made. Refining is not possible
in the induction furnace, so a carefully selected charge must be used, but
any type of steel may be melted.

Arc furnace melting

The furnace consists of a refractory-lined mild steel shell, circular in section
and having a dished bottom. The refractory roof, usually water-cooled, has
three graphite electrodes projecting through it into the charge. The electrodes
can be raised and the roof lifted hydraulically and swung aside to allow the
furnace to be charged by a drop bottom charge bucket. The furnace is tilted
forward for tapping and backward for slag removal (Fig. 10.1).
Electrode arms above the roof supply three-phase power to the electrodes
and allow each electrode to be independently raised or lowered mechanically.
The length of the arc governs the power input to the furnace, regulators
measure the current in each electrode and control it automatically by
continuously adjusting the position of the electrode to maintain an optimum
arc length.
500–600 kWh of electricity is required to melt and raise to casting
temperature 1 tonne of steel. Furnaces are usually rated at 500 kVA per
tonne to give a melting time of about 11/2 hours.
The electrodes are made of graphite and are consumed during operation
through oxidation, volatilisation and breakage so they must be replaced as
necessary. A 3 tonne furnace typically uses electrodes of 200 mm diameter.
Electrode consumption is a significant factor in the cost of arc melting and
figures vary from 3–10 kg/tonne of steel melted depending on the type of
steel being produced and the practice used.
122 Foseco Ferrous Foundryman’s Handbook

Silica brick or
high alumina brick Silica brick
Unburnt metal encased Water cooled
magnesite-chrome brick arches & jambs

Magnesite brick

Tapping hole
sleeve Work
door

Top of
sill plate

Fireclay brick Silica brick

Magnesite brick Fireclay brick
Grain magnesite Ground silica ganister
Basic furnace lining Acid furnace lining

Figure 10.1 Schematic diagram of an arc furnace showing typical refractories

used in acid and basic practices. (From Jackson, W.J. and Hubbard, M.W., Steelmaking
for Steelfounders, 1979, SCRATA. Courtesy CDC.)

Refractories

Arc furnace refractories can be ‘acid’, ‘basic’ or ‘neutral’ depending on the

melting practice used. Basic linings of dolomite or magnesite are the most
commonly used in steel foundries. A safety lining of firebrick is installed in
the steel shell of the furnace, followed by an inner lining of magnesite or
dolomite bricks. The working hearth is then formed by ramming tarred
dolomite or magnesite. Side walls are built up from tarred dolomite or steel-
clad magnesite bricks. The lining is then dried and fired. Roofs are expected to
have a long, maintenance-free life and may be made of high alumina refractory.

Melting practice in the basic lined arc furnace

The use of basic slags, oxidising or reducing in nature, allows phosphorus

and sulphur to be reduced to low levels so that the basic lined furnace is
very versatile, allowing good quality steel to be made from poor quality scrap.
Melting in a basic arc furnace can be a one or two stage process using first
an oxidising slag then, if necessary, a reducing one. The charge, usually of
foundry returns and purchased steel scrap, is made up in a drop-bottom
charging bucket adding limestone and sometimes iron ore to the charge to
make the first oxidising slag which will remove phosphorus. Prior to the
 Melting and treatment of steel for casting 123

scrap being dropped into the furnace, some form of recarburiser such as
anthracite, coke or crushed electrodes is placed on the furnace bottom to
ensure sufficiently high carbon content at melt-out to give an adequate
‘boil’. When the charge is melted, the boil is induced by adding iron ore or
gaseous oxygen to the melt. The oxygen reacts with carbon in the melt
forming bubbles of carbon monoxide which rise through the metal giving
the appearance of boiling. The vigorous stirring action of the boil ensures
effective reaction of the metal with the slag and removes hydrogen and
nitrogen from the melt. The boil also removes carbon from the melt. The
carbon content of the melt is taken below the required finishing carbon and
the boil stopped by adding a deoxidant such as silicon and/or aluminium.
The slag is then removed, taking the phosphorus from the charge with it.
If sulphur removal is required, a second, reducing slag is formed by
adding limestone with fluorspar to keep it fluid and reducing agents such
as powdered ferrosilicon and calcium carbide. The chemistry of the melt is
adjusted by adding carburiser, Fe-Si, Fe-Mn together with Fe-Cr and Fe-Mo
if required. During time under the reducing slag hydrogen and nitrogen
pick-up can occur, so the reducing period should be as short as possible.
The slag is removed before pouring the steel into the ladle. The steel is
finally deoxidised in the ladle with aluminium.
There are many variations of the basic arc steelmaking process. The practice
used depends on the type of steel to be made and the quality of the scrap
steel used for the charge.

Induction furnace melting

The medium frequency coreless induction furnace (Fig. 4.5) is the most
flexible steel melting unit and is widely used in steel foundries. Small sized
scrap can be melted from cold and the furnace can be used for successive
melts of widely differing analyses. Melting is rapid since high power can be
applied. There is a strong stirring action in the molten steel due to the
induced current, so that a slag cover cannot be maintained. For this reason
there is no refining stage, the liquid steel having similar composition to the
charged scrap except for carbon, silicon and manganese which are lost by
oxidation and must be made up to the required levels by adding pig iron, or
other carburisers and ferroalloys. Alloying elements may also be added and
adjustments to temperature can be rapidly made. High alloy steels are
frequently made in the induction furnace since the melting losses of expensive
alloying elements are low. Extra-low carbon stainless steels can be melted
without pick-up of carbon.

Design of the furnace

The principles of design and operation of coreless induction furnaces are
described in Chapter 4, pp. 55–60.
124 Foseco Ferrous Foundryman’s Handbook

Induction furnaces for steel melting may have a melt capacity of up to

6 tonnes. Frequencies of 1000 Hz are usually used up to about 3 tonnes and
500 Hz for larger furnaces, being powered at up to 750 kW per tonne of
capacity. Such furnaces can be started from cold and have a charge-to-tap
time of about 1 hour.

Refractories

The linings for induction furnaces are nowadays usually made of magnesium-
alumina spinel, which can expand or contract on firing. The usual grade is
30% alumina, 70% magnesia which has an expansion sufficient to produce
metal tightness without damaging the coil.
Linings are made from refractory ramming mixes. The grading of the
refractory is important since it controls packing density and sintering. A
grading of around 45% coarse, 10% medium and 45% fines gives the maximum
packing density. The fines control the sintering of the bond which may be
formed by additions of boric acid or boric oxide, often incorporated in a
proprietary ramming mix. Before installing the lining, the coil and its insulation
should be checked for damage. The usual practice with medium frequency
furnaces is to coat the copper coil of the furnace with a layer of ‘mudding’
about 6 mm thick, of a medium to high alumina cement. This remains in
place when the hot face is knocked out. Between this and the hot face is a
layer of ceramic fibre insulation. The furnace bottom is first formed by
pouring in refractory to a depth of 50–60 mm which is rammed to give the
correct depth. The working face is formed by compacting the refractory
behind a steel former concentrically placed within the coil. Formers are
normally constructed from mild steel sheet according to the furnace
manufacturer’s design. Refractory is poured between the former and the
coil and compacted using vibratory ramming tools or manual compaction
adding more refractory as compaction proceeds. The top layer of the lining
is usually finished by mixing the granular lining refractory with sodium
silicate.
The former is left in the furnace and melted out with the first heat. Power
is slowly increased giving time for the bond to form before the former
melts.

Charge materials

The composition of steel melted in the induction furnace changes little during
melting so that careful selection of charge materials is necessary. Scrap should
be clean, often shot blasted to remove sand, and free from oil and moisture
which may cause hydrogen pick-up and fume. The scrap is charged together
with non-oxidisable alloys. When melted, the oxidisable alloys are added
and the melt brought to the required temperature then tapped as quickly as
 Melting and treatment of steel for casting 125

possible. Holding the melt at high temperature should be avoided to minimise

hydrogen pick-up which occurs quite rapidly above 1600°C.
The recycling of scrap in an induction furnace can lead to the pick-up of
gases since no refining takes place. Where possible, scrap from induction
furnaces should be returned to arc furnaces for refining.

AOD refining
AOD refining (argon oxygen decarburisation) has been developed to process
alloy steels for demanding applications. The metal is melted in a primary
arc or induction furnace then transferred to a special AOD vessel with a
controlled slag cover. Oxygen diluted with argon is then injected through a
tuyere at the bottom of the vessel. By reducing the partial pressure of oxygen,
carbon can be removed to low levels without oxidising the chromium and
other elements in the melt. Removal of S and P by slag/metal reaction is
aided by the vigorous stirring of the bath induced by the gas injection.
Steel with precise chemistry can be produced using AOD but at high cost
since the extra vessel is expensive and refractories must be replaced frequently.
It is not widely used in steel foundries outside the USA.

Melting and casting quality

Aspects of casting quality affected by melting and steelmaking practice
include

Deoxidation effects
Sulphides
Inclusions
Gas porosity
Hot tearing

Deoxidation

Liquid steel dissolves oxygen; as the steel cools, the dissolved oxygen combines
with carbon in the steel forming carbon monoxide:
[C] + [O] = CO (g)
If allowed to occur during solidification this would cause gas porosity in
the castings. The amount of dissolved oxygen in the liquid steel depends on
the carbon content of the melt, low carbon steel containing higher oxygen.
The dissolved oxygen is also affected by the silicon and manganese content
of the steel. To avoid gas evolution during solidification, the melt must be
deoxidised with an element which combines preferentially with the dissolved
126 Foseco Ferrous Foundryman’s Handbook

oxygen forming a stable oxide. Silicon, calcium silicide, titanium, zirconium

and aluminium are commonly used deoxidants with aluminium being the
most powerful. Ferrosilicon, calcium silicide or aluminium may be used as
a ‘block’ in the furnace before tapping but the use of aluminium for the final
deoxidation of liquid steel is practically universal.
The addition is usually made in the ladle. Aluminium is commonly used
in stick form for rapid solution. Additions vary from about 0.1% for medium
carbon steel to 0.2% for low carbon steel. Recovery is between 35 and 80%.
Mechanical feeding of aluminium wire is also used, precise control of the
addition is possible since the wire is fed rapidly deep into the melt giving
predictable yields. It is important to control the residual aluminium left in
the melt after deoxidation. Too little may allow the formation of harmful
forms of sulphide and intergranular AlN. Too much will harmfully affect
the steel properties. While aluminium is a powerful deoxidant, the products
of deoxidation are solid crystalline inclusions which are slow to separate
from the molten steel. They reduce the fluidity of the steel and have a
tendency to precipitate in the nozzle bore, reducing the rate at which steel
discharges from the ladle.
CaO fluxes alumina and controls the shape and size of the oxides allowing
them to float out more effectively in the casting ladle. It is difficult to introduce
sufficient calcium into the steel because when a lump of calcium alloy contacts
molten steel, most of the calcium boils off before it has a chance to enter the
steel where it can be effective. Calcium silicide cored wire may be added to
the ladle using a mechanical wire injector. In this way the calcium is introduced
deep into the steel where the ferrostatic head suppresses the rapid vaporisation
of calcium so that it is more effective than bulk additions. Removal of
inclusions is more effective if the steel is ‘rinsed’ with argon after the calcium
treatment through a porous diffuser in the ladle.

Sulphides

As the tensile strength of the steel increases, so the harmful effect of sulphur
increases. In carbon steel castings, 0.06%S may be tolerated but in aircraft
specifications sulphur is limited to 0.02%max. Careful selection of charge
materials is necessary to ensure low sulphur in induction melted steels.
Sulphur is soluble in liquid steel but on solidification it precipitates as MnS.
The sulphides formed may be globular, chain and film, or angular and are
known respectively as Types I, II or III. Type II sulphides are the most
harmful and must be avoided since they seriously affect the Charpy V-
notch values of the steel.
The form of the sulphides is related to the residual aluminium content of
the steel after deoxidation. The least harmful Type I globular sulphides are
formed when residual Al is very low, but this is difficult to control in practice.
The harmful Type II sulphides form at intermediate residual Al levels while
the less harmful Type III angular sulphides are formed at higher residual Al
 Melting and treatment of steel for casting 127

levels. The carbon content of the steel also affects the amount of residual Al
needed to ensure Type III sulphides. Low carbon steel requires higher residual
Al (Table 10.1).

Table 10.1 Residual Al% needed to ensure

formation of Type III sulphides in steels of
different carbon content

C (%) Residual Al(%)

0.1 0.04
0.2 0.03
0.3 0.02
0.4 0.02
0.5 0.02

From Jackson & Hubbard p. 235.

Steels with carbon above 0.5% should be deoxidised with 0.04% Al to ensure
100% type III sulphides. The aluminium addition must be increased as the
carbon decreases but the addition should not be too high or the impact
properties may suffer. Steels with carbon contents below 0.2% may
advantageously be deoxidised using Al plus calcium silicide.

Inclusions

Inclusions in steel castings can arise from slag entrapment or from the erosion
of furnace or ladle linings and refractories. They also arise from deoxidation,
being either direct deoxidation products or products arising from reaction
of the killed steel with refractories. For example manganese in killed steel
will reduce the silica in siliceous refractories, the MnO so formed has a
severe fluxing action on firebrick refractories.
Reoxidation of steel during teeming is also a source of inclusions. It has
been estimated that in each pouring or tapping operation, 1–4 kg of oxide
are produced for each tonne of steel transferred. Systems of shrouding the
metal stream with an inert gas have been developed but are not very effective.
The use of calcium silicide in conjunction with controlled additions of
aluminium and argon rinsing are considered effective in reducing inclusions
but can be expensive. Filtration of liquid steel using ceramic foam filters,
(see Chapter 18) is now widely used to reduce the presence of inclusions in
steel castings.

Gas porosity

Gas porosity in steel castings can be attributed to one or both of the following
reactions:
128 Foseco Ferrous Foundryman’s Handbook

[C] + [O] = CO (g) (1)

2[H] = H2(g) (2)
A reaction similar to (2) is possible for nitrogen.
To minimise porosity, the hydrogen and nitrogen contents of the liquid
steel must be minimised and the carbon/oxygen reaction prevented by a
sufficiently strong deoxidation practice.

Hydrogen

The hydrogen level at which gas holes occur varies according to deoxidation
practice and steel composition. A casting can be made sound as follows:

up to 9 ppm of hydrogen for silicon killed steel castings

9–13 ppm hydrogen for Si and Al killed steel castings
over 13 ppm hydrogen, impossible to obtain solid castings

Hydrogen in liquid steel arises from contact with moisture, usually from
damp refractories and additives. Excessive bath temperatures and prolonged
refining periods should be avoided. All late additions should be dry and
ladles should be thoroughly dried. High atmospheric humidity is a factor
since it may lead to moisture pick-up from refractories (and later from the
sand moulds). A new furnace lining may cause excessive hydrogen pick-up
and it is good practice to melt a ‘wash heat’ before the first charge proper.

Nitrogen

The main source of nitrogen in steel is the atmosphere and the nitrogen
content of the steel depends to a large extent on the length of time that the
bath is in contact with the atmosphere. While nitrogen alone can contribute
to porosity, its normal level is far below the limit of solid solubility. In
0.10%C plain carbon steel castings, when nitrogen is at the 0.013% level,
there is a distinct tendency towards porosity. Nitrogen together with hydrogen
can sometimes be the cause of porosity in high alloy steels, particularly
those containing large amounts of chromium, if high nitrogen ferrochromium
has been used.
Nitrogen pinholes occur in castings made in moulds made using nitrogen-
containing resins such as UF-furanes and phenolic urethanes, and it is
important to use low nitrogen binders, particularly when reclaimed sand is
used (see Chapter 13).
In addition to the possibility of causing porosity, nitrogen in combination
with aluminium may influence the ductility of the steel due to intergranular
fracture caused by AlN precipitating at grain boundaries as the steel cools
through the temperature range 800–900°C. In commercial steel castings, the
minimum amounts of nitrogen combined as aluminium nitride that can
give rise to intergranular fracture are 0.002% for low alloy steels and 0.004%
for plain carbon steels. It is not generally practical to reduce nitrogen to
 Melting and treatment of steel for casting 129

such a low level. However, if there is a large enough excess of aluminium in

the steel after deoxidation, AlN can form in the melt and act as nuclei for
precipitation in the solid state. The large AlN particles so formed do not
give rise to intergranular fracture.

Carbon monoxide

The more thorough the deoxidation of the steel, the less the pinholing tendency,
and proper deoxidation with aluminium is the most effective control. There
must be an excess of deoxidant to compensate for the higher oxygen content
that develops in the surface layer of a casting as it comes in contact with the
mould wall. It is possible for silicon and aluminium to be removed locally
which may result in pinhole porosity.

Hot tearing

Hot tearing is a common foundry problem caused by the contraction of the

steel on freezing being hindered by the mould or cores. Chemical composition
of the steel affects hot tearing in a minor way, sulphur having the greatest
effect. Lowering sulphur to below 0.02% can be beneficial in carbon and low
alloy steels, and a manganese content of 0.7–1.3%Mn ensures that any sulphur
present precipitates as MnS rather than the Fe-FeS eutectic which increases
the tendency to hot tear. In high alloy steels, austenitic 18/8 steels do not
tear as severely as plain carbon or 13%Cr steels. As in the case of carbon
steels, sulphur has the greatest adverse effect and manganese improves tear
resistance.
Since hot tearing is caused by mechanical restraint preventing metal
shrinkage occurring, casting design changes, such as use of adequate fillets
at junctions, can be of great help. The minimum binder should be used in
cores and moulds for any casting susceptible to the problem.
The above description of the melting and treatment of steel for casting is
far from complete and for further details the book Steelmaking for Steelfounders,
by W.J. Jackson and M.W. Hubbard, published in 1979 by the Steel Castings
Research and Trade Association, Sheffield is recommended. Thanks are due
to the Castings Development Centre, 7 East Bank Road, Sheffield for
permission to use data from this book and from Steel Castings Design Properties
and Applications referred to above.
Chapter 11
Molten metal handling

Iron foundries

Metal handling systems

The aim of the metal handling system is to deliver clean liquid metal to the
mould at the required temperature. Automatic pouring systems are used
increasingly, but manual pouring is still the commonest method of filling
moulds. Iron foundries almost always use lip-pouring ladles although bottom-
pouring may be used for very large castings such as ingot moulds.
For efficient energy use, molten metal tapping temperature should be as
near as practicable to the desired pouring temperature, having regard to
unavoidable heat losses. In general this is achieved by:

Use of the largest ladle size practicable for pouring moulds.

Ladles should always be covered.
Use of transfer ladles should be avoided, if possible.
Avoid intermittent use of pouring ladles.
Use efficient ladle-heating to prepare ladles for use and when unavoidable
delays occur.
Bucket type ladles with insulated covers lose less heat than drum-type
ladles.

Each transfer of liquid iron from one ladle to another results in a temperature
loss of about 10–20°C.

Ladle lining
The ladle lining in an iron foundry is normally expected to last for a full
shift, in which time it may be used between 50 and 100 times. The lining
material must therefore be of good quality and be applied carefully.
Traditionally, iron foundry ladles have been lined with ganister (crushed
silica rock), naturally bonded sand or firebrick or a combination of these
materials.
Ganisters are usually pre-mixed with a clay binder and are applied with
the minimum amount of water, using a removable former of metal or wood.
 Molten metal handling 131

If ladles are used daily on high production plants, the old lining should be
removed completely from the ladle body, which should then be coated
lightly with a ganister wash and the new lining material rammed into position
round the former. The material must be properly vented, by hanging rods
between the ladle shell and the lining. These are removed before drying to
allow easy passage of steam from the lining.
Good quality naturally bonded sand may be used as an alternative to
ganister but the bottom of the ladle should preferably be either brick or
ganister lined to withstand the erosion during filling. The lip and rim of the
ladles are also best lined with ganister. Again, venting is necessary to assist
drying.
Larger ladles are lined with silica firebrick faced with a layer of ganister.
CO2/silicate sand linings may also be used, they are applied using a
former and cured by gassing. Because the free moisture is low, less drying
is needed before use.
The greater the capacity of the ladle, the thicker should be the lining.
Typical thicknesses are:

Ladle capacity (kg) Thickness (mm)

500 25–50
500–5000 50–75
over 5000 75–150

Linings can be damaged by incorrect drying. Properly designed ladle heaters

are better than simple gas-torches. The initial drying must be slow, to avoid
cracking, the gas flame can be increased in intensity as the lining heats up.
When the ladle is completely dry, it is ready for pre-heating. The lining
temperature should be raised to about 900–1000°C before filling with molten
metal. Heat losses are greatest when the ladle is filled for the first time after
pre-heating or if it is only used intermittently, since heat from the metal is
used in heating the refractories to temperature. If ladles are used continuously,
temperature losses stabilise to a predictable value as heat is lost by radiation
from the metal surface and by conduction through the ladle.
The actual cooling rate for any ladle depends on factors such as:

Capacity; larger ladles cool more slowly

Shape; narrow, deep ladles lose less heat by radiation from the metal
surface
How full the ladle is; the fuller, the less the heat loss
The use of covers; a properly fitted cover can halve the cooling rate.

Typical temperature losses from open bucket ladles, well-lined and full are:

Capacity 25 kg 100 kg 500 kg 30 tonnes

Temp. loss 30°C/min 10°C/min 6°C/min <2°C/min
132 Foseco Ferrous Foundryman’s Handbook

These figures are only a rough guide, for greater accuracy, actual
measurements should be made on the ladles in use. The use of a cover will
approximately halve temperature losses.
Heat losses can be reduced by covering the surface of the metal with an
insulating layer (note that it is not desirable to leave slag or dross on the
surface because of the possibility of metal/slag reaction). RADEX 460L
refractory insulator has been developed for this purpose, it is insulating
and inert, having low carbon content and is virtually aluminium free. The
metal should be skimmed clean and a layer of RADEX 460L scattered over
the surface to give a layer about 10 mm thick, or roughly 1.5 kg/tonne of
iron, though this depends on the ladle shape.

KALTEK ladle lining system for iron foundries

The KALTEK system uses a thermally insulating material as a liner to reduce
the time taken to prepare foundry ladles and to provide consistent pouring
temperatures (Fig. 11.1). Pre-heating of ladles is eliminated and installation
is quick and easy. One-piece moulded KALTEK liners are available for many
sizes of iron ladles. The lining is designed to fit accurately into the ladle
shell. A permanent safety lining may sometimes be used as well. A coarse
grained sand (10–20 AFS) is recommended to back-fill the lining to ensure
adequate venting as the organic binders in the liner system burn out. A
sodium silicate bonded core sand, adequately vented, can be used as a sand
cap.
When molten metal is first poured into a cold ladle lined with KALTEK,
heat is absorbed by the lining, which initially causes the metal to cool. On
successive fills, the lining quickly reaches steady state and the cooling becomes
less. Cooling of the metal in the ladle is less than in conventionally lined
ladles because of the insulating properties of the KALTEK material. When
used for iron casting, the life of the lining depends on the metal grade, slag
type and pouring temperature but is usually several hundred fills, allowing
multi-shift use. After use, the one piece lining is easily removed by breaking
the sand cap and overturning the ladle. Cooling is usually unnecessary and
the ladle can be relined immediately. To complement the KALTEK system,
a one-piece insulating lid made from PROCAL can be used to reduce
temperature losses even further.
Two-piece KALTEK ladle liners are available for sizes larger than can be
moulded as a single liner. For even larger ladles, linings can be formed from
KALTEK boards, supplied in the form of segments and bottom boards (Fig.
11.2). The ladle must be furnished with a permanent base lining of alumina
or firebrick to act as a safety lining. The KALTEK segments and base boards
are fitted inside the permanent lining, joints being filled with KALSEAL
refractory cement. The gap between the KALTEK and the permanent lining
is filled with coarse, dry silica sand, 15 AFS is recommended, or BAKFIL.
Multiple filling is possible as long as the ladle lining is not allowed to cool
down between fillings.
 Molten metal handling 133

(a) (a)

(b) (c)

(d) (d)

(e) (e)

Figure 11.1 Installation of the KALTEK one-piece ladle liner

(a) Coarse grade sand (or BAKFIL) is placed in the base of the ladle
(b) The KALTEK liner is placed in position.
(c) The gap between the KALTEK lining and the shell is filled with coarse
(10–20 AFS) sand
(d) A rammable material such as sodium silicate bonded sand, adequately vented,
seals the top
(e) The FOSCAST refractory dam board is fitted and sealed
134 Foseco Ferrous Foundryman’s Handbook

Figure 11.2 Panel ladle made using KALTEK boards.

Comparison of KALTEK cold ladle lining system with

conventional ladle lining

A ductile iron foundry making automotive castings in the weight range 1 kg

to 40 kg (average 5 kg) used pouring teapot ladles of 1100 kg capacity lined
with high quality refractory concrete poured around an internal former,
dried and fired with gas burners. Ladles were cleaned and repaired daily,
each ladle had a lifetime of about 500 tonnes of liquid iron. The pouring
practice was
 Molten metal handling 135

Time t = 0 Tap 1100 kg of liquid iron from the furnace to

the treatment ladle
Time t = 2 minutes Transfer to the pouring ladle with inoculation
into the metal stream
Time t = 5 minutes Begin pouring
Time t = 16 minutes End pouring and coat the ladle lip with graphite
wash
Back to t = 0 Repeat cycle, two ladles are in use simultaneously

The cycle remained the same when the foundry changed from using concrete
liners to using the KALTEK cold disposable liner system. For trials, a panel
ladle was made using KALTEK boards (Fig. 11.2). This was later replaced
by a shaped KALTEK two-piece lining (Fig. 11.3) with capacity 1100 kg. The
new ladle lining consisted of three elements: a bottom bowl, a top ring and
a concrete dam plate. Installation of the lining involved:

The bottom bowl was laid on a coarse sand bed

A refractory glue joint was deposited on the horizontal joint
The top ring was installed on the glue joint
The gap between the KALTEK lining and the shell was filled with coarse
(15 AFS) sand
The top of the sand was sealed with a rammable material to prevent sand
leakage when the ladle was tilted. The rammed material was vented by
1 mm holes, 50 mm apart

Figure 11.3 1100 kg capacity teapot KALTEK ladle.

136 Foseco Ferrous Foundryman’s Handbook

After several years of experience with the KALTEK ladles a comparison of

the two lining processes showed:

Furnace temperature reduced by 20°C with the same average pouring

temperature
Less wear of the furnace refractory lining
Lower variance of pouring temperature
Lower ladle shell external temperature, better working environment
Lower direct labour costs on lining operations
Less gas consumption for preheat
Savings in FeSiMg alloy additions because of the lower tap temperature
Significant reduction in non-metallic inclusions
Better process control

KALTEK insulating lining system for

automatic pouring boxes
Many automatic pouring systems, such as the DISAMATIC system, use
refractory lined pouring boxes using a clay-graphite stopper rod and graphite
nozzle. The box has a capacity of 1800 kg with an average fill of 600–900 kg.
The conventional pouring box is lined with a 70% alumina castable having
a life of about 25 shifts with a certain amount of patching needed daily. The
nozzle requires maintenance every 1–11/2 hours to remove slag and metal
build-up.
The KALTEK insulating board system consists of three components:

A back-up refractory safety liner in the metal shell of the pouring box
KALTEK boards custom-designed to fit the safety liner
Coarse sand to separate the KALTEK boards from the safety liner

The stopper rod, nozzle and pouring procedure remain the same.
The back-up 50% alumina liner is cast into the metal shell, since it has no
direct contact with the molten metal, it lasts for years without maintenance.
The KALTEK boards are supplied as custom-fitted sides, ends and bottom
(Fig. 11.4). Metal banding is used to secure the assembled board set until
placement in the pouring box. Before the boards are placed into the box, the
graphite nozzle is set into the nozzle well and secured with a graphitic,
thermally setting, ramming mix.
25 mm of coarse sand is spread evenly across the box floor and the pre-
assembled KALTEK set lowered into the box (Fig. 11.5). More sand is poured
into the cavity between side boards and safety liner. A refractory lid is
attached and any exposed sand capped with a thin layer of plastic ram,
vented with small holes to vent gases arising from organic binders in the
boards. Finally the nozzle is finished according to standard practice.
The lining requires no heating before use, a small gas torch is used to
 Molten metal handling 137

Figure 11.4 Custom fitted KALTEK boards for DISAMATIC automatic pouring system.

Figure 11.5 Lowering the pre-assembled KALTEK boards into the DISAMATIC
automatic pouring box.

preheat the stopper rod and nozzle prior to the first cast. Pouring and
casting are carried out as standard.
Benefits of the KALTEK lining system, in a ductile iron foundry, are:

30°C lower holding temperature

Alloy usage reduced because of lower holding temperature
Rate of heat loss one-third as fast
Refractory costs lowered
Labour costs lowered
Reduction of slag defects, misruns and shrinkage defects
Gas costs drastically lowered
Zero nozzle maintenance

Casting defects due to poor ladle maintenance

Slag and dross inclusions

These arise from slag carried over from the melting furnaces, from slag
deposits left in the ladles after pouring and from fusion of the ladle lining
material. They can be reduced by:
138 Foseco Ferrous Foundryman’s Handbook

removal of adhering slag before use;

with tea-pot ladles, molten slag should be removed by back-tilting, to
avoid contamination of the spout;
use of good quality ganister and other lining materials.

Hydrogen pinholing

Caused by molten metal being poured from improperly dried ladles or

damp furnace launders. Hydrogen from the moisture is dissolved in the
liquid metal and released during solidification. The problem is increased by
trace amounts of aluminium in the metal. The defect is avoided by correct
ladle drying and pre-heating.
The use of the KALTEK ladle lining system avoids the above problems.

Adding alloy additions in the ladle

Silicon and molybdenum additions can be made to cast iron by adding

LADELLOY fluxed additives.
SILICON LADELLOY 33 is a fluxed ladle addition allowing the silicon
content of the iron to be increased quickly and accurately. The amount of
silicon to be added is determined by a wedge/chill test. SILICON LADELLOY
33 contains about 62% Si with a flux to ensure optimum pick-up. The required
quantity of additive is slowly poured into the metal stream as it passes
down the launder to the ladle. The use of ladle additions simplifies melting
operations since a standard charge can be adjusted at the ladle to cover a
wide variety of requirements.
MOLYBDENUM LADELLOY 33 contains about 66%Mo and a small
amount of flux to ensure optimum pick-up of the element. The required
number of units should be placed in the bottom of the ladle and metal
tapped onto them. The phosphorus content of the iron must be less than
0.2% to avoid the formation of molybdenum phosphide.

Molten metal handling in steel foundries

Ladle practice
There are three types of ladle used for steel casting: lip pour, teapot and
bottom pour.

Lip pour ladles

In this type of ladle the metal is discharged over the lip, flow is controlled
 Molten metal handling 139

by tilting the ladle using a geared handwheel (Fig. 11.6a). Since the metal
flows from the top of the ladle, the metal surface must be slag-free or a
skimmer must be used to prevent slag entering the mould. The advantages
of lip pouring ladles are that there are no narrow passages in which the
metal can freeze so they are ideal for handling small quantities of metal.
They are used for pouring small steel castings. They are inexpensive easy to
prepare and are immediately ready for use. The disadvantage is that slag
can easily be entrained in the metal stream.

Lifting bail

Safety catches
in open position
pouring stream

Liquid metal

Refractory lining

Mild steel shell

Figure 11.6a Section through a lip pour ladle. (From Jackson, W.J. and Hubbard,
M.W., Steelmaking for Steelfounders, 1979, SCRATA. Courtesy CDC.)

Teapot ladles

A refractory dam before the ladle lip ensures that metal is drawn from the
bottom of the ladle so that the stream is slag free (Fig. 11.6b). Deoxidation
products have more time to float away from the area where metal is being
140 Foseco Ferrous Foundryman’s Handbook

withdrawn so the steel is generally cleaner than from a lip pour ladle. The
disadvantage is that the narrow ‘spout’ may occasionally permit the liquid
steel to freeze if the heat is tapped cold or pouring is prolonged.

Lifting bail

Safety catches
in open position

Pouring stream

Slag

Liquid metal

Mild steel
shell

Refractory lining

Figure 11.6b Section through a teapot ladle. (From Jackson, W.J. and Hubbard,
M.W., Steelmaking for Steelfounders, 1979, SCRATA. Courtesy CDC.)

With both lip pour and teapot ladles, it is only necessary to invert the
ladle to remove all slag and metal before refilling or reheating.

Bottom pour ladles

The ladle is fitted with a pouring nozzle in its base, closed by a refractory
stopper rod (Fig. 11.6c). The metal is drawn from the bottom and is therefore
slag-free and non-metallics such as deoxidation products are able to float
out of the melt. The metal stream flows vertically downwards from the
ladle so that there is no movement of the stream during pouring. The
 Molten metal handling 141

disadvantage is that the velocity and rate of flow change during pouring as
the ferrostatic head changes.

Slag

Liquid metal
Slide

Refractory lining
Pouring handle

Mild steel shell

Stopper rod
Stopper rod covers
Stopper rod end

Pouring nozzle

Figure 11.6c Section through a bottom pour ladle. (From Jackson, W.J. and Hubbard,
M.W., Steelmaking for Steelfounders, 1979, SCRATA. Courtesy CDC.)

Unless a reusable system is fitted, the nozzle and stopper rod assembly
must be changed after each use, thereby increasing the turn-round time,
costs and the number of ladles required to handle a given output of metal.
The stopper rod may also distort or erode making it impossible to shut off
the stream completely. It is not practical to handle less than 100 kg of steel
due to the chilling effect of the stopper rod assembly. The flow from a
bottom pour ladle depends on the size of the nozzle and the height of metal
in the ladle so the flow rate and velocity of the metal stream reduces as the
ladle empties. The nozzle/stopper rod is an excellent on-off valve but is not
an effective flow control valve. Attempts to use it to control flow result in
breakup of the metal stream with consequent risk of reoxidation of the steel.
However it is possible to calculate the discharge rate and metal velocity for
each ladle and nozzle and its variation throughout the pour so that ladles
can be suited to the mould sizes that are being cast (see Chapter 18).
142 Foseco Ferrous Foundryman’s Handbook

Ladle linings

The ideal lining is highly refractory, non-reactive with metal or slag and of
low thermal conductivity and heat capacity. Fireclay is the traditionally
used material either in the form of pre-fired bricks or as a plastic ramming
grade. Brick linings must be installed by skilled personnel to ensure tight
joints. Monolithic refractories are easier to install and eliminate the weak
spots associated with bricks.
Although fireclay is the lowest cost refractory, it is possible to raise the
level of as-cast quality considerably by using better lining materials. High
alumina monolithic linings are popular because of their better refractoriness
and their longer life.
Whether bricks, rammed monolithic refractories or castables are used,
there is a long preparation and drying time needed. New ladle linings must
be dried before use to remove moisture. After preliminary air drying, gas
heaters are used to heat the lining to 600–800°C. Final firing of the lining
occurs during its first use.
Steel rapidly loses temperature when held in ladles. A 5-tonne bottom
pouring ladle lined with fireclay will lose 20°C in 5 minutes, while the same
ladle lined with high alumina refractory will lose as much as 50°C in 5
minutes.

The KALTEK ladle lining system

The KALTEK range of products are disposable, insulating, refractory ladle
linings for steel, iron and non-ferrous applications. The low thermal mass
and insulating properties of KALTEK linings eliminate the need for pre-
heat in virtually all ladle sizes and with most alloys. KALTEK inner linings
can be quickly stripped and replaced when necessary.
KALTEK one-piece ladle linings are used for lip pouring in a wide range
of ladle sizes. A FOSCAST castable alumina dam board can be integrated
with the KALTEK liner to create a teapot ladle. Larger ladles, up to 15
tonnes, are lined with KALTEK in the form of board segments and bottom
boards designed to suit individual ladles. KALTEK boards and linings are
supplied in silica based refractory and also alumino-silicate for severe
applications and high magnesia for special alloys.
The ladle to be lined with a pre-formed one piece lining must be free
from holes. The KALTEK one-piece lining is fitted inside the shell with a
minimum of 15 mm backfill of coarse, dry silica sand or BAKFIL, a coarse
graded aggregate (AFS 35). The top exposed ring between lining and shell
should be capped with a suitable mixture such as silicate bonded core sand
and adequately vented with holes every 10–12 cm formed using 6 mm wire
(Fig. 11.1).
Larger ladles to be lined with KALTEK boards must first be stripped and
then furnished with a permanent base lining of alumino silicate bricks or
 Molten metal handling 143

castable refractory to act as a safety lining. The permanent lining must be

dried and fired carefully. Vent holes in the ladles must be cleaned or new
holes bored where necessary of 6–8 mm diameter at 20 cm centres.
The KALTEK segments and base boards are then carefully fitted inside
the permanent lining, the joints between them being filled with KALSEAL
refractory cement. The gap between the refractory and backing brick safety
lining should be filled with BAKFIL or coarse, dry silica sand (Fig. 11.2).
Refractory nozzle assemblies can be fitted into bottom pouring KALTEK
lined ladles using special KALPACK ramming material (Fig. 11.7). The nozzle
must be pre-heated to dry the KALPACK product before use.
The four methods available for a bottom pour ladle are shown here:

(1) Normal ladle (2) Normal ladle

nozzle is flush with nozzle raised using
permanent lining packing pieces

(3) Modified ladle (4) Ladle with nozzle

nozzle is raised so well-block as (3) but KALTEK
that its top is flush is butted to well-block
with KALTEK lining rather than nozzle

Permanent Backfill Nozzle KALPACK

lining
Various configurations in bottom pour ladles

Figure 11.7 Methods of installing nozzles in bottom pour ladles.

KALTEK ladle linings are used under cold start conditions, pre-heating
is unnecessary and would destroy the binder components of the boards.
The thermal capacity of a KALTEK lined ladle is lower than conventional
refractories and the thermal insulation twice as good so that while there is
some initial chill when steel is tapped into the cold ladle, the superior
insulation properties soon compensate so that lower tapping temperatures
are possible (Fig. 11.8).
Initially, the concept of the KALTEK cold ladle lining system for steel
foundries was based on single use of the disposable lining, but now in most
cases, multiple life is possible. Pot ladles can be used many times as long as
they are not allowed to cool between fillings. In many cases, bottom pour
ladles can be used more than once as long as multi-life stopper rods such as
the Roto-rod isostatically pressed one-piece alumina-graphite rod are used.
With KALTEK there is:
144 Foseco Ferrous Foundryman’s Handbook

Heat loss comparison

5 ton bottom pour ladle
(steel tapped from AOD vessel)
1700

Tapped at 1693°C

1650 Tapped at 1650°C

Temperature °C

1600

KALTEK
No preheat
1550 70% AI2O2
Brick-Preheated
to 800°C
0 5 10 15 20
Time (minutes)

Figure 11.8 Heat loss comparison, 5 ton bottom pour ladle, KALTEK v. brick-lined.

No pre-heat
Faster ladle turnaround
Easier ladle maintenance
Better temperature control
Better working environment
Lower inclusion levels

Pouring temperature for steels

The temperature at which steel castings are poured is at least 50°C above
the liquidus temperature. Further superheat is needed to allow for cooling

Table 11.1 Variation of liquidus temperature with carbon content for Fe-C alloys

Carbon Liquidus temperature Carbon Liquidus temperature

(%) (°C) (%) (°C)

0.05 1533 0.55 1490

0.10 1528 0.60 1486
0.15 1524 0.65 1483
0.20 1520 0.70 1480
0.25 1515 0.75 1477
0.30 1511 0.80 1473
0.35 1507 0.85 1470
0.40 1502 0.90 1466
0.45 1498 0.95 1463
0.50 1494 1.00 1459
 Molten metal handling 145

Table 11.2 Depression of liquidus temperature caused by the presence of 0.01% of

alloying elements

Element Depression (°C) Element Depression (°C)

P 0.300 Mo 0.020
S 0.250 Si 0.080
Mn 0.050 Cu 0.050
Cr 0.015 Sn 0.080
Ni 0.040 V 0.030

Example: a steel containing

0.06C, 1.0Si, 1.2Mn, 0.03P, 0.02S, 18Cr, 2.0Mo, 10.5Ni would have a liquidus temperature of
1532 – (8.0 + 6.0 + 0.90 + 0.50 + 27 + 4 + 42) = 1532 – 88.4 = 1444°C
The pouring temperature should be at least 1444 + 50 = 1494° say 1500°C

in the ladle during casting, which can be as high as 10°C per minute in high
alumina lined ladles though much lower for KALTEK lined ladles.
The liquidus temperature of a steel can be estimated from Tables 11.1 and
11.2.
Chapter 12
Sands and green sand

Silica sand
Most sand moulds and cores are based on silica sand since it is the most
readily available and lowest cost moulding material. Other sands are used
for special applications where higher refractoriness, higher thermal
conductivity or lower thermal expansion is needed.

Properties of silica sand for foundry use

Chemical purity

SiO2 95–96% minimum The higher the silica the more

refractory the sand
Loss on ignition 0.5% max Represents organic impurities
Fe2O3 0.3% max Iron oxide reduces the refractoriness
CaO 0.2% max Raises the acid demand value
K2O, Na2O 0.5% max Reduces refractoriness
Acid demand value 6 ml max High acid demand adversely
to pH4 affects acid catalysed binders

Size distribution

The size distribution of the sand affects the quality of the castings. Coarse
grained sands allow metal penetration into moulds and cores giving poor
surface finish to the castings. Fine grained sands yield better surface finish
but need higher binder content and the low permeability may cause gas
defects in castings. Most foundry sands fall within the following size range:

Grain fineness number 50–60 AFS  Yields good surface finish at


Average grain size 220–250 microns  low binder levels
Fines content, below 2% max Allows low binder level to be
200 mesh used
Clay content, below 0.5% max Allows low binder levels
20 microns
Size spread 95% on 4 or 5 screens Gives good packing and
resistance to expansion defects
Specific surface area 120–140 cm2/g Allows low binder levels
Dry permeability 100–150 reduces gas defects
 Sands and green sand 147

Grain shape

Grain shape is defined in terms of angularity and sphericity. Sand grains

vary from well rounded to rounded, sub-rounded, sub-angular, angular
and very angular. Within each angularity band, grains may have high, medium
or low sphericity. The angularity of sand is estimated by visual examination
with a low power microscope and comparing with published charts
(Fig. 12.1).

High
sphericity

Medium
sphericity

Low
sphericity

Very angular Angular Sub-angular Sub-rounded Rounded Well rounded

Figure 12.1 Classification of grain shapes.

The best foundry sands have grains which are rounded with medium to
high sphericity giving good flowability and permeability with high strength
at low binder additions. More angular and lower sphericity sands require
higher binder additions, have lower packing density and poorer flowability.

Acid demand

The chemical composition of the sand affects the acid demand value which
has an important effect on the catalyst requirements of cold-setting acid-
catalysed binders. Sands containing alkaline minerals and particularly
significant amounts of sea-shell, will absorb acid catalyst. Sands with acid
demand values greater than about 6 ml require high acid catalyst levels,
sands with acid demand greater than 10–15 ml are not suitable for acid
catalysed binder systems.

Typical silica foundry sand properties

Chelford 60 Sand (a sand commonly used in the UK as a base for green sand
and for resin bonded moulds and cores)
148 Foseco Ferrous Foundryman’s Handbook

Grain shape: rounded, medium sphericity

Bulk density, loose: 1490 kg/m3 (93 lb/ft3)
GF specific surface area: 140 cm2/g
Angle of repose: 33°

Chemical analysis:

SiO2 Fe2O3 Al2O3 K2 O Na2O CaO TiO2 Cr2O3 LOI

97.1 0.11 1.60 0.73 0.15 0.10 0.06 15 ppm 0.3

Acid demand (number of ml 0.1N HCl):

to: pH3 pH4 pH5 pH6 pH7

ml: 2.0 1.8 1.4 1.0 0.8

Sieve grading of Chelford 60 sand

Aperture size (µm) BSS mesh No. % wt retained

1000 16 nil
700 22 0.4
500 30 2.3
355 44 10.0
250 60 25.7
210 72 23.8
150 100 28.7
105 150 7.6
75 200 1.3
–75 –200 0.2

AFS Grain Fineness No. 59

Base permeability: 106

Table 12.1 gives size gradings of typical foundry sands used in the UK and
Germany.

Safe handling of silica sand

Fine silica sand (below 5 microns) can give rise to respiratory troubles.
Modern foundry sands are washed to remove the dangerous size fractions
and do not present a hazard as delivered. It must be recognised, however,
that certain foundry operations such as shot blasting, grinding of sand covered
castings or sand reclamation can degrade the sand grains, producing a fine
quartz dust having particle size in the harmful range below 5 microns.
Operators must be protected by the use of adequate ventilation and the
wearing of suitable face masks.
 Sands and green sand 149

Table 12.1 Typical UK and German foundry sands

Sieve size Sand type

UK sands German sands

microns BSS No. Chelford 50 Chelford 60 H32 H33 F32

1000 16 trace nil

700 22 0.7 0.4
500 30 4.5 2.3 1.0 0.5 1.0
355 44 19.8 10.0 15.0 7.5 7.0
250 60 44.6 25.7 44.0 30.0 30.0
210 72 21.6 23.8 39.0 60.0 60.0
150 100 8.2 28.7
100 150 2.6 7.6
75 200 nil 1.3 1.0 2.0 2.0
75 –200 nil 0.2 nil nil nil
AFS grain fineness No. 46 59 51 57 57
Average grain size 0.275 mm 0.23 0.27 0.23 0.23

Note: Haltern 32, 33 and Frechen 32 are commonly used, high quality German sands.
German sieve gradings are based on ISO sieves.
The German sands have rounder grains and are distributed on fewer sieves than UK
sands, they require significantly less binder to achieve the required core strength.

Segregation of sand

Segregation, causing variation of grain size, can occur during sand transport
or storage and can give rise to problems in the foundry. The greatest likelihood
of segregation is within storage hoppers, but the use of correctly designed
hoppers will alleviate the problem.

1. Hoppers should have minimum cross-sectional area compared to height.

2. The included angle of the discharge cone should be steep, 60–75°.
3. The discharge aperture should be as large as possible.

Measurement of sand properties

Acid demand value

Acid demand is the number of ml of 0.1 M HCl required to neutralise the

alkali content of 50 g of sand.

Weigh 50 g of dry sand into a 250 ml beaker

Add 50 ml of distilled water
Add 50 ml of standard 0.1 M hydrochloric acid by pipette
Stir for 5 minutes
Allow to stand for 1 hour
150 Foseco Ferrous Foundryman’s Handbook

Titrate with a standard solution of 0.1 M sodium hydroxide to pH values

of 3, 4, 5, 6 and 7
Subtract the titration values from the original volume of HCl (50 ml) to
obtain the Acid Demand Value

Grain size

See Section I, pp. 15, 16 for the method of measuring average grain size and
AFS grain fineness number.

Thermal characteristics of silica sand

Silica sand has a number of disadvantages as a moulding or coremaking

material.

It has a high thermal expansion rate (Fig. 12.2) which can cause expansion
defects in castings, such as finning or veining and scabbing.
It has a relatively low refractoriness (Table 12.2) which can cause sand
burn-on, particularly with steel or heavy section iron castings.
It is chemically reactive to certain alloys; for example, ferrous alloys
containing manganese. The oxides of Mn and Fe react with silica to form
low melting point silicates, leading to serious sand burn-on defects.

1.6
Silica
1.4
Linear expansion (%)

1.2

1.0

0.8 Olivine

0.6
Chromite
0.4

0.2 Zircon

200 400 600 800

Temperature (°C)

Figure 12.2 Thermal expansion characteristics of zircon, chromite and olivine

sands compared with silica sand. (Courtesy CDC.)

For certain types of casting, it may be necessary to use a non-silica sand,

even though all other sands are more expensive than silica.
 Sands and green sand 151

Table 12.2 Sintering points of silica sand

Sand Sintering point (°C)

High purity silica sand, >99% quartz 1450
Medium purity silica sand, 96% quartz 1250
Sea sand (high shell content) 1200
Natural clay bonded sand 1050–1150

Non-silica sands (Table 12.3)

Table 12.3 Properties of non-silica sands (compared with silica)

Property Silica Zircon Chromite Olivine

AFS grain size No. 60 102 74 65
Grain shape rounded rounded angular angular
Specific gravity 2.65 4.66 4.52 3.3
Bulk density (kg/m3) 1490 2770 2670 1700
(lb/ft3) 93 173 167 106
Thermal expansion 1.9% 0.45% 0.6% 1.1%
20–1200°C non-linear
Application general refractoriness resistance to Mn steel
chill metal
penetration
chill

Zircon, ZrSiO4

Zircon sand has a high specific gravity (4.6) and high thermal conductivity
which together cause castings to cool faster than silica sand. The chilling
effect of zircon sand can be used to produce favourable thermal gradients
that promote directional solidification giving sounder castings. The thermal
expansion coefficient of zircon is very low (Fig. 12.2) so that expansion
defects can be eliminated. Zircon has higher refractoriness than silica,
moreover it does not react with iron oxide, so sand burn-on defects can be
avoided. Zircon sand generally has a fine grading, with AFS number between
140 and 65 (average grain size 115–230 microns), the most frequently used
grade is around AFS 100.
Zircon is probably the most widely used of the non-silica sands. It is used
with chemical binders for high quality steel castings and for critical iron
castings such as hydraulic spool valves which contain complex cores, almost
totally enclosed by metal, making core removal after casting difficult. Zircon
has low acid demand value and can be used with all chemical binder systems.
The Cosworth casting process uses the low thermal expansion of zircon
sand cores and moulds to cast dimensionally accurate castings. The high
cost of zircon sand makes reclamation necessary and thermal reclamation
of resin bonded moulds and cores is frequently practised.
152 Foseco Ferrous Foundryman’s Handbook

Zircon sands contain low levels of naturally occurring radioactive materials,

such as uranium and thorium. Any employer who undertakes work with
zircon mineral products is required, by law, to restrict exposure of workers
to such naturally occurring contaminants so far as reasonably practical. The
primary requirement is to prevent the inhalation of zircon dust. Suitable
precautions are set out in the UK in HSE Guidance Note EH55: Dust –
general principles of protection. Guidance on the possible hazards associated
with the use, handling, processing, storage, transport or waste disposal of
such naturally occurring radioactive materials and the control measures
that are recommended to minimise exposure should be obtained from the
supplier.

Chromite, FeCr2O4

The high specific gravity (4.5) and high thermal conductivity of chromite
provide a pronounced chilling effect. Thermal expansion is low so expansion
defects are unlikely to occur. Chromite sand has a glossy black appearance,
it has greater resistance to metal penetration than zircon in spite of its
generally coarser grading (typically AFS 70). It has somewhat higher acid
demand than other sands which entails greater additions of acid catalyst
when furane resin is used. Apart from this the sand is compatible with all
the usual binder systems. Chromite is generally used for steel casting to
provide chilling. It is difficult to reclaim chromite sand since, if it becomes
contaminated with silica, its refractoriness is seriously reduced.

Olivine, Mg2SiO4

Olivine sand is used mainly for the production of austenitic manganese

steel castings (which react with silica and other sands to give serious burn-
on defects). It has also been used to avoid the health hazards possible with
silica sand. Olivine has a very high acid demand and is not suitable for use
with acid catalysed binders such as furan resins, it tends to accelerate the
curing of phenolic urethane binders. Being a crushed rock, it is highly angular
and consequently requires high binder additions. Thermal expansion is
regular and quite low.

Green sand
The earliest method of bonding sand grains to form a sand mould was to
use clay and water as a binder. The moulds could be used in the ‘green’ or
undried state (hence the term green sand moulding) or they could be baked
in a low temperature oven to dry and strengthen them to allow heavy
castings to be made. Nowadays, dried, clay bonded sand is little used,
 Sands and green sand 153

having been replaced by chemically bonded sand, but green sand is still the
most widely used moulding medium, particularly for iron castings. 40 years
ago, 90% of all steel castings were produced in green sand moulds but this
has now declined and the majority of steel castings are now made in chemically
bonded sand moulds. This reflects the declining tonnage of high volume
repetition steel castings now being made.
Originally, naturally occurring clay-sand mixtures were used, containing
10% or more of clay. It was found that by adding coal dust, the ease of
stripping iron castings from the mould and the surface finish of the castings
could be greatly improved. The heating of the coal dust by the liquid iron
causes the formation of a type of carbon called lustrous carbon which is not
wetted by the liquid iron, so the cast surface is improved. Coal dust additives
are not used for steel casting since they would cause unacceptable carbon
pick-up on the casting surface.
Clay bonded moulding sand can be used over and over again by adding
water to replace that which is lost during casting, and re-milling the sand.
However, clay which is heated to a high temperature becomes ‘dead’, that
is it loses its bonding power. The coal dust is partly turned to ash by heat,
so new clay, coal dust and water must be added and the sand re-milled to
restore its bonding properties. As the sand is re-used, dead clay and coal
ash build up in the sand, reducing its permeability to gases so that eventually
water vapour and other mould gases are unable to escape from the mould
and defective castings are produced.
Natural sands are little used nowadays (except for some aluminium
castings) and most iron foundry green sands are ‘synthetic’ mixtures based
on washed silica sand to which controlled additions of special moulding
clays (bentonites) and low ash coal dust are made. Alternatively, special
blended additives such as BENTOKOL or CARSIN, which combine the clay
binder with lustrous carbon formers, can be used. Steel foundry green sands
contain bentonites, starches and dextrins. The moulding sand becomes a
‘sand system’ which is continuously recycled, with suitable additions and
withdrawals made as required. The control of system sand to achieve constant
moulding properties has become an important technology of its own for it
determines the quality of the castings produced.

Green sand additives

Base sand: Silica sand of AFS grain size 50–60 is usually used. The particle
size distribution is important, a sand spread over 3 to 5 consecutive sieve
sizes with more than 10% on each sieve gives the best results (see Table
12.1). Rounded or sub-angular sands are the best since the more round the
grains, the better the flowability and permeability of the sand. The base
sand should have the same size grading as the core sand used in the foundry,
so that burnt core sand entering the system does not alter the overall size
grading.
154 Foseco Ferrous Foundryman’s Handbook

Clay: The best bonding clays are bentonites which can either have a calcium
or a sodium base. Sodium bentonites occur naturally in the USA as Wyoming
or Western bentonite and also in other countries particularly in the
Mediterranean area. Green sand produced with this clay has medium green
strength and high dry strength which increases the resistance to erosion of
metal but can give problems at shakeout. Sodium bentonites are commonly
used for steel casting production.
Calcium bentonites are more widely distributed, they produce green sands
with rather high green strength but low dry strength so they have low
erosion resistance and are prone to scabbing and other expansion defects.
Calcium bentonites can be converted to sodium bentonite by adding soda
ash, the calcium base is replaced by sodium and the clay then has properties
approaching those of natural sodium bentonite. Such clays are known as
activated clays.
Blended or mixed bentonites are commercial blends of sodium bentonite
with calcium bentonite or a sodium-activated bentonite. Compositions can
be varied to suit particular applications. Most iron foundries use blended
bentonites.
Clays absorb water from the atmosphere so they should be stored in dry
conditions, they do not deteriorate on storage, even for long periods.

Coal dust: Used mainly in iron foundries with some being used in non-
ferrous foundries. The formation of ‘lustrous carbon’ from the thermal
degradation of the volatiles given off during casting improves casting surface
finish and strip. Good coal dust has the following properties:

Volatile content: 33–36%

Ash: less than 5%
Fixed carbon: 50–54%
Sulphur: less than 1%
Chlorine: less than 0.03%
Size: 75 or 100 grade (AFS grain fineness No.)

It is important that the size grading of the coal dust used should not be too
fine. Coal dust increases the moisture requirement of the sand and the finer
it is, the more moisture is needed, which may have harmful effects on the
castings. Fine coal dust will also reduce the overall permeability of the
sand. Coal dust levels in green sand vary from 2 or 3% for small castings to
7 or 8% for heavy section castings. Too much coal dust can give rise to gas
holes in the castings or misruns.
Coal dust must be stored dry to prevent the risk of fire. Damp coal dust
can ignite spontaneously. Large stocks are undesirable and stocks should be
rotated; first in, first out.

Coal dust replacements: Consist of blends of high volatile, high lustrous carbon
materials blended with clays. They are generally more environmentally
acceptable than coal dust, producing less fume during casting. They can be
 Sands and green sand 155

designed for particular applications such as high pressure moulding.

BENTOKOL and CARSIN additives are blends of natural clays to which
have been added specially selected essential volatiles to provide both bond
and volatile content for iron foundry green sand. They are in powder form,
virtually dust free with flow properties and bulk densities similar to clay,
they are supplied in bulk. Use of BENTOKOL/CARSIN often means that a
‘one shot’ sand addition is sufficient and always improves working conditions.
Fume after casting is reduced and the foundry is cleaner and pleasanter
while casting finish is maintained or enhanced.
Many of the BENTOKOL/CARSIN formulations contain a blend of natural
sodium bentonite which has high burn-out temperature and results in lower
concentrations of dead clay. This improves permeability, reduces moisture
requirements and gives better all-round sand properties over a longer life
cycle. Sodium bentonites take longer than other clays to develop their bond.
To offset this BENTOKOL/CARSIN formulations may include dispersion
agents so that the clay is rapidly spread and the bond quickly developed.
BENTOKOL/CARSIN is compatible with sand systems based on clay and
coal dust. Certain parameters of the sand system may change with the
establishment of BENTOKOL/CARSIN; there may be a slight reduction of
volatiles, LOI and clay fraction and the optimum moisture content will fall
by up to 0.5%.
BENTOKOL 80 is a high volatile : clay ratio sand conditioner for the
production of iron castings where exceptional surface finish is required. It
is suitable for use in high pressure or boxless moulding lines. It is a free
flowing grey powder. Because of its high volatile content, one part of
BENTOKOL 80 can usually replace 1.8 parts of coal dust and will additionally
replace a proportion of the clay addition. The use of BENTOKOL 80 ensures
a positive and rapid response to addition rate changes so allowing immediate
alteration of sand properties to match changes in production rate or pattern
mix.
LUCIN is a coal dust replacement product having no bentonite.

Cereal binders: Used mainly in steel foundries to increase strength and

toughness of the green sand. There are two main types of cereal binder:
starch and dextrin. Starch is the basic material and is produced from a
number of plant materials with maize starch being the most commonly
used for foundry purposes. Starch is treated by heating to form ‘pre-gelatinised
starch’ for foundry use. In this form it develops a gel when mixed with
water at about 10% concentration. Dextrin is produced by treating starch
with a weak acid and cooking at 120–150°C. The starch is re-polymerised
and converted to dextrin which is freely soluble in water, forming syrups.
Starch and dextrin additions affect the bonding and moulding properties
of green sand mixtures (Table 12.4).
The main benefits to moulding properties obtained by adding a cereal
binder are:

Providing greater toughness to the mix and, therefore, the compacted

156 Foseco Ferrous Foundryman’s Handbook

Table 12.4 The effect of starch and dextrin on green sand mixes

Property Starch addition Dextrin addition

Green strength Increases moderately Increases slightly

Toughness Increases moderately Increases markedly
Green deformation Increases Increases
Pattern stripping Improves Improves markedly
Mould compaction Little effect Can improve compaction
with high pressure moulding
Sensitivity to change Lowered Lowered
in moisture
Dry strength Increases slightly Increases markedly at
higher moisture content
Friability of mould Increases slightly Markedly reduces
surfaces

From SCRATA Technical Bulletin No. 16, Additives to Green Sand.

sand can deform more easily without fracture, i.e. the cereals correct
the brittleness of the sand. This enables the sand (mould) to strip from
difficult pockets in the pattern without failing. Starch and dextrin are
both effective in this way.
Providing harder and less friable mould surfaces and edges where green
sand moulds are allowed to air dry. Starch is not effective for this
purpose but dextrin, being water soluble migrates to the mould surface
forming a hard, less friable crust. Friable mould surfaces can cause
dirty moulds and sand inclusions in the casting and dextrin corrects
friability problems.

Effect of cereal binders on casting surface quality

Cereal additions do not improve the erosion resistance of the sand or resistance
to metal penetration. They are very good anti-scabbing agents and the higher
the amount of starch or dextrin the greater is the resistance to expansion
scabbing. Starch is more effective in preventing expansion scabbing than a
dextrin and should be employed when making larger green sand moulds
and where radiation effects from the rising steel are most severe.

Use of cereals in moulding mixes

Effective additions of cereal binders to new green sand facing mixes are
between 0.5 and 0.75%. Accurate weighing of the cereal addition is a primary
requirement as is efficient mulling to develop the full properties of the
relatively tough mixes.
In unit type green sands part of the cereal is destroyed during the casting
process. Cereal binders lose their beneficial properties when heated beyond
 Sands and green sand 157

225°C. A recommended addition on each recycle is 0.1 to 0.25% depending

upon the amount of cereal burn-out and the dilution by cores and new
sand.
Choice of the type of cereal binder to use depends upon the intended
purpose for the green sand in the foundry. Dextrin is to be preferred for
general green sand work and where moulds air dry before pouring. Starch
is best for larger work where expansion scabbing is possibly a more serious
problem. As a compromise and to give the beneficial effects from both
types, a percentage of starch and dextrin should be added to the moulding
mixes. A suitable proportion is two-thirds starch to one-third dextrin.

Water: Water is needed to develop the clay bond but it can cause casting
defects. Where there is strong, localised heating, e.g. in the vicinity of ingates
or on horizontal mould surfaces exposed to radiant heat from the metal,
moisture is driven back from the mould surface, condensing in a wet, weak
underlying layer that can easily fracture to produce expansion defects in
castings such as scabs, rat-tails and buckles.

MIXAD additive: MIXAD 61 additive has been developed to eliminate sand

expansion defects. By improving the wet strength of the condensed water
layer, MIXAD 61 eliminates the associated defects. MIXAD 61 is added
early in the milling cycle, usually 0.5–1.0% is used in the facing sand. If
MIXAD 61 is used in a unit sand system, an addition of 0.05–0.10% is usually
sufficient. MIXAD 61 is used principally in sands for iron and copper base
alloys.

The green sand system

The ‘sand system’ in green sand foundries is illustrated in Fig. 12.3. It

comprises the following items of equipment;

The bulk sand storage hopper

Returned knockout sand is stored in a hopper which ideally has capacity
for about 4 hours of usage. Returned sand is very variable in properties,
since some will have been heated very strongly by the hot metal while sand
which came from the edges of the mould will have been heated very little.
Some mixing of the sand occurs in the storage hopper. It is better to have
several smaller volume hoppers rather than one large unit since mixing of
the return sand is improved. Hoppers should be designed in such a way
that sand flows evenly through without sticking to the walls.

The sand mill

At the mill, the additions of new sand, clay, coaldust and water are made to
the returned sand. Sand mills may be continuous or batch but batch mills
are nowadays preferred because better sand control is possible. Several
158 Foseco Ferrous Foundryman’s Handbook

designs of mill are available, their purpose is to mix the sand and spread the
moistened clay over the surface of the sand grains in order to develop the
bond. Older sand mills used heavy vertically set mulling wheels and plough
blades (Fig. 12.4). A certain optimum milling time is necessary and this has
to be determined for each green sand system and adhered to carefully. Short
milling times underuse the additives and produce sand having low and
variable green strength. Modern sand mills use intensive mixers to develop
the clay bond quickly and efficiently. Whatever type of mill is used, regular
maintenance is essential to ensure that the milling efficiency does not change.
Additions of new sand, clay and coal dust (or coal dust replacement) are
made at the mill. Batch mills use weighed additions while continuous mills
use calibrated feeders to make continuous additions. Water is added, often
via an automatic moisture controller which measures the moisture content
of the sand as it feeds into the mixer batch hopper prior to mixing then
calculates the exact amount of water to be added to the sand batch discharged
into the mixer to bring the moisture to the required level. Prepared sand is
usually passed through an aerator which ‘fluffs up’ the sand before feeding
the moulding machines.

Buffer hopper
A buffer hopper is usually provided before the moulding machines, so that
a moulding machine stoppage does not necessitate running freshly milled
sand straight back to the main hopper.

The moulding machine(s)

These can be jolt-squeeze, high pressure, impact moulding machines etc.
Each has its advantages and each requires special moulding sand properties.
See p. 164.

Casting
The moulds are cast then allowed to cool for a suitable time, often 30–40
minutes, before knocking out the castings. The cooling time not only allows
the castings to solidify completely but also reduces internal stresses within
the casting caused by differential cooling of sections of varying thickness.

Shake-out
The shakeout separates the castings from the sand. It may be a vibrating
grid or a rotating drum, the latter also having a cooling function since water
is sprayed onto the sand in the drum. The shakeout is provided with copious
air extraction, to prevent dust from entering the foundry. This also removes
some fines from the sand and so plays an important part in the control of
the sand system.

Return sand conveyor

Sand from the shakeout contains lumps of partially burnt cores, very hot
burnt core sand, hot moulding sand from near the casting and cooler moulding
sand from the edges of moulds. A magnetic separator removes metallic
 Sands and green sand 159

Rotary
screen
Fluid bed cooler
Road Bulk
Tanker sand
Fill
pipes storage

Coal New
Clay
dust sand
Sand
weigh
hopper
Moisture
controller
Weigh
Water
in
Mixer
Aerator

Elevator
Prepa
Feed red s
and conve
hopper yor

Knockout Water
Moulding sprays
Mag.
machine
Sep.
r
v eyo
con
Spill sand conveyor s and
urn
Ret

Figure 12.3 Flow diagram for a typical green sand plant. (From Foundryman,
March 1998, p. 102. Courtesy Foundry and Technical Liaison Ltd.)

Muller wheel

Plough blade

Figure 12.4 Vertical wheel batch muller. (Sixth Report of Institute Working Group
T30, Mould and Core Production. Foundryman, Feb. 1986.)
160 Foseco Ferrous Foundryman’s Handbook

particles and water sprays are used to effect some preliminary cooling before
elevation, since very hot sand can damage the elevator.

Screen
The return sand is elevated to a screen where residual core lumps and other
unwanted material are removed before the sand is returned to the hopper.
Non-magnetic metallics are also removed by the screen.

Sand cooling
Hot sand causes problems of excessive moisture loss and condensation of
moisture on patterns and cores, so the returned sand must be cooled. The
only effective way of cooling green sand is by evaporation of water.
Evaporation of 1% water cools the sand by about 25°C, so water is sprayed
onto the sand and air drawn over the sand to evaporate it, either in the
rotary drum shakeout or at the screen or in a specially designed fluidised
bed cooler. Returned green sand cannot be fluidised by air alone. Sand is
delivered onto a screen which is vibrated to transport the sand along it.
Water is added by sprays and evaporated by blowing air through the
perforated screen.

Sand removal or addition

The sand system must have a facility for removal of excess sand, since
foundries making highly cored castings build up the sand level in the system
through the introduction of burnt core sand. Regular addition of new sand
is needed to maintain control of green sand properties.

Green sand properties

Green sand for iron foundries typically has the following properties:

jolt/squeeze high pressure

machines (DISA etc.)

water 3–4% 2.5–3.2%

green strength 70–100 kPa 150–200 kPa
10–15 psi 22–30 psi
compactability 45–52% 38–40%
permeability 80–110 80–100
live clay 5.0–5.5% 6.0–10.0%
volatiles 2.5% 2.0%
LOI 7.0–7.5% 6.0%

Steel foundries use sand having similar properties except for reduced volatiles
and LOI since coal dust is not used.
A typical grading of a sand suitable for iron or steel castings is:
 Sands and green sand 161

Sieve mesh %
size (µm)

710 0.14
500 1.42
355 5.44
250 22.88
212 19.64
150 26.74
106 9.70
75 2.84
pan 1.10

Control of green sand systems

When iron or steel is poured into green sand moulds, the heat from the
metal drives off some water from the sand and clay, burns some of the coal
dust to coke and ash (in the case of iron), and burns a proportion of the clay
so that its bonding properties are destroyed. Before the sand can be used
again its properties must be restored by removing burnt clay, coal dust and
ash, and milling in new clay, coal dust (or dextrin) and water. The following
additions are usually necessary at the sand mill in order to maintain the
required moulding properties of the sand.

0.3–0.5% of new clay

0.3–0.5% of coal dust (or 0.2% dextrin)
1.5–2.5% of water

In addition, it is desirable to add a proportion (up to 10%) of new silica sand

to the system and to dispose of a corresponding amount of used system
sand. If highly cored work is made, the amount of new sand addition can
possibly be reduced since well-burnt core sand may serve as new sand, but
care must be taken since certain core binder residues (particularly from
phenolic-isocyanate binders) can harm the green sand properties.
The precise amount of the additions that must be made depends on
many factors, including:

The weight and type of castings being made, which affects the burn-out
The amount of core sand entering the system
Whether hold-ups on the moulding line have allowed freshly milled
sand to return directly to the sand hopper

Good sand control depends on carefully monitored experience of the particular

sand system, since every system is different. The normal additions of clay,
coal dust or BENTOKOL/CARSIN at each cycle represent only about 10%
of the total active clay or coal present in the system. It is not possible to
change the total clay or coal dust content quickly since any change in the
162 Foseco Ferrous Foundryman’s Handbook

addition rate takes about 20 cycles to work its way fully into the system. For
example; an addition of 0.3% clay is usually sufficient to maintain the total
clay level at 3.0%. If the clay addition is increased to 0.4%, the total clay
content after one cycle will only rise to about 3.1% and it will take 20 cycles
(about 1 week) for the full effect of the change of addition to be felt and for
the clay level to rise to around 4.0%.
The operator must find by experience what additions are needed during
normal operation of the foundry to maintain the required physical properties
of the sand; changes will normally only be necessary if some change of
practice has occurred, such as a change to heavier castings, which will cause
more burn-out of clay and coal. A change to more highly cored castings is
likely to result in more burnt core sand entering the system, which will
require greater additions to compensate.
Because sand systems have such built-in inertia, or resistance to change,
the only way that quick changes can be made to the moulding properties of
the sand is by increasing or reducing the addition of water to the sand mill.
This has an immediate effect on sand properties and as long as the other
constituents of the sand are in control, the sand’s mouldability can quickly
be corrected.

Sand testing
Regular testing of the properties of the sand is essential. One or two sand
tests do not truly indicate the condition of the whole sand system, since a
sand sample weighs only about 1 kg and cannot represent the whole 200
tonnes or so of the system. At least five samples should be taken per shift
and measured for moisture, green strength, compactability and permeability.
LOI and volatiles should be measured once per day. Active clay, twice per week.
Records of additions should also be kept:

Weight of clay, coal dust and new sand added each day
Number of moulds made
Weight of iron poured
Weight of used sand removed from the system each day

The three-ram test method developed by the AFS in the 1920s is still widely
used to measure green strength and compactability although it is not really
suitable for sand used in modern high pressure moulding machines. The
AFS ramming test is being replaced by tests on 50 mm diameter specimens
produced by squeeze machines which can reproduce the pressures actually
developed on the mould by the machine that the foundry uses.

Control graphs
Individual figures mean rather little, but daily average sand properties should
 Sands and green sand 163

be plotted together with weekly figures of active clay and average additions
of clay, coal dust and new sand. After a few weeks of plotting the data, it
will be possible to draw control lines. Variation within the lines is permissible
but if results appear outside the control lines, then action must be taken
such as increasing or reducing clay or coal or new sand. If action is taken,
it must be remembered that it will be 1 or 2 days before the full effect of the
action will be seen.
The sand properties, moisture, green strength, compactability and
permeability are of first importance and must be maintained. The remaining
graphs are of secondary importance, but provide useful information which
may allow problems to be anticipated. It is also necessary to keep a log book
for the sand system listing information such as:

When the sand mill was maintained

When new sand was added and old sand removed
Changes in foundry practice, such as change of core binder, or change
in type of casting made.

Parting agents
Unless a water-repellent material is present on the pattern surface to prevent
wetting, green sand will tend to stick to the pattern when rammed. This will
cause damage to the mould when the pattern is withdrawn. To prevent this,
it is usual to spray the pattern with a liquid parting agent such as one of the
SEPAROL grades. The frequency of application depends on the type and
quality of the pattern, but every 10th or 12th mould is typical.

Special moulding materials, LUTRON

LUTRON moulding sand is a waterless sand containing mineral oils designed
for use with fine sands to produce a superfine finish in castings such as
name plates, plaques, medallions and other art castings made in aluminium
or copper base alloys.
The ready-mixed LUTRON moulding sand is based on a very fine sand
(AFS No. 160). The sand has excellent flowability due to the lubricating
properties of the bond. It can be used alone or as a facing sand backed with
normal green sand or other type of sand. If used as a unit sand, the mix can
be reconstituted after casting by adding a compensating amount of LUTRON
binder and milling.
The LUTRON binder can be used with any suitable fine, dry sand; additions
of 10–12% are normally needed.
164 Foseco Ferrous Foundryman’s Handbook

Green sand moulding machines

The moulding machine must compact the green sand evenly around the
pattern to give the mould sufficient strength to resist erosion while liquid
metal is poured, to withstand the ferrostatic pressure exerted on the mould
when full and to resist the expansion forces that occur on solidification of
the iron. Green sand does not move easily under compression forces alone
so that achieving uniform mould strength over a pattern of complex shape
by simply squeezing is not possible. Various combinations of jolting and
vibrating with squeeze applied simultaneously or sequentially have been
used to produce uniform strength moulds at up to 400 moulds per hour.
Green sand moulds have traditionally been made in steel flasks, but flaskless
moulding is now widely used for smaller castings.

Moulding in flasks

The basic principles used are:

Jolt-Squeeze A pattern plate carrying the pattern surrounded by a moulding

flask is fitted onto a jolt piston. Green sand from a hopper above the machine
fills the moulding box loosely. The assembly is raised by the pneumatic jolt
cylinder and allowed to fall against a stop. The jolt action is repeated a pre-
set number of times causing the sand to be compacted to some extent, the
density being highest nearest the pattern plate. Additional compaction is
then provided by a pneumatic or hydraulic squeeze plate (Fig. 12.5a). More
uniform squeeze can be achieved by using compensating squeeze heads
(Fig. 12.5b) which compact the sand more uniformly even though the depth
of sand over the pattern varies.

Vibration-squeeze Instead of a jolt table, the pattern and flask are vibrated
before and during squeezing by a multi-ram head.

Shoot-squeeze A shoot head above the flask contains loose green sand. Opening
a shoot valve allows compressed air stored in a reservoir to travel through
the shoot head causing a column of sand to be shot into the flask where the
kinetic energy of the rapidly moving sand compacts it against the pattern.
A hydraulic cylinder then squeezes the mould to complete the compaction.

Impulse compaction The flask is filled loosely with prepared sand which is
compacted by means of a pressure wave generated by explosion or by
compressed air. Compaction is greatest near the mould surface. Squeeze
pressure may then be applied to compress the back of the mould. Deep
pattern draws are possible with this method.

Vacuum squeeze A partial vacuum is created in the flask surrounding the

 Sands and green sand 165

Compensating head
Flask
Flask Stiffener
Moulding sand
Guide Moulding sand Peen
pin Guide block
pin
Pattern
Anvil Pattern plate
Anvil
Ram-jolt
piston Ram-joit
piston

Pattern plate Push-off pin Air cylinder Push-off pin

Air cylinder on stripping frame
on stripping frame
Jolt squeeze moulding machine Jolt squeeze moulding machine
with solid squeeze heads with compensating heads
(a) (b)

Figure 12.5 Jolt squeeze moulding machine: (a) with solid squeeze head; (b) with
compensating heads. (Sixth Report of Institute Working Group T30, Mould and Core
Production. Foundryman, Feb. 1996.)

pattern. A metered amount of sand is released into the chamber where the
vacuum accelerates the sand which impacts onto the pattern causing
compaction. A multi-ram head provides high pressure squeeze to complete
the compaction of the mould. The system is suitable for large moulds.

Flaskless moulding

Horizontally parted (match-plate moulding)

A matchplate is a pattern plate with patterns for both cope and drag mounted
on opposite faces of the plate. Both cope and drag halves of the mould are
filled with prepared sand in the machine before being brought together for
the high pressure squeeze with simultaneous vibration to compact the sand.
The completed mould is pushed out of the machine onto a shuttle conveyor.
Moulds can be made at up to 200 per hour.

Vertically parted moulding

The Disamatic flaskless moulding machine introduced in the late 1960s

(now supplied by Georg Fischer Disa) revolutionised green sand moulding,
allowing high precision moulds to be made at up to 350 moulds/hour. The
method of operation is shown in Fig. 12.6. One pattern half is fitted onto the
end of a hydraulically operated squeeze piston with the other pattern half
fitted to a swing plate, so called because of its ability to move and swing
166 Foseco Ferrous Foundryman’s Handbook

1. Sand shot 2. Mould squeeze from 3. Stripping off the swing

two sides squeeze plate

4. Mould close-up and mould 5. Stripping of the rear 6. Closing the moulding
string transport squeeze plate chamber

Figure 12.6 Vertically parted flaskless moulding, the Disamatic machine. (Sixth
Report of Institute Working Group T30, Mould and Core Production. Foundryman,
Feb. 1996.)

away from the completed mould. Sand from a supply hopper above the
machine is blown into the moulding chamber by means of a variable pressure
compressed air supply stored in a nearby air receiver. Vacuum can be applied
to the moulding chamber to vent air and assist in drawing sand into deep
pattern recesses. Both halves of the pattern are hydraulically squeezed together
to compress the sand block. As the swing plate moves away, the piston
pushes the new mould to join ones previously made, to form a continuous
mould string. Mould sizes available are from 500 mm × 400 mm × 315 mm
on the smallest 2110 model, up to 950 mm × 800 mm × 635 mm on the largest
model manufactured, the 2070. Flexibility is available through variable mould
output, variable mould thickness, fast pattern change and core placing options.
Varying degrees of control sophistication are provided dependent on the
model. Cores can be placed in the mould using a mechanised core placer.
There are many variations on the moulding principles described above.
See Sixth Report of Institute of British Foundrymen Working Group T30
(Foundryman, Feb. 1996, p. 3) from which some of the above information has
been taken.
Chapter 13
Resin bonded sand

Chemical binders
A wide variety of chemical binders is available for making sand moulds
and cores. They are mostly based either on organic resins or sodium silicate
(see Chapter 14), although there are other inorganic binders such as cement,
which was the earliest of the chemical binders to be used; ethyl silicate,
which is used in the Shaw Process and for investment casting and silica sol,
which is also used for investment casting.
The binders can be used in two ways:

As self-hardening mixtures; sand, binder and a hardening chemical are

mixed together; the binder and hardener start to react immediately, but
sufficiently slowly to allow the sand to be formed into a mould or core
which continues to harden further until strong enough to allow casting.
The method is usually used for large moulds for jobbing work, although
series production is also possible.
With triggered hardening; sand and binder are mixed and blown or rammed
into a core box. Little or no hardening reaction occurs until triggered by
applying heat or a catalyst gas. Hardening then takes place in seconds.
The process is used for mass production of cores and in some cases, for
moulds for smaller castings.

Self-hardening process (also known as self-set,

no-bake or cold-setting process)
Clean, dry sand is mixed with binder and catalyst, usually in a continuous
mixer. The mixed sand is vibrated or hand-rammed around the pattern or
into a core box; binder and catalyst react, hardening the sand. When the
mould or core has reached handleable strength (the strip time), it is removed
from the pattern or core box and continues to harden until the chemical
reaction is complete.
Since the binder and catalyst start to react as soon as they are mixed, the
mixed sand has a limited ‘work time’ or ‘bench life’ during which the mould
or core must be formed (Fig. 13.1). If the work time is exceeded, the final
strength of the mould will be reduced. Work time is typically about one-
168 Foseco Ferrous Foundryman’s Handbook

third of the ‘strip time’ and can be adjusted by controlling the type of
catalyst and its addition rate. The work time and strip time must be chosen
to suit the type and size of the moulds and cores being made, the capacity
of the sand mixer and the time allowable before the patterns are to be re-
used. With some binder systems the reaction rate is low at first, then speeds
up so that the work time/strip time ratio is high. This is advantageous,
particularly for fast-setting systems, since it allows more time to form the
mould or core.

psi kPa
500 3500

80% max
Compression
strength

50 350

1.5 10
Tw Ts Tc Tmax
Time

Figure 13.1 Typical hardening curve for self-hardening sand:

Tw = work time
Ts = strip time
Tc = casting time
Tmax = time to achieve maximum strength.

Stripping is usually possible when the sand has reached a compression

strength of around 350 kPa (50 psi) but the actual figure used in practice
depends on the type of binder system used, the tendency of the binder to
sag before it is fully hardened, the quality of the pattern equipment and the
complexity of the moulds and cores being made.
It is advisable to strip patterns as soon as it is practical, since some binder
chemicals attack core box materials and paints after prolonged contact. The
properties of chemical binders can be expressed in terms of:
 Resin bonded sand 169

Work time (bench life): which can be conveniently defined as the time after
mixing during which the sand mixture has a compressive strength less
than 10 kPa, at this stage it is fully flowable and can be compacted easily.
Strip time: which can be defined as the time after mixing at which a
compressive strength of 350 kPa is reached, at this value most moulds
and cores can be stripped without damage or risk of distortion.
Maximum strength: the compressive strength developed in a fully hardened
mixture, figures of 3000–5000 kPa are often achieved.

It is not necessary to wait until the maximum strength has been achieved
before moulds can be cast, the time to allow depends on the particular
castings being made, usually casting can take place when 80% of the maximum
strength has been reached.

Testing chemically bonded self-hardening sands

Units

Compressive strength values may be reported in

SI units kPa = kN/m2

cgs units kgf/cm2
Imperial units psi = lbf/in2

Conversion factors:

100 kPa (kN/m2) = 1.0197 kgf/cm2

= 14.5038 psi (lbf/in2)
1 kgf/cm2 = 98.0665 kPa
= 14.22 psi (lbf/in2)
1 psi (lbf/in2) = 6.895 kPa (kN/m2)
= 0.07032 kgf/cm2

Conversion table
kPa (kN/m2) kgf/cm2 psi (lbf/in2)
10 0.10 1.5
50 0.51 7.3
100 1.02 14.5
200 2.04 29.0
300 3.06 43.5
400 4.08 58.0
500 5.10 72.5
600 6.12 87.0
700 7.14 101.5
170 Foseco Ferrous Foundryman’s Handbook

800 8.16 116.0

900 9.18 130.5
1000 10.20 145.0
2000 20.39 290.1
3000 30.59 435.1
4000 40.79 580.1
5000 50.99 725.2

The curing properties (work time, strip time and maximum strength) are
measured by compression tests using 50 mm diameter specimen tubes with
end cups, or AFS 2 inch diameter tubes, with a standard rammer. Sand is
mixed in a food mixer or small coresand mixer; catalyst being added first
and mixed, then the resin is added and mixed.

Measurement of ‘work time’ or ‘bench life’

Mix the sand as above, when mixing is complete, start a stopwatch and
discharge the sand into a plastic bucket and seal the lid.
After 5 minutes, prepare a standard compression test piece and immediately
measure the compressive strength.
At further 5 minute intervals, again determine the compressive strength,
stirring the mixed sand in the bucket before sampling it.
Plot a graph of time against strength and record the time at which the
compressive strength reaches 10 kPa (0.1 kgf/cm2, 1.5 psi); this is the work
time or bench life.
The sand temperature should also be recorded.
For fast-setting mixtures, the strength should be measured at shorter
intervals, say every 1 or 2 minutes.

Measurement of strip time

Prepare the sand mixture as before.

When mixing is complete, start a stop-watch.
Prepare 6–10 compression test pieces within 5 minutes of completion of
mixing the sand.
Cover each specimen with a waxed paper cup to prevent drying.
Determine the compressive strength of each specimen at suitable intervals,
say every 5 minutes.
Plot strength against time.
Record the time at which the strength reaches 350 kPa (3.6 kgf/cm2, 50 psi),
this is the ‘strip time’.

The sand temperature should also be recorded.

Measurement of maximum strength

Prepare the sand mixture as before.

 Resin bonded sand 171

Record the time on completion of mixing.

Prepare 6–10 specimens as quickly as possible covering each with a waxed
cup.
Determine the strength at suitable intervals say, 1, 2, 4, 6, 12, 24 hours.
Plot the results on a graph and read the maximum strength.

The sand temperature should be held constant if possible during the test.

While compressive strength is the easiest property of self-hardening sand

to measure, transverse strength or tensile strength are being used more
frequently nowadays, particularly for the measurement of maximum strength.

Mixers

Self-hardening sand is usually prepared in a continuous mixer, which consists

of a trough or tube containing a mixing screw. Dry sand is metered into the
trough at one end through an adjustable sand gate. Liquid catalyst and
binder are pumped from storage tanks or drums by metering pumps and
introduced through nozzles into the mixing trough; the catalyst nozzle first
then binder (so that the binder is not exposed to a high concentration of
catalyst).

Calibration of mixers

Regular calibration is essential to ensure consistent mould and core quality

and the efficient use of expensive binders. Sand flow and chemical flow
rates should be checked at least once per week, and calibration data recorded
in a book for reference.

Sand: Switch off the binder and catalyst pumps and empty sand from the
trough. Weigh a suitable sand container, e.g. a plastic bin holding about
50 kg. Run the mixer with sand alone, running the sand to waste until a
steady flow is achieved. Move the mixer head over the weighed container
and start a stop watch. After a suitable time, at least 20 seconds, move the
mixer head back to the waste bin and stop the watch. Calculate the flow
in kg/min. Repeat three times and average. Adjust the sand gate to give
the required flow and repeat the calibration.
Binders: Switch off the sand flow and the pumps except the one to be
measured. Disconnect the binder feed pipe at the inlet to the trough,
ensuring that the pipe is full. Using a clean container, preferably a polythene
measuring jug, weigh the binder throughput for a given time (minimum
20 seconds). Repeat for different settings of the pump speed regulator.
Draw a graph of pump setting against flow in kg/min.
Repeat for each binder or catalyst, taking care to use separate clean
containers for each liquid. Do not mix binder and catalyst together, since
172 Foseco Ferrous Foundryman’s Handbook

they may react violently. Always assume that binders and catalysts are
hazardous, wear gloves, goggles and protective clothing.
When measuring liquid flow rate, the pipe outlet should be at the same
height as the inlet nozzle of the mixer trough, so that the pump is working
against the same pressure head as in normal operation.

Mixers should be cleaned regularly. The use of STRIPCOTE AL applied to

the mixer blades, reduces sand build-up.

Sand quality

In all self-hardening processes, the sand quality determines the amount of

binder needed to achieve good strength. To reduce additions and therefore
cost, use high quality sand having:

AFS 45–60 (average grain size 250–300 microns)

Low acid demand value, less than 6 ml for acid catalysed systems
Rounded grains for low binder additions and flowability
Low fines for low binder additions
Size distribution, spread over 3–5 sieves for good packing, low metal
penetration and good casting surface.

Pattern equipment

Wooden patterns and core boxes are frequently used for short-run work.
Epoxy or other resin patterns are common and metal equipment, usually
aluminium, may be used for longer running work. The chemical binders
used may be acid or alkaline or may contain organic solvents which can
attack the patterns or paints. STRIPCOTE AL aluminium-pigmented
suspension release agent or silicone wax polishes are usually applied to
patterns and core boxes to improve the strip of the mould or core. Care
must be taken to avoid damage to the working surfaces of patterns and
regular cleaning is advisable to prevent sand sticking.

Curing temperature

The optimum curing temperature for most binder systems is 20–25°C but
temperatures between 15 and 30°C are usually workable. Low temperatures
retard the curing reaction and cause stripping problems, particularly if metal
pattern equipment is used. High sand temperatures cause reduction of work
time and poor sand flowability and also increase the problem of fumes from
the mixed sand. If sand temperatures regularly fall below 15oC, the use of
a sand heater should be considered.
 Resin bonded sand 173

Design of moulds using self-hardening sand

Moulds may be made in flasks or flaskless. Use of a steel flask is common

for large castings of one tonne or more, since it increases the security of
casting. For smaller castings, below one tonne, flaskless moulds are common.
Typical mould designs are illustrated in Fig. 13.2. The special features of
self-hardening sand moulds are:

Large draft angle (3–5°) on mould walls for easy stripping

Incorporation of a method of handling moulds for roll-over and closing
Means of location of cope and drag moulds to avoid mismatch
Reinforcement of large moulds with steel bars or frames
Clamping devices to restrain the metallostatic casting forces
Use of a separate pouring bush to reduce sand usage
Mould vents to allow gas release
Sealing the mould halves to prevent metal breakout
Weighting of moulds if clamps are not used
Use of minimum sand to metal ratio to reduce sand usage, 3 or 4 to 1 is
typical for ferrous castings

Foundry layout

With self-hardening sand; moulds and cores are often made using the same
binder system, so that one mixer and production line can be used. A typical
layout using a stationary continuous mixer is shown in Fig. 13.3. The moulds
may or may not be in flasks. Patterns and core boxes circulate on a simple
roller track around the mixer. The length of the track is made sufficient to
allow the required setting time, then moulds and cores are stripped and the
patterns returned for re-use.
For very large moulds, a mobile mixer may be used.

Sand reclamation
The high cost of new silica sand and the growing cost of disposal of used
foundry sand, make the reclamation and re-use of self-hardening sands a
matter of increasing importance. Reclamation of sand is easiest when only
one type of chemical binder is used. If more than one binder is used, care
must be taken to ensure that the binder systems are compatible. Two types
of reclamation are commonly used, mechanical attrition and thermal.
Wet reclamation has been used for silicate bonded sand. The sand is
crushed to grain size, water washed using mechanical agitation to wash off
the silicate residues, then dried. The process further requires expensive
water treatment to permit safe disposal of the wash water so its use is not
common.
174 Foseco Ferrous Foundryman’s Handbook

(a)
(d)

(b)

(c)

Figure 13.2 Typical designs of self-hardening moulds. (From Foundry Practice

Today and Tomorrow, SCRATA Conference, 1975.) (a) Method of moulding-in steel
tubes for ease of handling boxless moulds. (b) Sockets moulded into boxless moulds
for ease of lifting, roll-over and closing. (c) Steel reinforcement frames for handling
large boxless moulds. (d) Method of locating mould halves and preventing runout.

The difficulty and cost of disposing safely of used chemically bonded

sand has led to the growing use of a combination of mechanical and thermal
treatment. Mechanical attrition is used to remove most of the spent binder.
Depending on the binder system used, 60–80% of the mechanically reclaimed
sand can be rebonded satisfactorily for moulding, with the addition of clean
sand. The remaining 20–40% of the mechanically treated sand may then be
 Resin bonded sand 175

Mould hardening

Mixer
Roll-over and
stripping

Empty box return

pattern clean

Mould
close

Casting
area

Figure 13.3 Foundry layout for self-hardening sand moulds.

thermally treated to remove the residual organic binder, restoring the sand
to a clean condition. This secondarily treated sand can be used to replace
new sand. In some cases, all the used sand is thermally treated.

Mechanical attrition

This is the most commonly practised method because it has the lowest cost.
The steps in the process are:

Lump breaking; large sand lumps must be reduced in size to allow the
removal of metal etc.
Separation of metal from the sand by magnet or screen.
Disintegration of the sand lumps to grain size and mechanical scrubbing
176 Foseco Ferrous Foundryman’s Handbook

to remove as much binder as possible, while avoiding breakage of

grains.
Air classification to remove dust, fines and binder residue.
Cooling the sand to usable temperature.
Addition of new sand to make up losses and maintain the quality of the
reclaimed sand.

Reclamation by attrition relies on the fact that the heat of the casting burns
or chars the resin binder close to the metal. Even at some distance from the
metal, the sand temperature rises enough to embrittle the resin bond. Crushing
the sand to grain size followed by mechanical scrubbing then removes much
of the embrittled or partially burnt binder. The more strongly the sand has
been heated, the more effectively is the sand reclaimed.
Mechanical attrition does not remove all the residual binder from the
sand, so that continued re-use of reclaimed sand results in residual binder
levels increasing until a steady state is reached which is determined by:

the amount of burn-out which occurs during casting and cooling

the effectiveness of the reclamation equipment
the percentage of new sand added
the type of binder used.

The equilibrium level of residue left on the sand is approximately expressed

as:

P= TB
1 – TR
P is the maximum percentage of resin that builds up in the sand (the LOI
of the reclaimed sand)
B is the binder addition (%)
T is the fraction of binder remaining after reclamation
R is the fraction of sand re-used

Example: In a typical furane binder system

B = 1.4% resin + 0.6% catalyst = 2.0%
T = 0.7 (only 30% of the binder residue is removed)
R = 0.90 (90% of reclaimed sand is re-used with 10% new sand)

P= 0.7 × 2.0
1 – (0.7 × 0.9)
= 3.78% (residual binder that builds up on the sand)
This represents an inefficient reclaimer. Ideally P should not exceed 3.0%.
Even with an inefficient reclaimer P = 3% can be achieved by reducing R,
that is, by adding more new sand. For example, reducing R to 0.75 (25%
addition of new sand) reduces P to 2.95%
Regular testing of reclaimed sand for LOI, acid demand, grain size and
 Resin bonded sand 177

temperature is needed, together with regular maintenance of the reclaimer

to ensure that consistent mould quality is achieved.
Binder systems containing inorganic chemicals, e.g. silicate based systems,
alkaline phenolic resins or binder systems containing phosphoric acid are
difficult to reclaim at high percentages because no burn-out of the inorganic
material occurs.
Use of reclaimed sand with high LOI may cause problems due to excessive
fumes at the casting stage, particularly if sulphonic acid catalysed furane
resins are used.

Thermal reclamation

Sand bonded with an entirely organic binder system can be 100% reclaimed
by heating to about 800°C in an oxidising atmosphere to burn off the binder
residues, then cooling and classifying the sand. Thermal reclaimers are usually
gas heated but electric or oil heating can also be used. The steps in the
process are:

Lump breaking
Metal removal
Heating to about 800°C for a certain time in a fluidised bed furnace or
rotary kiln
Cooling the sand, using the extracted heat to preheat the incoming sand
or the combustion air
Classification
Addition of new sand to make up losses in the system.

Thermal reclamation is costly because of the large amount of heat needed

and the relatively expensive equipment. The ever growing cost of sand
disposal, however, leads to its increasing use.

Sand losses
Whatever method of reclamation is used, there is always some loss of sand
so that 100% reclamation can never be achieved. Sand losses include: burn-
on, spillage, inefficiencies in the sand system and the need to remove fines.
Dust losses of around 5% can be expected and total sand losses of up to 10%
may be expected.

Typical usage of sand reclamation

Furane bonded sand

Mechanical attrition allows up to 90% of sand to be re-used. Only sulphonic

178 Foseco Ferrous Foundryman’s Handbook

acid catalysed sand can be reclaimed. Reclaimed sand may have up to 3%

LOI. Binder additions on rebonding can be reduced by 0.15–0.2% (from say,
1.2% on new sand to 1.0%) with a proportionate reduction of catalyst. Nitrogen
and sulphur levels build up and may cause casting defects and excessive
fume if allowed to exceed 0.15% when the sand is rebonded. Low nitrogen
resins must be used.
Thermal treatment at 800°C can be used on the residual sand. Nitrogen
and sulphur are removed with the remaining organic content.

Phenolic urethane bonded sand

Up to 90% of sand can be reclaimed by mechanical attrition. If iron oxide is

added to the sand mix, care must be taken to ensure that the iron content of
the sand does not rise too high or the refractoriness may suffer.
Thermal reclamation can be used on the residual sand as long as the iron
oxide content is low.
Cold box core sand, from core assembly casting processes, can be thermally
reclaimed satisfactorily. If uncontaminated core lumps can be separated
from green sand, they can also be reclaimed.

Alkaline phenolic bonded sand

These binders contain potassium which, if allowed to rise above 0.15% in

the reclaimed sand, will cause unacceptable reduction of work time and
final strength. This limits attrition reclamation to about 70%.
The residue after attrition reclamation can be thermally reclaimed if it is
treated with FENOTEC ADT 1, an anti-fusion additive. This additive prevents
sintering of the sand during thermal treatment and aids removal of potassium
salts.

Resin shell sand

Shell moulding and core sand (silica or zircon) is fully reclaimable by thermal
means. The used sand often has as much as 2% of residual resin left after
casting. Useful heat can be obtained from the combustion of the resin residues
which significantly reduces the quantity of heat that must be supplied. The
quality of the reclaimed sand must be regularly checked for:

Size grading
LOI (the visual appearance of the sand, which should be the colour of
new sand, gives a good indication of LOI)
Inorganic residues, e.g. calcium stearate additions, which are often made
to lubricate shell sand, burn to CaO during reclamation, giving rise to
high acid demand.

Foundries wishing to thermally reclaim shell sand must have equipment to

re-coat the sand for re-use.
 Resin bonded sand 179

Silicate-ester bonded sand

Standard silicate-ester bonded sand can be reclaimed mechanically only at

rather low levels, less than 50% because the build-up of soda in the sand
reduces its refractoriness. The presence of reclaimed sand in the re-bonded
mix reduces the work time and final strength and increases the tendency for
sagging.

VELOSET special silicate ester binder (see Chapter 14)

Up to 90% reclaimed sand can be re-used with the VELOSET system. Shakeout
sand is reduced to grain size in a vibratory crusher which provides the
primary attrition stage. The sand is then dried in a fluidised bed drier.
Secondary attrition takes place next in a hammer mill. The sand is finally
passed through a cooler-classifier ready for re-use. The reclaimed sand is
blended with new sand in the proportion 75 to 25. During the first 10 cycles
of re-use, the sand system stabilises and the bench life of the sand increases
by a factor of up to 2. Also, mould strength should improve, and it is usually
possible to reduce the binder addition level by up to 20% yet still retaining
the same strength as achieved using new sand. Once the process has become
established, it becomes possible to re-use up to 85–90% of the sand (Figs
14.4, 14.5).

Wet reclamation

Some silicate bonded sands are particularly difficult to reclaim mechanically

because Na2O builds up and lowers refractoriness. (VELOSET sand is an
exception.) Thermal reclamation is ineffective, but water washing can be
used. The steps in the process are:

Lump breaking
Metal separation by magnet or screen
Disintegration to grain size
Water wash with mechanical agitation to wash off the silicate residues
Separation of sand and water
Drying of the sand
Agglomeration of the alkaline residues in the water to allow settlement
and separation
Water treatment to permit safe disposal of the water

Wet reclamation is expensive and its use is not common.

180 Foseco Ferrous Foundryman’s Handbook

Self-hardening resin binder systems

Furanes

Foseco products: FUROTEC, ESHANOL Binders.

Principle: Self-setting furane sands use a furane resin and an acid catalyst.
The resins are urea-formaldehyde (UF), phenol-formaldehyde (PF), or UF-
PF resins with additions of furfuryl alcohol (FA). Speed of setting is controlled
by the percentage of acid catalyst used and the strength of the acid.

Sand: Since the resins are acid catalysed, sands should have a low acid
demand, less than 6 ml.

Resin: A wide range of resins is available having different nitrogen content.

The UF base resin contains about 17–18% N; furfuryl alcohol is N-free, so
the N content of a UF-FA resin depends on its FA content. Nitrogen can
cause defects in steel and high strength iron castings, so it is advisable to
use high FA resins (80–95% FA, 3.5 – 1% N) although they are more expensive
than lower FA resins. These resins are particularly useful with sands of high
quality, such as the German Haltern and Frechen sands, which are round-
grained and can be used with low resin additions (<1.0%) so the total nitrogen
on the sand is low.
A range of UF-PF furanes is also available, having low nitrogen content,
they are frequently used with sands of rather lower quality, which require
quite high resin additions (>1.2%). The total nitrogen on the sand can be
kept to a lower level than is possible with UF-FA resins.
Phenol-formaldehyde resin is N-free and a range of totally N-free PF-FA
resins is available, these are used for particularly sensitive castings, or when
sand is reclaimed. Where the presence of phenol in the resin presents a
problem, due to restrictions on used sand disposal, special phenol-free FA-
Resorcinol resins (FA-R resins) can be used. They contain less than 0.5%N
and are used for steel and high quality iron casting. While they are expensive,
they are used at low addition rates.

Catalyst: UF-FA resins can be catalysed with phosphoric acid, which has the
advantage of low odour and no sulphur but reclamation of the sand is
difficult. FA-R, PF-FA, UF-FA with high FA and UF-PF-FA resins are catalysed
with sulphonic acids, either PTSA (para-toluene sulphonic acid), XSA (xylene
sulphonic acid) or BSA (benzene sulphonic acid), the last two are strong
acids and are used when faster setting is required. Sulphonic acid catalysed
resin sand can be reclaimed but the sulphur content of the sand rises and
may cause S defects in ferrous castings particularly in ductile iron.
Environmental problems may also arise due to the SO2 gas formed when
moulds are cast. Mixed organic and inorganic acids may also be used.

Addition rate: Resin: 0.8–1.5% depending on sand quality. Catalyst: 40–60%

 Resin bonded sand 181

of resin, depending on sand temperature and speed of setting required. If

fast setting is required, XSA or BSA catalyst should be used.

Pattern equipment: Wood or resin is commonly used. Metal patterns can be

used but if the pattern is cold, setting may be retarded. FA is a powerful
solvent and will attack most paints, patterns must be clean and preferably
unpainted. Special release agents such as STRIPCOTE AL can be used.

Temperature: The optimum is 20–30°C. Cold sand seriously affects the setting
speed and the final strength.

Strength: Depends on resin type and addition, typically 4000 kPa (600 psi)
compression strength.

Speed of strip: Can be from 5–30 minutes. Short strip times require high
speed mixers and may cause problems of sand build-up on the mixer blades.

Work time/strip time: UF-FA resins are better than PF-FA, the higher the FA
content, the better.

Coatings: Water or spirit based coatings may be used, alcohol based coatings
may cause softening of the mould or core surface and it is advisable to
allow 1/2 –1 hour between stripping and coating.

Environmental: Formaldehyde is released from the mixed sand, so good

extraction is needed around the mixer. When sulphonic acid catalyst is
used, SO2 gas is evolved when moulds are cast, particularly if reclaimed
sand is used.

Reclamation: Attrition reclamation works well and it is often possible to

reduce the resin addition on reclaimed sand. Low N resin and sulphonic
acid catalyst must be used. The LOI of the reclaimed sand must be kept to
less than 3.0%. N and S in the sand should not exceed 0.15% when the sand
is rebonded.

Casting characteristics: N above 0.15% in the sand will cause defects in iron
and steel castings. S above 0.15% in the sand may cause reversion to flake
graphite at the surface of ductile iron castings. Metal penetration can be a
problem if work time is exceeded.

General: Probably the most widely used self-hardening system because of

the easy control of set times, good hot strength, erosion resistance and the
ease of reclamation.
182 Foseco Ferrous Foundryman’s Handbook

Phenolic-isocyanates (phenolic-urethanes)

Foseco product: POLISET Binder.

Principle: The binder is supplied in three parts. Part 1 is a phenolic resin in

an organic solvent. Part 2 is MDI (methylene diphenyl diisocyanate). Part 3
is a liquid amine catalyst. When mixed with sand, the amine causes a reaction
between resin and MDI to occur, forming urethane bonds which rapidly
sets the mixture. The speed of setting is controlled by the type of catalyst
supplied in Part 1.

Sand: The binder is expensive, so good quality sand is needed to keep the
cost of additions down. AFS 50–60 is usually used.

Addition rate: The total addition is typically 0.8% Part 1, 0.5% Part 2, more or
less being used, depending on the sand quality.

Pattern equipment: Wood, resin or metal can be used. Paints must be resistant
to the strong solvents in Part 1 and Part 2.

Temperature: Sand temperature affects cure rate, but not as seriously as other
binders. The optimum temperature is 25–30°C.

Speed of strip: Can be very fast, from 2–15 mins. A high speed mixer is
needed to handle the fast cure rates.

Work-time/Strip-time: Very good.

Strength: Compression strength is typically over 4000 kPa (40 kgf/cm2,

600 psi).

Coatings: If water-based coatings are used, they should be applied immediately

after strip and dried at once. Spirit based coatings should be applied 15–20
minutes after strip and are preferably air-dried.

Casting characteristics: Lustrous carbon defects may occur on ferrous castings.

Part 2 (MDI) contains 11.2% N and gas defects may occur in steel castings.
The binder has rather low initial hot strength, so erosion defects may occur.
All these problems can be reduced by the addition of 1 or 2% of iron oxide
to the sand.

Reclamation: Attrition works well, 90% re-use of sand is possible. If iron

oxide is added to the sand mix, care must be taken to ensure that the iron
content of the sand does not rise too high or the refractoriness may suffer.

Environmental: Solvent release during mixing and compaction may be

troublesome, use good exhaust ventilation. Isocyanates can cause respiratory
problems in sensitised individuals. Once an individual has become sensitised,
 Resin bonded sand 183

exposure, however small, may trigger a reaction. MDI has very low vapour
pressure at ambient temperature, so exposure to the vapour is unlikely to
cause problems, however uncured isocyanate may be present on airborne
sand particles at the mixing station. Good exhaust ventilation is essential.

General: This system has never been as popular in the UK and Europe as in
the USA due to the difficulty of stripping and the low hot strength. Steel
foundries find it useful for cores, where the low hot strength reduces hot
tearing.

Alkaline phenolic resin, ester hardened

Foseco products: FENOTEC, FENOTEC hardener.

Principle: The binder is a low viscosity, highly alkaline phenolic resole resin.
The hardener is a liquid organic ester. Sand is mixed with hardener and
resin, usually in a continuous mixer. The speed of setting is controlled by
the type of ester used.

Sand: Can be used with a wide range of sands including zircon, chromite
and high acid demand sand such as olivine.

Resin addition: 1.2–1.7% depending on the sand quality; 18–25% hardener

based on resin.

Nitrogen content: Very small or zero.

Pattern equipment: Wood or resin, patterns strip well.

Temperature: Low sand temperature slows the cure rate, but special hardeners
are available for cold and warm sand.

Speed of strip: 3 min to 2 hours depending on the grade of hardener used.

The work time/strip time ratio is good.

Strength: 24 h strength: transverse 1600 kPa 16 kgf/cm2 230 psi

tensile 900 9 130
compression 4000 40 600

Coatings: Some iron and steel castings can be made without coatings. Both
water based and spirit based coatings can be used, but some softening-back
of the sand surface is possible. The stripped core or mould should be allowed
to harden fully before applying the coating. Water based coatings should be
dried as quickly as possible. Alcohol based coatings should be fired as soon
as possible after application.
184 Foseco Ferrous Foundryman’s Handbook

Casting characteristics: Good as-cast finish on all metals. Hot tearing and
finning defects are eliminated. No N, S or P defects. Good breakdown,
particularly on low melt point alloys. Widely used for steel castings as well
as iron and aluminium.

Reclamation: FENOTEC binders allow up to 70–90% sand reclamation by

attrition, there is some loss of strength and careful management of the alkali
content of the reclaimed sand is needed. The remainder of the sand can be
thermally reclaimed if it is treated with FENOTEC ADT 1, an anti-fusion
additive which aids potassium removal.

Environmental: Low fume evolution at mixing, casting and knockout stages.

General: Stripping from all pattern types is excellent; this, together with the
good casting finish achieved and the low fume evolution, makes the system
popular for all types of casting but most of all for steel.

Triggered hardening systems

Cores for repetition foundries are usually made using a triggered hardening
system with the mixed sand being blown into the core box. The cores must
be cured in the box until sufficient strength has been achieved to allow
stripping without damage or distortion, usually the core continues to harden
after stripping. Transverse breaking strength or tensile strength are used to
assess the properties of triggered core binder systems since they represent
the properties needed to strip and handle cores. The strength requirements
needed depend on the particular type of core being made. Thin section
cores, such as cylinder head water jacket cores, require high stripping strength
because of their fragile nature. Tensile strengths of 1000–2000 kPa (150–300
psi) are typical, equating roughly to transverse strengths of 1500–3000 kPa.
Final strengths may be higher, but some binder systems are affected by
storage conditions (humidity in particular) and strengths may even fall if
the storage conditions are poor. Surface hardness is also important but difficult
to measure precisely.
Heavy section cores and moulds made by core blowing can be successfully
made with lower strength binders, and factors such as fume released on
casting and compatibility of core residues with the green sand system may
be more important than achieving the very highest strength.

Heat triggered processes

The sand and binder are mixed then introduced into a heated core box or
pattern. The heat activates the catalyst present in the binder system and
cures the binder. Core boxes and patterns must be made of metal, normally
 Resin bonded sand 185

grey cast iron, evenly heated by means of gas burners or electric heating
elements. The working surface of the core box is usually heated to 250°C,
higher temperatures may overcure or burn the binder. Sand is a rather poor
conductor of heat and heat penetration is rather slow (Fig. 13.4), it is therefore
difficult to cure thick sections of sand quickly and fully. When the core is
ejected from the core box, residual heat in the sand continues to penetrate
into the core, promoting deeper cure. To achieve the fastest cure times,
heavy section cores are often made hollow, using heated mandrels or ‘pull-
backs’ to reduce core thickness. Core boxes are usually treated with silicone
release agents to improve the strip. The thermal expansion of the metal core
box must be taken into account when designing core boxes. A cast iron core
box cavity 100 mm long will expand by 0.27 mm when heated from 25 to
250°C. This change becomes significant on large cores.

0.7

16
0.6
50°C
14
0.5
70°C 12

Depth (mm)
Depth (in)

0.4 10
100°C
8
0.3

6
0.2
4

0.1
2

0
0 20 40 60 80 100 120
Heating time (s)

Figure 13.4 Temperature rise in a sand core with one surface in contact with a
heated core box at 250°C (theoretical).

Other methods of applying heat to sand cores have been tried. Microwave
or dielectric heating is difficult because electrically conducting metal core
boxes cannot be used. Certain resins can be used for core boxes but they
pick up heat from the cores and may distort.
Blowing heated air through the core has also been used, but a large
volume of air is needed and the method becomes slow and impractical if
core sections above about 30 mm are to be cured. Approximately 1 kg (800
litres at STP) of heated air is needed to heat 1 kg of sand to curing temperature.
186 Foseco Ferrous Foundryman’s Handbook

Gas triggered systems

Sand and binder are mixed and blown into a core box then a reactive gas is
blown into the core box causing hardening of the binder. Hardening occurs
at room temperature, avoiding the need for heated core boxes. Many of the
gases used to trigger chemical reactions are highly reactive and toxic so that
specially designed gas generators are needed to meter the gas accurately
into the core box. Core boxes must be sealed to contain the gases and often
scrubbers must be used to absorb the exhaust gas to prevent it from
contaminating the atmosphere both inside and outside the foundry. Frequently,
the gassing cycle includes an air purge to remove residual gas from the core
so that it can be handled and stored safely.
Gas cured binder systems often allow very fast curing, even of thick
section cores.

Heat triggered processes

The Shell or Croning process

Principle: Sand is pre-coated with a solid phenolic novolak resin and a catalyst
to form a dry, free-flowing material. The coated sand is blown into a heated
core box or dumped onto a heated pattern plate causing the resin to melt
and then harden. Cores may be solid or they may be hollow with the uncured
sand shaken out of the centre of the core. Shell moulds are normally only
20–25 mm thick.

Sand: The base sand is usually a fine silica sand of AFS 60-95. To achieve the
highest strength with the minimum resin addition, the sand should be pure,
rounded grain and free from surface impurities. Zircon sand is also pre-
coated for special applications.

Resin and catalyst additions: The resins are solid phenolic novolaks which
melt at around 100°C. The catalyst is hexamine, a white powder. Resin
additions are 2.5–4.5% (depending on the application and the strength
required), hexamine is added at 11–14% of the resin content.

Nitrogen content: The resins are N-free but hexamine contains 40% N. Special
low N systems with little or no hexamine are available.

Pre-coating procedure: There are two processes; hot coating and warm coating.
In the hot coating procedure, the sand is heated to around 130°C and charged
into a batch mixer. Solid resin granules are added, the heat from the sand
melts the resin which then coats the sand grains. Aqueous hexamine solution
plus release agents are added, cooling the sand below the melt point of the
resin. The coated sand is broken down to grain size and finally cooled.
 Resin bonded sand 187

In warm coating, the sand is heated to 50°C and charged into a batch
mixer together with hexamine and release agents. Resin in alcohol solution
is added and warm air blown through the mixer to vaporise the solvent,
leaving the sand evenly coated. After breaking down the sand lumps and
cooling, the coated sand is ready for use.
Many foundries purchase ready coated sand from specialist suppliers.

Bench life: Pre-coated sands last indefinitely if stored dry and not exposed to
excessive heat, which may cause clumping.

Core blowing: The free-flowing sand can be blown at low pressure, 250–350
kPa(35–50 psi). Core boxes should be made of cast iron heated to 250°C.
Copper or brass must be avoided since ammonia released during curing
will cause corrosion.

Curing times: Minimum curing time is 90 seconds but 2 minutes is common.

The longer the curing time, the thicker is the shell build-up. Increasing
curing temperature does little to speed up cure rate and runs the risk of
overcuring the sand adjacent to the core box or pattern plate.

Core strength: Surface hardness and strength is high on ejection. A 3.5% resin
content will give a tensile strength of 1400 kPa (200 psi). Storage properties
of cured cores is excellent.

Gas evolution: 4–5 ml/g of gases for each 1% of resin.

Casting characteristics: Core and mould coatings are unnecessary. The surface
finish of castings is excellent. Breakdown is good, particularly with hollow
cores. Accuracy and reproduction of pattern detail is extremely good because
of the free-flowing properties of pre-coated sand. Some distortion, particularly
of moulds, can occur on casting and support of shell moulds by steel shot
or loose sand may be needed for heavy section castings. N defects may
occur on steel castings unless low-N coated sand is used. Excellent surface
finish is achieved on aluminium castings.

Health hazards: Phenol and ammonia are released during curing so good
ventilation is needed at core and mould making machines. Casting into
shell moulds produces unpleasant and hazardous fumes and often shell
moulds are cooled in an enclosed tunnel, allowing fumes to be collected
and exhausted outside the foundry.

Reclamation: When both moulds and cores are made by shell process, the
sand can be reclaimed thermally.

Shell moulding: Shell moulds are made by dumping pre-coated sand onto an
iron pattern plate heated to 240–260°C. After a suitable time, usually about
2 minutes, the mould is overturned, returning the uncured sand to the
188 Foseco Ferrous Foundryman’s Handbook

hopper and leaving a shell mould 20–25 mm thick which is ejected from the
pattern plate. Cores are placed and the two half-moulds are glued together
with hot-melt adhesive (CORFIX).
Shell moulds are usually cast horizontally, unsupported. If cast vertically,
support is usually necessary to prevent mould distortion. Steel shot is
frequently used to support the mould.
KALMIN pouring cups are often used with horizontally poured shell
moulds.

General: The shell process was one of the first synthetic resin foundry processes
to be developed. Although it is slow and rather expensive, it is still widely
used, because of the excellent surface finish and dimensional accuracy of
the castings produced.

Hot box process

Principle: The binder is an aqueous PF-UF or UF-FA resin, the catalyst is an

aqueous solution of ammonium salts, usually chloride and bromide. Sand
is mixed with the liquid resin and catalyst and blown into a heated core box.
The heat liberates acid vapour from the catalyst which triggers the hardening
reaction. Hardening continues after removal of the core from the box causing
the sand to be cured throughout.

Sand: Clean silica sand of AFS 50-60, acid demand should be low. The binder
is aqueous so traces of moisture in the sand do not affect the cure. Low sand
temperatures (below 20°C) slow the cure rate.

Resin addition: 2.0–2.5% depending on sand quality. Catalyst additions are

20–25% of the resin weight.

Nitrogen content: The resin contains 6–12%N. The catalysts are usually
ammonia chloride solutions and also contain N. The total N in the system
is 10–14% depending on the resin used. Low-N or N-free systems are available
but their performance is not as good as the N-containing binders.

Mixing procedure: Batch or continuous mixers can be used, add catalyst first
then the resin.

Bench life: 1–2 hours, mixed sand must be kept in closed containers to prevent
drying out. High ambient temperatures reduce bench life.

Core blowing: High blowing pressure, 650–700 kPa (90–100 psi) is needed to
blow the wet sand into core boxes.

Core boxes: Cast iron or steel, fitted with gas or electric heaters to maintain
 Resin bonded sand 189

the temperature at 250–300°C. Silicone parting agents are usually applied to

improve the strip of the core.

Curing time: Thin section cores cure in 5–10 seconds. As cores increase in
section, curing time must be extended up to about one minute for 50 mm
section. For larger sections still, it is advisable to lighten-out cores with
heated mandrels, otherwise the centre of the core will remain uncured.

Core strength: Surface hardness and strength is high on ejection. Final tensile
strength is 1400–2800 kPa (200–400 psi). Storage properties are good if cores
are kept dry.

Casting characteristics
Ferrous castings: Cores are usually coated to prevent burn-on. Water based
coatings, such as RHEOTEC are used. The hot strength and breakdown is
good, particularly for the UF-PF-FA resins, the higher the PF content, the
worse the breakdown. The rather high nitrogen can cause gas holes and
fissures in iron castings and additions of 1–3% of a coarse grained form of
iron oxide are often used to minimise N-defects. Finely powdered red iron
oxide can also be used but it causes some loss of strength.
Hot box cores made using phenolic resin expand slightly during casting
and this can cause distortion, for example on large water jacket cores. Iron
oxide additions reduce the expansion, so does the use of UF/FA resins.
Aluminium castings: Special hot box resins are available which break
down readily at aluminium casting temperatures.
Copper-base castings: Brass water fittings are commonly made with special
hotbox resin cores.

Health hazards: Formaldehyde is liberated during curing, so coreblowers

and the area around them must be well ventilated and exhausted. Avoid
skin contact with liquid resins and mixed sand.

Reclamation: It is not normal to reclaim hot box cores when they are used in
green sand moulds. Core sand residues in green sand are well tolerated.

General: The hot box process was the first high speed resin coremaking
process. Gas-cured cold box processes have superseded it in many cases,
particularly for thick section cores, but hot box is still used because of its
good surface hardness, strength in thin sections and excellent breakdown
properties.

Warm box process

Principle: The binder is a reactive, high furfuryl alcohol binder. The catalyst
is usually a copper salt of sulphonic acid. Sand, binder and catalyst are
190 Foseco Ferrous Foundryman’s Handbook

mixed and blown into a heated core box. The heat activates the catalyst
which causes the binder to cure.

Sand: Clean silica sand of AFS 50–60 is used. Low acid demand is advisable.

Resin addition: 1.3–1.5% resin and 20% catalyst (based on resin) depending
on sand quality.

Nitrogen content: The resin contains about 2.5% N. The catalyst is N-free.

Mixing procedure: Continuous or batch mixers. Catalyst should be added

first, then resin.

Bench life: Typically 8 hours.

Core blowing: The binder has low viscosity and blow pressures of around
500 kPa (80 psi) may be used.

Core boxes: Cast iron or steel heated to 180–200°C. Use a silicone based
parting agent. A release agent is sometimes added to the sand/binder mix.

Curing time: 10–30 seconds depending on the thickness. Thick section cores
continue to cure after ejection from the core box.

Core strength: Surface hardness and strength is high on ejection. Final tensile
strength can be 3000–4000 kPa (400–600 psi).

Casting characteristics: Gas evolution is low, the low nitrogen content reduces
incidence of gas related defects in ferrous castings. Surface finish of castings
is good with low incidence of veining defects. Breakdown after casting is
good.

Environmental: Emission of formaldehyde is low at the mixing and curing

stages. Avoid skin contact with binder, catalyst and mixed sand.

Reclamation: Core residues in green sand are not harmful.

General: Warm box cores have high strength and resistance to veining and
are used in critical applications, usually for iron castings such as ventilated
brake discs.

Oil sand
Principle: Certain natural oils, such as linseed oil, known as ‘drying oils’,
polymerise and harden when exposed to air and heat. Natural oils can be
chemically modified to accelerate their hardening properties. Silica sand is
 Resin bonded sand 191

mixed with the drying oil, a cereal binder and water. The resulting mixture
is either manually packed or blown into a cold core box. The cereal binder,
or sometimes dextrin, gives some green strength to the core which is then
placed in a shaped tray or ‘drier’ to support it during baking. The baking
hardens the oil and the core becomes rigid and handleable.

Sand: Clean silica sand, AFS 50-60.

Binder addition: 1–2% of drying oil

1–2% pre-gelatinised starch  or 2% dextrin

2–2.5% water  0.5–1% water

Nitrogen content: Zero.

Mixing: Batch mixers are preferred in order to develop the green bond strength,
mixing times may be 3–10 minutes.

Bench life: As long as the mixture does not lose moisture, the bench life is up
to 12 hours.

Core blowing: Oil sand mixtures are sticky and difficult to blow, the highest
blow pressure possible (around 700 kPa, 100 psi) is used. They are frequently
hand-rammed into core boxes.

Core boxes: May be wood, plastic or metal. Wooden boxes require a paint or
varnish to improve the strip. Metal boxes are preferably made of brass or
bronze to aid stripping of the fragile green cores.

Curing: A recirculating air oven is needed since oxygen is necessary to harden

the oil. The temperature is normally 230°C, allowing 1 hour for each 25 mm
section thickness. Burning and consequent friable edges may occur on thin
section cores, if this happens, a lower temperature for a longer time should
be used.

Core strength: Green strength is low so sagging will occur if the cores are not
supported during baking. Correctly baked cores develop tensile strength of
1340 kPa (200 psi).

Casting characteristics: Breakdown after casting is excellent. The gas evolution

is high, particularly if cores are under-baked, venting is often necessary.
Water based coatings can be used. While for most applications, oil sand has
been superseded by synthetic resin processes, it still remains a valuable
process in applications where particularly good breakdown of cores is needed,
for example for high conductivity copper and high silicon iron castings in
which hot tearing is a serious problem.

Health hazards: Unpleasant fumes are emitted during baking and ovens must
192 Foseco Ferrous Foundryman’s Handbook

be ventilated. Some proprietary core oils contain a proportion of mineral

oil, which may be harmful if skin contact is prolonged.

Gas triggered processes

Phenolic-urethane-amine gassed (cold box) process

Foseco products: POLITEC and ESHAMINE cold box resin

Principle: The binder is supplied in two parts. Part 1 is a solvent based

phenolic resin, Part 2 is a polyisocyanate, MDI (methylene di-phenyl di-
isocyanate) is a solvent. The resins are mixed with sand and the mixture
blown into a core box. An amine gas (TEA, triethylamine or DMEA, dimethyl
ethyl amine or DMIA, dimethyl isopropyl amine) is blown into the core,
catalysing the reaction between Part 1 and Part 2 causing almost instant
hardening.

Sand: Clean silica sand of AFS 50–60 is usually used but zircon and chromite
sands can be used. The sand must be dry, more than 0.1% moisture reduces
the bench life of the mixed sand. High pH (high acid demand) also reduces
bench life. Sand temperature is ideally about 25°C, low temperature causes
amine condensation and irregular cure. High temperature causes solvent
loss from the binder, and loss of strength.

Resin addition: Total addition is 0.8–1.5% depending on sand quality. Normally

equal proportions of Part 1 and Part 2 are used.

Nitrogen content: Part 2, the isocyanate, contains 11.2% N.

Mixing procedure: Batch or continuous mixers can be used, add Part 1 first
then Part 2. Do not overmix since the sand may heat and lose solvents.

Bench life: 1–2 hours if the sand is dry.

Core blowing: Use low pressure, 200–300 kPa (30–50 psi), the blowing air
must be dry, use a desiccant drier to reduce water to 50 ppm. Sticking of
sand and resin to the box walls can be a problem due to resin blow-off. The
lowest possible blow pressure should be used. Use of a special release agent
such as STRIPCOTE is advised.

Core boxes: Iron, aluminium, urethane or epoxy resin can be used. Wood is
possible for short runs. Use the minimum vents that will allow good filling,
since reduced venting gives better catalyst gas distribution. The exhaust
vent area should be 70% of the input area to ensure saturation of the core.
Core boxes must be sealed to allow the amine catalyst gas to be collected.
 Resin bonded sand 193

Gas generators: The amine catalysts are volatile, highly flammable liquids.
Special generators are needed to vaporise the amine and entrain it in air or
CO2. The carrier gas should be heated to 30–40°C to ensure vaporisation.
Controlled delivery of amine by pump or timer is desirable.

Gas usage: Approximately 1 ml of amine (liquid) is needed per kilogram of

sand. The amine usage should be less than 10% of Part 1 resin. DMEA is
faster curing than TEA. Typical curing is a short gassing time (1–2 seconds)
followed by a longer (10–20 seconds) air purge to clear residual amine from
the core. Over-gassing is not possible but simply wastes amine.
Typical gassing times:

core wt (kg) total gas+purge (s)

10 10
50 30
150 40

Core strength: Tensile strength immediately after curing is high, 2000 kPa
(300 psi), transverse strength 2700 kPa (400 psi). Storage of cores at high
humidity reduces the strength considerably. The use of water based coatings
can cause loss of strength.

Casting characteristics
Ferrous castings Good surface and strip without coatings
Low hot strength but this can be improved by adding
iron oxide
Tendency to finning or veining
Lustrous carbon formation may cause laps and
elephant skin defects on upper surfaces of castings.
Addition of red iron oxide (0.25–2.0%) or a coarse-
grained form of iron oxide at 1.0–4.0% reduces defects
Breakdown is good
The N content may cause problems on some steel
castings.

Aluminium castings Good surface and strip

Poor breakdown, the resin hardens when heated at
low temperature. Aluminium castings may need heat
treatment at 500°C to remove cores
There are no hydrogen problems.

Reclamation: Excessive contamination of green sand with cold box core residues
can cause problems. Cold box cores and moulds can be thermally reclaimed
if uncontaminated with iron oxide.
194 Foseco Ferrous Foundryman’s Handbook

Environmental: TEA, DMIA and DMEA have objectionable smells even at 3

or 6 ppm. Good, well sealed core boxes, good exhaust and good exhaust gas
scrubbers are necessary. The cores must be well purged or amine will be
released on storage. Liquid amine is highly flammable, treat like petrol.
Air/amine mixtures may be explosive. MDI (Part 2) acts as a respiratory
irritant and may cause asthmatic symptoms but it has low volatility at
ambient temperature, and is not normally a problem. Avoid skin contact
with resins or mixed sand.

General: In spite of its environmental and other problems, the cold box
process is so fast and produces such strong cores that it is the most widely
used gas triggered process for high volume core production. Good engineering
has enabled the environmental problems to be overcome. Reduced free-
phenol resins are being produced to assist with sand disposal problems.

The ESHAMINE Plus process

This patented process is a development of the amine gassed phenolic-urethane
cold box process which overcomes the problems of using volatile liquid
amine catalysts. The liquid amines TEA, DMEA and DMIA must be vaporised
to function effectively. Under cold ambient conditions, amine condensation
in the gas supply lines or in the core box itself may occur.
In the ESHAMINE Plus process, a gaseous amine, TMA (trimethylamine)
is used as catalyst. TMA is the most reactive of the tertiary amine family and
has a boiling point of 3°C (compared to 89°C for TEA, 65°C for DMIA and
35°C for DMEA). Being a gas at ambient temperature simplifies the design
of gassing equipment, reduces gas consumption and improves productivity.
Gassing times are reduced by up to 78% and purge times by up to 54%
compared to DMEA, DMIA or TEA catalysed cores. These reductions have
been found to have a significant impact on the overall cycle time of the core
machine, raising core output by 30%. A further benefit of the reduced gas
usage is the lower cost of amine scrubber maintenance.

ECOLOTEC process (alkaline phenolic resin

gassed with CO2)
Foseco product: ECOLOTEC resin

Principle: ECOLOTEC resin is an alkaline phenol-formaldehyde resin

containing a coupling agent. The resin is mixed with sand and the mixture
blown into a core box. CO2 gas is passed through the mixture, lowering the
pH and activating the coupling agent which causes crosslinking and hardening
of the resin. Strength continues to develop after the core is ejected as further
crosslinking occurs and moisture dries out.

Sand: Clean silica sand of AFS 50-60 is used. The sand should be neutral (pH
 Resin bonded sand 195

7) with low acid demand. Zircon and chromite sands can be used, but
olivine sand is not suitable. Temperature is ideally 15–30°C.

Resin addition: 2.0–2.5% depending on the sand grade.

Nitrogen content: Zero.

Mixing procedure: Batch or continuous.

Bench life: Curing occurs only by reaction with CO2. The CO2 in the air will
cause a hardened crust on the surface of the mixed sand, but the soft sand
underneath can be used. Keep the mixed sand covered. The bench life is
usually 5 hours at 15°C reducing to 1 hour at 30°C.

Core blowing: Blow pressure 400–550 kPa (60–80 psi) is needed.

Core boxes: Wood, metal or plastic. Clean them once per shift and use SEPAROL
or a silicone based release agent.

Gassing: Cores are blown at 400–550 kPa (60–80 psi) then gassed for 20–40
seconds at 100–300 litres/minute. CO2 consumption is about 2% (based on
weight of sand). Gas should not be forced through the core box at high
velocity since the gas must react with the binder. No purge is necessary
before extracting the core.

Core strength
as gassed
(kPa) (kgf/cm2) (psi)
Transverse 800–1000 8–10 110–140

Transverse strength doubles after one hour. Cores should be stored in dry
conditions.

Casting characteristics: Steel, iron, copper based and light alloy castings can
be made. ECOLOTEC is free from N, S and P. Finning and lustrous carbon
defects do not occur. Breakdown after casting is good.

Coating: Water or solvent based coatings can be used without affecting the
strength. Methanol coatings should be avoided.

Environmental: Gassed cores have no odour. Fume is low at mixing, casting

and knockout.

Reclamation: Not normally practised when cores are used in green sand
moulds. The inflow of alkaline salts into green sand can result in strong
activation of the bentonite clay, a move to a partially activated bentonite
will counteract the effect.
196 Foseco Ferrous Foundryman’s Handbook

General: Tensile strength is not as high as the amine gassed isocyanate process
but the excellent casting properties, freedom from nitrogen, lustrous carbon
and finning defects and above all its environmental friendliness make
ECOLOTEC an attractive process particularly for larger cores and moulds
for iron steel and aluminium castings.

The SO2 process

Principle: Sand is mixed with a furane polymer resin and an organic peroxide,
the mixture is blown into the core box and hardened by passing sulphur
dioxide gas through the compacted sand. The SO2 reacts with the peroxide
forming SO3 and then H2SO4 which hardens the resin binder.

Sand: Clean silica sand of AFS 50-60. Other sands may be used if the acid
demand value is low. The temperature should be around 25°C, low
temperature slows the hardening reaction.

Resin addition: Typically 1.2–1.4% resin, 25–60% (based on resin) of MEKP

(methyl ethyl ketone peroxide).

Nitrogen content: Zero.

Mixing: Batch or continuous mixers, add resin first then peroxide. Do not
overmix since the sand may heat and reduce bench life.

Bench life: Up to 24 hours.

Core blowing: Blowing pressure 500–700 kPa (80–100 psi).

Core boxes: Cast iron, aluminium, plastics or wood can be used. A build up
of resin film occurs on the core box after prolonged use, so regular cleaning
is needed by blasting with glass beads or cleaning with dilute acetic acid.

Curing: The SO2 gas is generated from a cylinder of liquid SO2 fitted with a
heated vaporiser. The gas generator must also be fitted with an air purge
system so that the core can be cleared of SO2 before ejection. SO2 is highly
corrosive and pipework must be stainless steel, PTFE or nylon. For large
cores, a separate gassing chamber may be used in which the chamber pressure
is first reduced then SO2 is injected and finally the chamber is purged.

Gas usage: 2 g (ml) liquid SO2 is needed per kg of sand. Cores are normally
gassed for 1–2 seconds followed by 10–15 seconds air purge. Overgassing is
not possible.

Core strength: Tensile strength is 1250 kPa (180 psi) after 6 hours. Storage
properties are good.
 Resin bonded sand 197

Casting characteristics: Coatings are not normally necessary and surface quality
of castings is good. Breakdown of cores is excellent with ferrous castings. It
is also good with aluminium castings, better than cold box cores, and this
has proved to be one of the best applications. The sulphur catalyst may
cause some metallurgical problems on the surface of ductile iron castings.

Reclamation: Thermal reclamation is possible. Mechanically reclaimed sand

should not be used for SO2 core making but may be used with furane self-
setting sand.

Environmental: SO2 has an objectionable smell and is an irritant gas. Core

boxes must be sealed and exhaust gases must be collected and scrubbed
with sodium hydroxide solution. Cores must be well purged to avoid gas
release during storage.
MEKP is a strongly oxidising liquid and may ignite on contact with
organic materials. Observe manufacturer’s recommendations carefully. Avoid
skin contact with resin and mixed sand.

SO2 cured epoxy resin

Principle: Modified epoxy/acrylic resins are mixed with an organic peroxide,

the mixture is blown into the core box and hardened by passing sulphur
dioxide gas through the compacted sand. The SO2 reacts with the peroxide
forming SO3 and then H2SO4 which hardens the resin binder.

Sand: Clean silica sand of AFS 50-60. Other sands may be used if the acid
demand value is low. The temperature should be around 25°C, low
temperature slows the hardening reaction.

Resin addition: Typically 1.2–1.4% resin, 25–60% (based on resin) of MEKP

(methyl ethyl ketone peroxide).

Nitrogen content: Zero.

Other details: Similar to SO2/furane process. Bench life is up to 24 hours.

Ester-cured alkaline phenolic system

Foseco product: FENOTEC resin

Principle: The resin is an alkaline phenolic resin (essentially the same as the
self-hardening resins of this type). Sand is mixed with the resin and blown
or manually packed into a corebox. A vaporised ester, methyl formate, is
passed through the sand, hardening the binder.
198 Foseco Ferrous Foundryman’s Handbook

Sand: Highest strengths are achieved with a clean, high silica sand of AFS
50-60. Sand temperature should be between 15 and 30°C.

Resin addition: Typically 1.5% total.

Nitrogen content: Zero, sulphur is also zero.

Mixing procedure: Batch or continuous.

Bench life: Curing only occurs by reaction with the ester-hardener, so the
bench life is long, 2–4 hours.

Core blowing: Blow pressure 350–500 kPa (50–80 psi) is needed.

Core boxes: Wood, plastic or metal. Clean once per shift and use SEPAROL or
a silicone based release agent.

Gassing: Methyl formate is a colourless, highly flammable liquid, boiling at

32°C. It has low odour. A specially designed vaporiser must be used to
generate the methyl formate vapour. Usage of methyl formate is about 20–
30% of the resin weight. Core boxes and gassing heads should be sealed
correctly and the venting of the core box designed to give a slight back
pressure so that the curing vapour is held for long enough for the reaction
to take place.

Core strength: Compression strengths of 5000 kPa (700 psi) are possible.
Tensile and transverse strengths are not available, they are not as high as
phenolic isocyanate resins.

Casting characteristics: Finning and lustrous carbon defects are absent. Good
high temperature erosion resistance and good breakdown. Mostly used for
steel and iron castings.

Environmental: Low odour, but methyl formate is flammable and care is

needed.

Reclamation: Reclamation by attrition is possible but the reclaimed sand is

best re-used as a self-hardening alkaline-phenolic sand where strengths are
not as critical. Core sand residues entering green sand causes no problems.

General: Strengths are relatively low so the process has mainly been used for
rather thick section moulds and cores where high handling strength is not
necessary.

Review of resin coremaking processes

The heat activated resin coremaking processes, Croning/Shell and hot box
 Resin bonded sand 199

were developed in the 1950s and 1960s and rapidly supplanted the older oil
sand coremaking process. The attraction was speed of cure and the fact that
cores could be cured in the box, eliminating the dimensional problems of oil
sand cores. By the mid-1970s the amine-urethane cold box process was
becoming firmly established, bringing great benefits of speed as well as an
environmental burden. Since 1976 there have been many further developments
in the amine-urethane systems and in other cold gas hardened core binder
systems.
Reasons for this intense interest include:

The flexibility of gas-hardened processes which makes them suitable for

mass production of cores with automatic equipment
Fast curing of cores in the box by controlled injection of reactive gases
High strengths on ejection, often 80% of the final strength
Dimensional accuracy resulting from cold curing
Availability of an extensive range of specially designed core-making
equipment
Energy saving through cold operation
Rapid tooling changes made easy by low tooling temperatures

By 1990 the majority of cores were produced by the cold box, amine isocyanate
process (Table 13.1).

Table 13.1 Coremaking processes used in 48

automobile foundries in Germany, 1991

Amine cold box 44

Hot box 10
Shell/Croning 9
CO2 – silicate 3

Tables 13.2a and b summarise the production features of the many

coremaking processes in use.
Figure 13.5 shows how the use of chemical binders has changed in the
USA over the last 40 years.

Shell sand is declining slightly, holding up remarkably for such an old

process.
Core oil has declined in the face of hot box and cold box.
Silicates received a boost in 1980, due to the introduction of ester-hardening,
but have declined slowly since.
Hot box has declined due to the rise in use of cold box phenolic urethane
core binders, but still has a significant place.
Acid cured furanes have held steady since the maximum in 1980.
Urethane no-bakes are popular in the US, although their use in Europe is
limited.
New binders are predicted to increase substantially by the year 2000.
Table 13.2a Production features of core binders

Process Core-box Curing Typical Flowability Breakdown Integration

requirements binder (%) into
green sand

Shell Metal, heated 260° 2–5 1 2 2

Hot box Metal, heated 230°C 1.2–2.5 2 1 2
Warm-box Metal, heated 150–180°C 1.2–2.0 2 1 2
Amine-urethane Any material, vented, sealed Amine gas 0.9–2.0 1 1 2
SO2-epoxy Any material, vented, sealed SO2 gas and peroxide 1.0–1.2 1 1 2
SO2-furane Any material, vented, sealed SO2 gas 1.0–1.5 1 1 2
Methyl-formate Any material, vented, sealed Methyl formate gas 1.4–1.6 1 2 1
Alkaline phenolic
CO2-silicate Any, vented CO2 gas 2.5–3.5 3 3 1
ECOLOTEC Any, vented CO2 gas 2.0–2.5 1 2 1

1 Good – unlikely to restrict application

2 Satisfactory – suitable for most applications
3 Restrictions – may limit applications
Table 13.2b Properties of core binders

Process Strength Gas Typical Typical Minimum Pin-hole Veining Lustrous

evolution bench strip strip to tendency carbon
properties life time pour time

Shell 1 3 Indefinite 2 mins 30 min. 3 1 1

Hot box 1 2 4–6 hrs 1/2–1 min. 1/2–2hrs 3 2 1
Warm box 1 1 4–8 hrs 1/2–1 min. 0–2 hrs 3 1 1
Amine-urethane 1 1 1–6 hrs 10–30 sec. 0–1 hr 2 2 2
SO2-epoxy 1 2 24 hrs 10–30 sec. Instant 1 1 1
SO2-furane 2 2 24 hrs 10–30 sec. Instant 2 2 2
CO2-silicate 3 1 2–3 hrs 10–60 sec. 1 hr 1 1 1
ECOLOTEC 2 1 3–4 hrs 5–30 sec. Instant 1 1 1

1 Good – unlikely to restrict application

2 Satisfactory – suitable for most applications
3 Restrictions – may limit applications
202 Foseco Ferrous Foundryman’s Handbook

U.S. foundry industry consumption of sand binder by system

U.S. sand binder consumption, Ib × 106

300

200
New binders*
Phenolic urethane cold box
Phenolic urethane no bake
Acid cured furan and phenolic no
100 bake
Hot box
Silicate CO2 and no bake
Core oil
Shell
0 *Includes phenolic/ester, warm box,
1960 1970 1980 1990 2000 polyol urethane, and SO2 cured
Year
systems.

Figure 13.5 US foundry industry consumption of sand binder by system. (From

Foundry Management & Technology, Jan. 1995 p. D-5).

Exhaust
stacks

Exhaust Reverse
fans jet
filters

Fume extraction
systems

Oxidiser
exhaust
stack

Reverse
jet filter

Thermal
Machine oxidiser
enclosures
(11 off)
Amine
collection
system

Figure 13.6 Collection and control of amine gas in a cold box core shop.

The environmental difficulties of the amine-isocyanate process have been

solved, at a cost, by engineering. Figure 13.6 shows how a large coreshop
collects and disposes of amine gas. Smaller foundries find the problems of
handling amine gas difficult so there is a need for a more environmentally
 Resin bonded sand 203

friendly process. Up to the present, only the CO2-hardened ECOLOTEC

system meets the requirements.
A growing concern of all users of chemical binders is that of disposing of
used sand. All the developed countries impose severe restrictions on the
materials that can be safely dumped on landfill sites. Pollution of water
supplies by phenol and other leachable chemicals is a growing concern, so
all resin manufacturers are reducing the free phenol content of their products.
The use of sand reclamation, both mechanical and thermal is being increasingly
used by foundries to reduce the quantity of sand that must be disposed to
waste sites.
Chapter 14
Sodium silicate bonded sand

Sodium silicate is a water soluble glass available from suppliers in a wide

range of types specified by the silica (SiO2), soda (Na2O) and water content.
Manufacturer's data sheets specify the ‘weight ratio’ of silica to soda, the
percentage of water and the viscosity. For foundry use, sodium silicates
with ratios between 2 and 3 and water content around 56% are usually
used.
Sodium silicate can be hardened in a number of ways: by adding weak
acids (CO2 gas or organic esters), by adding various powders (di-calcium
silicate, anhydrite etc.) or by removing water. CO2 gas and liquid ester
hardeners are the most widely used of the silicate processes. Silicate binders
have no smell and few health hazards, but the bond strength is not as high
as that of resin binders and, being an inorganic bond, it does not burn out
with heat, so breakdown after casting can be a problem. For this reason,
various organic breakdown additives are often incorporated with the liquid
silicate or added during sand mixing.
Typical data for foundry grade sodium silicate:

Wt Na2O SiO2 H2O s.g. [Link] litres/

ratio (%) (%) (%) at 20°C tonne

Low ratio 2.0 15.2 30.4 54.4 1.56 850 641

Medium ratio 2.40 12.7 30.8 56.5 1.50 310 668
High ratio 2.85 11.2 31.9 56.9 1.48 500 677

CO2 silicate process (basic process)

Principle: Sand is mixed with sodium silicate and the mixture blown or
hand-rammed into a core box or around a pattern. Carbon dioxide gas is
passed through the compacted sand to harden the binder.

Sand: Silica sand of AFS 50-60 is usually used, the process is quite tolerant
of impurities and alkaline sand such as olivine can be used. Sand should be
clay free. Low temperature retards hardening, the sand temperature should
be above 15°C.

Binder addition: About 3.0–3.5% silicate addition is usually used. High ratio
 Sodium silicate bonded sand 205

silicates give short gassing times, low gas consumption and improved post-
casting breakdown, but high ratio silicate is easily over-gassed, resulting in
reduced core strength and poor core storage properties. Low ratio binders
give better core storage, but gassing times are longer and breakdown is not
as good. Low ratio silicates (2.0–2.2) are usually used.

Mixing: Batch or continuous mixers are used, avoid excessive mixing time
which may heat the sand and cause loss of water.

Bench life: Several hours if drying is prevented by storing in a closed container

or covering with a damp cloth.

Core boxes: Any material; wood, resin or metal may be used but the boxes
must be painted to prevent sticking. Polyurethane or alkyd paints are used
followed by the application of a wax or silicone-wax polish, or STRIPCOTE
AL.

Blowing: High pressure is needed, 650–700 kPa (90–100 psi).

Curing: The correct volume of gas is needed. Undergassing does not achieve
optimum strength. Overgassing leads to reduction of strength on standing.
To ensure correct hardening, a flow controller and timer should be used. If
cores are to be stored for some time before use, it is preferable to undergas.
Higher strength will then be developed on standing. Gassing times are
typically 10–60 seconds, depending on the size of the core.

CO2 gas consumption per kg of sodium silicate (good quality silica sand
AFS 50)

Silicate ratio Time to reach 700 kPa litres kg CO2

(100 psi) compression CO2
strength at 5 l/min

2.0 52 s 300 0.60

2.2 48 275 0.55
2.4 26 150 0.30
2.5 23 130 0.26

Core strength:

as-gassed after 6 h

Tensile 350 kPa 50 psi 1000 kPa 150 psi

Compressive 2100 300 4200 600

Core storage: Overgassing causes cores to become friable during storage

over 24 hours. Careful control of gassing is essential (Fig. 14.1).
206 Foseco Ferrous Foundryman’s Handbook

1 (1) Ratio SiO2 : Na2O 2.0:1

40 (2) 2.2:1
(3) 2.4:1
Compression strength (kg/cm2) (4) 2.5:1
30 2

20
3

10 4

0 60 120
Gassing time (seconds)

Figure 14.1 The effect of 24 hour storage of cores gassed for different times as a
function of silicate ratio.

Casting characteristics: Cores require coating to produce good surface finish

on ferrous castings. Spirit based coatings should be used because water can
soften the silicate bond. No metallurgical problems occur with ferrous or
non-ferrous castings, the binder contains no nitrogen or sulphur. Breakdown
after casting is poor (with ferrous castings) because silicate fuses with sand
above 400°C.

Reclamation: Reclamation is difficult because the binder does not burn out
during casting.

Health hazards: There is very little fume or smell during core manufacture,
storage or casting. Gloves should be worn when handling silicates or mixed
sand.

General comments: The CO2 process was the first of the gas-hardened processes
to be introduced into the foundry and is still one of the easiest and cleanest
processes to use. Foseco and other suppliers have made important
improvements to the process while still retaining its ease of use and excellent
environmental characteristics.

Gassing CO2 cores and moulds

The core boxes are prepared by coating with a suitable parting agent, silicone
waxes can be used but the best results are obtained with STRIPCOTE AL, an
aluminised self-drying liquid.
 Sodium silicate bonded sand 207

The simplest gassing techniques are based either on a simple tube probe
fitted through a backing plate, or a hood sealed to the open side of the box
with a rubber strip (Fig. 14.2). Gas at a pressure of about 70–140 kPa (10–20
psi) is passed until the sand is felt to have hardened or some white crystals
appear (indicating overgassing). Venting of the boxes and patterns to prevent
dead spots is essential. It is usual to rap the box before gassing to make
stripping easier.
Gas in Gas in
Wooden pad
Wooden pad

Metal tube
Single probe

Multiple probe

Gas in
Line coupler Hood
Gas in

Pad Hood

Figure 14.2 CO2-silicate process, methods of gas application.

Control of gassing is improved by incorporating gas heaters (to prevent

freezing of the valves), flow meters and timers so that automatic control of
the volume of gas passed is possible. Not only does this practice economise
on expensive CO2, but the best strength and core-storage life will only be
achieved if the correct amount of gas is used. It is important that overgassing
is avoided, any error should always be towards undergassing. The suppliers
of CO2 gas will advise on the best equipment to use.

Improvements to the CO2 silicate process

Foseco products: CARSIL  sodium silicate blended with special

SOLOSIL  additions
DEXIL breakdown agent

Principle: The main drawbacks of the basic CO2 process are:

poor breakdown of the bond after casting

poor core storage properties
rather low tensile strength
208 Foseco Ferrous Foundryman’s Handbook

These properties can be greatly improved by special additives, while retaining

the simplicity and user friendliness of the CO2 process.

The CARSIL and SOLOSIL range: These products are a range of sodium
silicate based binders for the CO2 process. They may be simple sodium
silicates which can be used with DEXIL breakdown agent if required, or
they may be ‘one-shot’ products which incorporate a breakdown agent, or
they may (like SOLOSIL) incorporate special additives to improve bond
strength as well as breakdown.
Binders containing high levels of breakdown additives give improved
post-casting breakdown but the maximum as-gassed strength is reduced
and core storage properties are likely to be impaired. The selection of an
optimum binder for a given application is therefore almost always a
compromise. The requirement for high production rates and high as-gassed
strength must be balanced against core storage properties and the need for
good breakdown. The range of binders includes some which are suitable
only for the CO2 process, some which are suitable for self-setting applications
and some which can be used for both processes.
The commonly used breakdown agents are organic materials which burn
out under the effect of the heat of the casting. While solid breakdown agents
such as dextrose monohydrate, wood flour, coal dust and graphite can be
used, powder materials are not easy to add consistently to sand in a continuous
mixer. Liquid breakdown agents are easier to handle, they usually consist of
soluble carbohydrates. The best improve gassing speed without loss of
strength. Some are also resistant to moisture pick-up and their use has
increased the storage life of high-ratio silicate bonded cores.
Sucrose is the only common carbohydrate soluble in sodium silicate without
a chemical reaction. It is readily soluble up to 25% and many sugar or
molasses-based binders are available. Use of sucrose increases gassing speed
but reduces maximum strength and storage properties. Nevertheless silicates
containing sugar are the most popular CO2 binders because of the convenience
of a binder in the form of a single liquid. Molasses can be used as a low cost
alternative to sugar, but it is subject to fermentation on storage. The Foseco
CARSIL range of silicate binders is based on sugar. Some are designed for
use with CO2, others for self-setting (SS) with ester hardeners. Some can be
used for both processes.

The CARSIL range of silicate binders

The extent to which a core will break down after casting varies depending
on the type of metal cast. Low temperature alloys such as aluminium do
not inject enough heat into the sand to burn out the breakdown agent
fully. Indeed, the low temperature heating may even strengthen the core.
In such cases it is useful to add additional breakdown agents such as
DEXIL.
 Sodium silicate bonded sand 209

Product Ratio Additive CO2/SS Comments

CARSIL 100 2.5:1 Sugar CO2/SS Higher ratio for faster gassing,
take care not to overgas. Can be
used with ester hardeners.
CARSIL 513 2.4:1 Sugar CO2/SS Low viscosity binder for easy
mixing in continuous mixers.
Moulds and cores.
CARSIL 520 2.0:1 High sugar CO2 High breakdown, low viscosity.
CARSIL 540 2.2:1 Low sugar CO2 Suitable for moulds or cores.
CARSIL 567 2.2:1 High sugar CO2 High breakdown, good for AI
casting.

Note: Some of the CARSIL binders were formerly known as GASBINDA binders in the UK.

DEXIL 34BNF is a powder additive developed for use with light alloys.
It also acts as a binder extender so reducing the silicate requirement. The
application rate is 0.5–1.5%. It should be added to the sand and pre-dispersed
before adding the silicate.
DEXIL 60 is a pumpable organic liquid which is particularly suitable for
use with continuous mixers.

SOLOSIL
SOLOSIL was developed to improve on the performance of silicates containing
sugar based additives. SOLOSIL is a complex one-shot sodium silicate binder
for the CO2 gassed process. It contains a high level of breakdown agent/co-
binder and offers a combination of high strength and rapid gassing with
good core storage properties and excellent post-casting breakdown.
The binder is best used with good quality silica sand. Addition levels of
3.0–4.5% are used depending on the application. To take full advantage of
the high reactivity, an automatic gassing system incorporating a vaporiser,
pressure regulator, flow controller and gassing timer is advisable. The high
rate of strength development is shown in Fig. 14.3. While the transverse and
tensile strength developed by SOLOSIL binders are still somewhat lower
than some organic resin binders, SOLOSIL generally proves more cost effective
and overcomes problems of poor hot strength, veining and finning, gas
pinholing and fume on casting which occur with some resin binders.

Self-setting sodium silicate processes

The first self-setting process used powder hardeners. The Nishiyama process
used finely ground ferrosilicon powder which reacts with sodium silicate
generating heat and forming a very strong bond. The reaction also generates
hydrogen which is dangerous. Other powder hardeners (which do not evolve
dangerous gases) include di-calcium silicate, certain cements (such as blast
210 Foseco Ferrous Foundryman’s Handbook

20
3.5% Solosil

Compression strength (kg/cm2)

15

10
3.5%
Conventional
5 silicate

40 80 120
Gassing time (seconds)

Figure 14.3 Strength development of SOLOSIL compared with a conventional

sugar/silicate system

furnace cement and sulphate resisting cement) and anhydrite. However, all
powder hardeners are difficult to add uniformly to sand in continuous
mixers, and their reactivity is difficult to control, since particle size and the
age after grinding affect the reactivity of the powder. When liquid hardeners
based on organic esters were introduced, the use of powder hardeners was
largely discontinued.

Ester silicate process

Foseco products: CARSIL sodium silicate binders

CARSET ester hardener
VELOSET special ester for very rapid setting

Principle: Sand is mixed with a suitable grade of sodium silicate, often

incorporating a breakdown agent, together with 10–12% (based on silicate)
of liquid organic ester hardener. The acid ester reacts with and gels the
sodium silicate, hardening the sand. The speed of hardening is controlled
by the type of ester used.

Sand: Dry silica sand of AFS 45-60 is usually used. As with all silicate processes,
the quality and purity of the sand is not critical, alkaline sand such as
olivine can be used. Fines should be at a low level. Sand temperature should
be above 15°C, low temperature slows the hardening.

Additions: Sodium silicates with ratios between 2.2 and 2.8 are suitable, the
higher the ratio, the faster the set. Silicates containing breakdown agents
are usually used, additions between 2.5 and 3.5% are used depending on
the sand grade. The ester hardener is commonly:
 Sodium silicate bonded sand 211

glycerol diacetate fast cure

ethylene glycol diacetate medium cure
glycerol triacetate slow cure

Proprietory hardeners may be blends of the above with other esters. The
addition level is 10–12% of the silicate.

Pattern equipment: Wood, resin or metal patterns can be used. Core boxes
and patterns should be coated with polyurethane or alkyd paint followed
by application of wax polish. STRIPCOTE parting agent may also be used.

Mixing: Continuous mixers are usually used, if batch mixers are used, the
ester hardener should be mixed with the sand before adding the silicate.

Speed of strip: 20–120 minutes is common with normal ester hardeners.

Attempts to achieve faster setting may result in lower strength moulds
because the work time becomes short. With certain esters there is a tendency
for core and mould distortion due to sagging if stripping occurs too early.
Faster setting can be achieved by using the special VELOSET hardener.

Strength: The final strength achieved is:

Tensile 700 kPa (100 psi)

Compression 2000–5000 kPa (300–700 psi)

Coatings: Spirit based coatings should be used.

Casting characteristics: No metallurgical problems arise with ferrous or non-

ferrous castings. Breakdown is poor unless a silicate incorporating a
breakdown agent is used.

Reclamation: As with all silicate processes, burnout of the bond does not
occur during casting and attrition does not remove all the silicate residue so
that build-up occurs in the reclaimed sand, reducing refractoriness and
leading to loss of control of work time and hardening speed. The VELOSET
system has been specially developed to permit reclamation (see below).

Environment: Silicate and ester have little smell and evolve little fume on
casting. Silicates are caustic so skin and eye protection is needed while
handling mixed sand.

CARSET 500 Hardeners: These are blends of organic esters formulated to

give a wide range of setting speeds when used with sodium silicates,
particularly the CARSIL series of silicates which incorporate a breakdown
agent. For the best results, the silicate addition should be kept as low as
possible in relation to the sand quality and the CARSET hardener maintained
212 Foseco Ferrous Foundryman’s Handbook

at 10% by weight of the silicate level. The speed of set is dependent on the
sand temperature, silicate ratio and grade of CARSET hardener used.

The CARSET 500 series of hardeners

Gel times (minutes) at 20°C

CARSET 500 CARSIL 540 CARSIL 513 CARSIL 100
series 2.2 ratio 2.4 ratio 2.5 ratio

500 8 7 5
511 9 8 6
522 13 12 8
533 19 15 9
544 105 53 21
555 – – 90

Note: The gel time is the time taken for gelling to occur when silicate liquid is mixed with an
appropriate amount of setting agent. The setting times may not be repeated exactly when sand
is present, due to the possibility of impurities, but it does provide a useful guide.

VELOSET hardeners: The VELOSET range is a series of advanced ester

hardeners for the self-setting silicate process. They have been designed to
give very rapid setting speed with a high strength, excellent through-cure
and a high resistance to sagging. Used in the VELOSET Sand Reclamation
Process, they provide the only ester silicate process in which the sand can
be reclaimed by a simple dry attrition process and reused at high levels
equal to those typical of resin bonded sands.

Additions: There are three grades of VELOSET hardener. VELOSET 1, 2 and

3 Binders of ratio 2.2–2.6 are used, lower ratios give inferior strength while
if higher ratios are used the bench life becomes too short. The bench life
obtained is independent of addition level. The level is usually 10–12% based
on the binder. If the sand is to be reclaimed, the addition level of 11% should
not be exceeded.

Bench life (minutes) at 20°C

CARSIL ratio VELOSET grade

1 2 3

2.2 10 7 4
2.4 7 4 2
2.6 4 2 1

When a choice is possible, always use the highest ratio CARSIL binder and
the slowest grade of VELOSET hardener. This provides optimum strength
development.
 Sodium silicate bonded sand 213

Mixer: Since VELOSET is rapid setting, it is preferable to use a continuous

mixer.

VELOSET sand reclamation process: With the conventional ester silicate process,
dry attrition reclamation has occasionally been practised but the level of
sand reuse is rarely more than 50%, which hardly justifies the capital
investment involved. With the VELOSET system, up to 90% reuse of sand is
possible using mechanical attrition.
The process stages are:

Crushing the sand to grain size

Drying
Attrition
Classification
Cooling

The reclaimed sand is blended with new sand in the proportion 75 to 25.
During the first 10 cycles of reuse, the sand system stabilises and the bench
life of the sand increases by a factor of up to 2. Also, mould strength should
improve, and it is usually possible to reduce the binder addition level by up
to 20% yet still retaining the same strength as achieved using new sand.
Once the process has become established, it may become possible to reuse
up to 85–90% of the sand (Figs 14.4 and 14.5). See also p. 179.

Adhesives and sealants

It is often necessary to joint cores together to form assemblies, or to glue
cores to moulds before closing the mould. A range of CORFIX adhesives is
available:

CORFIX Type Set time Temp (°C) Remarks

grade

4 Stove 30 180–220 High viscosity gap filling

hardening
8 Air hardening slow ambient For CO2 and self-set silicate
21 Air hardening fast ambient Any cold core
25 Hot melt open time 140–180 Core assembly at high rates,
15–120 s shell process

CORSEAL sealants
This is a group of core sealing or mudding compounds for filling out joint
lines, cracks and minor blemishes in cores. CORSEAL is available in two
forms.
214 Foseco Ferrous Foundryman’s Handbook

30

VELOSET 1

Sand: 70% reclaimed

20 30% New Chelford 50
Bench life (minutes)

Binder: 3.5% : 2.2 ratio silicate

Hardener: 0.35%

10

Conventional ester system

0
0 2 4 6 8 10
Re-use cycle

Figure 14.4 VELOSET reclamation, showing the variation in bench life after repeated
use of reclaimed sand, compared with conventional ester process.

p.s.i. kg/cm2
600
40

VELOSET 1

30
Compression strength

400
Sand: 70% reclaimed
30% New Chelford 50
20
Binder: 3.5% : 2.2 ratio silicate
Hardener: 0.35%

200
10
Conventional ester system

0
2 4 6 8 10
Re-use cycle

Figure 14.5 VELOSET reclamation, ultimate strength characteristics of reclaimed

sand, compared with conventional ester process.
 Sodium silicate bonded sand 215

CORSEAL 2 is a powder which is mixed with water to form a thick paste

(4 parts product to 3 parts water). The paste is applied by spatula or trowel
(or fingers) and allowed to dry for about an hour. It may be lightly torched
if required immediately.
CORSEAL 3 and 4 are ready-mixed self-drying putties which are sufficiently
permeable when fully dry to prevent blowing but strong enough to prevent
metal penetration into the joint. Drying time depends on local conditions
and the thickness of the layer applied but should be at least 30 minutes.

TAK sealant

Small variations in the mating faces of moulds due to flexing of patterns or

deformation of moulding boxes and moulding materials may result in gaps
into which liquid metal will penetrate causing runout and flash. This can be
prevented by the application of TAK plastic mould sealant which forms a
metal and gas-tight seal. TAK does not melt at high temperatures and, if
metal touches it, it burns to a compact, fibrous mass. The TAK strip is laid
around the upper surface of the drag mould, about 25 mm from the edge of
the mould cavity and the mould is then closed and clamped. TAK can also
be used to seal small core prints.
TAK 3 is supplied in cartridge form for extrusion from a hand gun, a
variety of nozzle sizes is available.
TAK 500 is ready-extruded material supplied in continuous lengths of
6 mm diameter.
Chapter 15
Lost foam casting

Principle of the process

Unlike any other sand casting process, no binders are used. Pre-forms of the
parts to be cast are moulded in expandable polystyrene or special expandable
copolymers. Complex shapes can be formed by gluing mouldings together.
The pre-forms are assembled into a cluster around a sprue then coated with
a refractory paint. The cluster is invested in dry sand in a simple moulding
box and the sand compacted by vibration. Metal is poured, vaporising the
preform and replacing it to form the casting (Fig. 15.1).

Foam sections Foam assembly Assembled cluster Coated cluster

Fluidize Pouring Compaction Sand fill

Extraction Quench Trim

Figure 15.1 The lost foam casting process.

Many problems hindered the development of the lost foam casting process.
By working closely together, designers, foundries, equipment engineers,
polymer manufacturers and coating suppliers have now removed these
barriers, making the process a cost effective way to manufacture quality
castings. The casting of iron components was initially very difficult due to
the formation of lustrous carbon defects on the surface and subsurface
 Lost foam casting 217

carbonaceous inclusions, which were only revealed on machining. Essential

to the advancement of ferrous lost foam (LF) casting was the development
of copolymers, such as the patented Foseco Low Carbon Bead, which helps
to eliminate these defects and so make the process viable.

Patternmaking
The raw bead to be moulded is purchased from a chemical supplier. It
consists of spherical beads of polystyrene (EPS) or copolymer of carefully
graded size. The bead is impregnated during manufacture with a blowing
agent, pentane. The first step is to pre-expand the bead to the required
density by steam heating. It is then moulded in a press rather like a plastic
injection moulding press. The moulding tool is made of aluminium, hollow-
backed to have a wall thickness of around 8 mm. The pre-expanded bead is
blown into the closed die, which is then steam heated causing the beads to
expand further and fuse together. After fusing, the die is cooled with water
sprays (often with vacuum assistance) so that the pattern is cooled sufficiently
to be ejected without distortion.
The cycle time is dependent on the heating and cooling of the die, a
process which is necessarily rather slow, taking 1–2 minutes for a cycle. The
moulding machines are large with pattern plate dimensions typically 800 ×
600 mm or 1000 × 700 mm so that multiple impression dies can be used to
increase the production rate.
Some lost foam foundries purchase foam patterns from a specialist supplier,
such as Foseco–Morval (in the USA and Canada). Others make their own
patterns. The casting reproduces in astonishing detail the surface appearance
of the foam pattern. A great deal of effort has been put into improving
surface quality and special moulding techniques such as Foseco–Morval’s
‘Ventless Moulding Process’ have been developed to minimise bead-trace.
EPS mouldings for casting have now reached levels of quality and complexity
far beyond that expected from a material designed originally for packaging.

Joining patterns

Where possible, patterns are moulded in one piece using the techniques
developed for plastic injection moulding such as metal ‘pull-backs’ and
collapsible cores, but many of the complex shapes needed to make castings
cannot be moulded in one piece. Sections of patterns are glued together
quickly and precisely using hot melt adhesives using special glue-printing
machines (Fig. 15.2). The adhesive is reproduced in the finished casting so
that it is important to lay down a controlled amount of glue to avoid an
unsightly glue line.
218 Foseco Ferrous Foundryman’s Handbook

Figure 15.2 Sections of patterns are glued together to form complex shapes.

Assembling clusters
Some large castings are made singly, the EPS pattern being attached to a
down-sprue of EPS or, in the case of ferrous castings, usually of hollow
ceramic fibre refractory. Smaller castings are made in clusters, the patterns
being assembled around the sprue with the ingates acting also as supports
for the cluster.

Coating the patterns

The foam pattern must be covered with a refractory coating before casting.
If no coating is used, sand erosion occurs which can lead to mould collapse
during casting. The coatings are applied by dipping and must be thoroughly
dried in a low temperature oven before casting.

Investing in sand
The coated pattern clusters are placed in a steel box and dry silica sand
poured around. As the box is filled, it is vibrated to compact the sand and
cause it to flow into the cavities of the pattern. It was the failure to vibrate
correctly that led to so many problems in the early days of lost foam. Following
much research by equipment suppliers, good vibration techniques have
been developed and patterns with complex internal form can now be reliably
invested. It is not possible to persuade sand to flow uphill by vibration so
patterns must be oriented in such a way that internal cavities are filled
horizontally or downwards.

The mechanism of casting into foam patterns

When low melting point metals such as aluminium alloys are cast into EPS
 Lost foam casting 219

foam, the advancing metal melts and degrades the polystyrene to lower
molecular weight polymers and monomers. These residues are then
transported through the coating as both liquids and gases into the sand.
This process is illustrated in Fig. 15.3. Failure to remove the gaseous residues
quickly enough results in slow mould filling and misrun castings, Fig.15.4.
If the gases escape too quickly, however, then the metal will fill the cavity in
an uncontrolled turbulent manner giving rise to oxide film defects and even
mould collapse (Fig. 15.5). If the liquid residues are not absorbed by the
coating, a thin carbon film may form on the liquid metal front which, if
trapped, may cause a ‘carbon-fold’ defect (Fig. 15.6).

Sand Gas Liquid

Coating

Metal Polymer
foam

Coating

Sand Gas Liquid

Figure 15.3 An illustration of the transport of gas and liquid polymer degradation
residues through STYROMOL and SEMCO PERM lost foam coatings.

Figure 15.4 A mis-run defect caused by the use of a coating with insufficient
permeability.

When iron is cast, the higher temperature causes rapid breakdown of the
EPS with much gas formation. Vacuum may be applied to the casting flask
to aid the removal of the gas. High temperature pyrolysis of EPS results in
tar-like carbonaceous products which are not always able to escape from
the mould. They break down further to a form of carbon called ‘lustrous
carbon’ which causes a defect characterised by pitting and wrinkling of the
220 Foseco Ferrous Foundryman’s Handbook

Figure 15.5 An oxide film defect (on aluminium) caused by turbulent mould filling
through use of a coating with too high permeability.

Figure 15.6 A cold fold defect (on aluminium) caused by poor liquid residue removal.

upper surface of the casting. High carbon irons, such as ductile iron, are
most prone to the defect. In the early 1980s it seemed as though the use of
the lost foam process would be seriously restricted by this problem. It was
then found that the polymer PMMA (polymethyl methacrylate) depolymerises
completely under heat to gaseous monomer leaving no residues in the casting.
PMMA is not easy to make in an expandable form and not easy to mould
either. Foseco has developed ‘Low Carbon Bead’, a copolymer of EPS-PMMA
which has moulding properties similar to EPS and eliminates lustrous carbon
in all but the heaviest section castings (Fig. 15.7).
Foseco and SMC have developed STYROMOL and SEMCO PERM lost
foam coatings, designed to remove both the liquid and gaseous residues at
the rate required to give controlled mould filling and defect-free castings
(see p. 239). These coatings include:

Insulating STYROMOL 169 series coatings for aluminium.

Non insulating STYROMOL and SEMCO PERM coatings for grey and
ductile iron.
 Lost foam casting 221

Figure 15.7 The use of Foseco Low Carbon Copolymer Bead eliminates lustrous
carbon defects on grey and ductile iron castings.

High refractoriness STYROMOL and SEMCO PERM coatings for chrome

irons, manganese and carbon steels.

Methoding

Patterns must be orientated to allow complete filling with sand. Ingates

must fulfil the dual role of supporting the fragile pattern cluster and controlling
the metal flow. In aluminium casting, it is mainly the permeability of the
coating that controls the filling of the casting and very gentle, turbulence
free filling is possible with direct pouring. Iron castings, having higher
density and heat content, are usually bottom gated to allow controlled mould
filling.

Advantages of lost foam casting

1. Low capital cost: The capital cost of a lost foam foundry is as little as
50% of a green sand plant of similar capacity.
2. Low tooling cost: Though tools are expensive, their life is long, so for
long-running, high volume parts like aluminium manifolds, cylinder
heads and other automotive parts, tool costs are much lower than for
gravity or low pressure dies which have a shorter life and require multiple
tool sets because of the long cycle time needed for each casting. For shorter
running parts the advantage is less and may even be a disadvantage.
3. Reduced grinding and finishing: There is a major advantage on most
castings since grinding is restricted only to removing ingates.
4. Reduced machining: On many applications, machining is greatly reduced
and in some cases eliminated completely.
222 Foseco Ferrous Foundryman’s Handbook

5. Ability to make complex castings: For suitable applications, the ability

to glue patterns together to make complex parts is a major advantage.
6. Reduced environmental problems: Lost foam is fume free in the foundry
and the sand, which contains the EPS residues, is easily reclaimed using
a simple thermal process.

Disadvantages
1. The process is difficult to automate completely; cluster assembly and
coating involves manual labour unless a complete casting plant is
dedicated to one casting type so that specialised mechanical handling
can be developed.
2. Methoding the casting is still largely hit and miss and a good deal of
experimentation is needed before a good casting is achieved.
3. Cast-to-size can be achieved but only after several tool modifications
because the contractions of foam and casting cannot yet be accurately
predicted.
4. Because of 2 and 3, long lead times are inevitable for all new castings.

Applications
The successful foundries have been extremely selective in choosing casting
applications. In general, lost foam is not regarded as a low-cost method of
casting. It is the final cost of the finished component that must be considered.

Aluminium castings

By far the largest proportion of Al castings are used in the automotive

industry so this has proved the biggest market for LF.

Inlet manifolds: This was the first successful application and is still being
used. The usual process for manifolds is gravity diecasting, a slow process
with high tooling cost. Lost foam is used to best advantage in water-jacketted
carburettor manifolds.

Cylinder heads: The fastest growing automotive application. Heads are

conventionally cast by gravity die with a complex package of cores set into
an iron die. Use of LF gives the designer rather more freedom to cool the
working face effectively, the combustion chambers can be formed ‘as-cast’,
avoiding an expensive machining operation and bolt holes can be cast.

Cylinder blocks: Automotive companies are moving away from grey iron
towards Al blocks. LF offers substantial design advantages; features can be
 Lost foam casting 223

cast in, such as the water pump cavity, alternator bracket, oil filter mounting
pad. Oil feed and drain line and coolant lines can also be cast more effectively.
A variety of other automotive parts are being made, water pump housings,
brackets, heat exchangers, fuel pumps, brake cylinders. Applications will
increase as designers become aware of the design potential. High strength
parts such as suspension arms have proved difficult because of metallurgical
differences between LF and conventional castings. New developments such
as the Castyral Process in which pressure is applied to the solidifying casting
are being adopted to improve strength.

Grey iron

Modern automated green sand is such a fast and efficient process that LF
cannot compete on cost of the casting alone. Foundries must look for castings
where the precision of LF gives savings in machining costs.

Stator cases: This has proved one of the most successful applications. There
are weight savings of up to 40% to be made through thinner cooling fins,
machining of the bore can be reduced, though not eliminated, mounting
feet can be cast complete with bolt holes.

Valves: Grey iron valve bodies, caps and gates are being cast in large numbers.
Flange faces are flat so by casting-in bolt holes etc. machining can be eliminated
completely in many cases.

Ductile iron

Pipe fittings: One of the most successful applications, the precision of LF is

so great that flanges can be cast flat complete with cast bolt holes so that
machining can be eliminated altogether. Bends, tees, spigots etc. up to at
least 300 mm diameter are being cast in large volume, replacing resin sand
moulding. Achieving success has not been easy because of distortion problems
both on foam patterns and castings, but the potential benefits are so great
that foundries in Europe and Japan have persisted and finally achieved
success.

Valves: Machining can be eliminated completely in many cases, particularly

with redesign of the part. LF is becoming the accepted way of making water
valves up to about 150 mm pipe diameter.

Hubs: Lower weight and reduced machining is possible.

Differential cases: Several automotive companies have persisted with this

difficult casting. The advantages are reduced machining (on the internal
surfaces) and better balance.
224 Foseco Ferrous Foundryman’s Handbook

Manifolds: Weight savings are achieved through control of wall thickness

and clear passages improve performance.

Brackets: Bolt holes can be cast so that machining can be eliminated completely
in some cases.

Other alloys
Steel: The process is not suitable for most steel castings because of the
danger of carbon pick-up from the foam pattern. The carbon pick-up is not
uniform, but carbon-rich ‘inclusions’ are found in carbon steel castings made
using the process. Residues from the pyrolysis of the EPS become trapped
in the liquid steel giving rise to areas of high carbon, up to 0.7%C, often
under the upper surfaces of the casting where the EPS residue has been
unable to float out of the steel. Such defects are unacceptable in most steel
castings. Even the use of foam patterns made from PMMA does not entirely
eliminate the problem. Some high carbon steels, such as austenitic manganese
steel which has around 1.25%C content, can be successfully cast using lost
foam.

Wear resistant castings: Elimination of flash and dimensional precision give

LF a major advantage over other casting methods. Grinding balls can be
made in enormous clusters. Shot blast parts are made in large numbers.
Slurry pump bodies can be cast with minimum machining saving many
hours of machining time. Wherever the rather high tooling costs can be
justified, LF has become the standard method of manufacture.

Duplex castings: The process lends itself to making castings containing

inserts, such as tungsten carbide or ceramic. The inserts are fitted into the
foam pattern before casting.

The future
LF casting has now passed through the critical development stage to become
a mature foundry process. While many of the early technical problems have
been overcome there are still some to be solved:

Cluster assembly, still requires too much manual work.

Methoding, still a matter of trial and error, though the current work on
modelling is helping.
Distortion of foam patterns can be a problem, but more and more the
coating is seen as a way of stiffening fragile patterns.
Glueing, expensive and not easy on non-flat joint lines.
Automation, better methods of mechanically handling foam patterns for
cluster assembly and coating are needed.
 Lost foam casting 225

Lost foam will not replace conventional casting processes. It is seen as a

clean process, all the used sand can be easily reclaimed and chemical residues
burnt off so that it is probably the most environmentally acceptable foundry
process at present available. This will prove to be a growing asset in the
future. But that is not the main reason. For the major automotive groups the
reason is the design flexibility offered.
Chapter 16
Coatings for moulds and
cores

The need for a coating

When liquid metal is cast into a sand mould or against a core, there may be
a physical effect and a chemical reaction at the sand/metal interface. Either
may result in surface defects on the finished casting.

Physical effects – metal penetration

Liquid metal penetrates the pores of the sand mould or core giving a rough
surface to the casting. The degree of penetration is dependent on:

metallostatic pressure
penetration is most severe in the lower parts of the casting and with
high density metals such as iron and steel
dynamic pressure
most severe where the metal stream impacts on the mould or core
pore size of mould or core surface
open pores arise from the use of coarse sand or sand with poor grading,
or from poorly compacted sand, due to imperfect core blowing or the
use of sand having a high viscosity binder or one which has exceeded
its correct work time
sand expansion
stresses can form in the bonded sand due to differential thermal
expansion, this can lead to the formation of mould or core surface
cracks allowing ingress of molten metal (finning or veining).

Chemical effects
(1) Burn-on
A chemical reaction occurs between sand and metal. It occurs with

impure sand
some impurities, particularly alkalis, reduce the refractoriness of the sand
some binders
mostly those based on sodium silicate which have poor refractoriness
so that liquid phases are produced at temperatures as low as 900°C
 Coatings for moulds and cores 227

certain cast metals

notably manganese steel which form oxides that react with silica sand
to form low melting point liquids.
(2) Carbonaceous defects
Organic mould and core binders degrade at high metal pouring
temperatures, forming carbon bearing gases which can lead to carbon
pick-up or surface pock marks due to lustrous carbon.

Metal penetration, sand burn-on and carbon defects occur most commonly
with ferrous metals, due to their high casting temperature, high density and
the reactivity of their oxides to silica sand. Conversely, moulds and cores for
aluminium alloy castings rarely display metal penetration and burn-on so
coatings are often un-necessary.
A refractory coating on the mould or core can reduce or eliminate metal
penetration and sand burn-on. For many years, foundries made their own
coatings, usually ‘blackings’, based on coke dust or graphite. The technology
of coatings has now become highly developed, and most foundries now
purchase coatings from specialist suppliers.
It is important to recognise the potential savings in shotblasting and
cleaning costs that can be made by using a good quality, well applied coating.

Coating selection

An effective coating should have the following characteristics:

Sufficiently refractory properties to resist the metal being poured.

Good adhesion to the substrate.
No blistering, cracking or scaling tendency on drying.
Good suspension and remixing properties.
Good stability in storage.
Good covering power.
Good application properties.

Choice of coating and form of supply

When choosing a coating, consideration must be given to:

Application method and equipment available

Carrier – water or spirit-based
Method of drying
Performance required
Mode of supply – tanker, tote tank, drums or powder.

Coatings are supplied in dry or liquid form.

228 Foseco Ferrous Foundryman’s Handbook

Powder coatings

The attraction of a powder coating is that it costs less to transport than a

liquid, it is easy to store and can reduce waste since precise quantities of the
mixed coating can be made to order. The characteristics of a coating prepared
from a powder pre-mix are highly dependent upon the mixing equipment
used and the procedure adopted. If simple paddle mixers are used, long
mixing times followed by ageing up to 24 hours may be required to develop
the correct suspension and rheology. It is preferable to use special high
shear mixers to develop the liquid coating and avoid the need for ageing.
Suppliers of powder coatings will advise on the best mixer to use and on
the correct testing procedure to ensure that the optimum coating properties
are achieved.

Creams and ready-for-use coatings

A ‘ready-for-use’ paint is the easiest to use. So-called ready-for-use paints

may also be supplied as light creams or slurries which require a small water
addition to allow the user to control the final consistency of the coating to
suit his particular application. During transport and storage of ready-for-
use coatings, their is always some separation from suspension of the refractory
filler, often a high density mineral, but such segregates are usually easily re-
incorporated by simple stirring.

Components of a coating
Foundry coatings consist of:

Refractory filler
Liquid carrier
Binder
Rheology control system.

The refractory filler

This is the most important constituent since its properties decide the efficiency
of the coating. The filler may be either a single material or a blend selected
for specific applications. Fillers are chosen for their particle size and shape,
sintering point, melting point, thermal conductivity, thermal expansion and
reactivity towards the metal being cast and the moulding material on which
it is being used. Zircon, being highly refractory, is used for the most demanding
applications such as steel or iron castings where section thickness is great.
Graphite and coke are commonly used for lighter iron castings, often in
 Coatings for moulds and cores 229

combination with zircon, silica or talc. Elimination or reduction of finning

can often be achieved through the inclusion of lamellar platelet fillers such
as talc and mica. A combination of silica and iron oxide is frequently used
as an inexpensive refractory blend for cores for automotive castings. Graphite
may be included to improve strip.

The liquid carrier

The liquid carrier is the vehicle for the total coating composition and serves
to transport the filler onto the sand substrate. It must be removed before
casting takes place. The most commonly used carriers are water and
isopropanol, although other solvents such as methanol, ethanol, hydrocarbons
and chlorinated hydrocarbons have also been used. Water is cheap and
readily available but drying in an oven is usually necessary to remove it
before casting. Foseco has introduced Rapid Drying Technology (RDT) to
many of its water based coatings, shortening the drying time needed. Oven
drying can be a problem on large moulds or cores. Isopropanol dries faster
than water and can be removed quickly by burning so it is frequently used
for large moulds and cores, but it is a more expensive carrier than water.
Methanol and ethanol have higher vapour pressures at 20°C than isopropanol
(Table 16.1) and dry faster still. Chlorinated hydrocarbons were at one time
introduced for fast air drying, but they are now obsolete due to environmental
concerns. The use of water based coatings has increased in recent years, also
for environmental reasons.

Table 16.1 Alcohols used as carrier liquids

Methanol Ethanol Isopropanol

Flash point (°C) 11 12 12

Boiling point (°C) 65.2 78.5 82.3
Vapour pressure 94 43 32
at 20°C mm Hg
Explosion limit
(Vol. %) 5.5–26.5 3.5–15 2.0–12
MAK value (ppm) 200 1000 400
Smell limit (ppm) 2000 350 100

The use of water based coatings on cores and moulds made with chemically
bonded sand can effect the strength of the bond, and may give rise to casting
defects due to surface friability of the core or mould. Silicate binders are
particularly badly affected by water based coatings and spirit based coatings
should be used instead. Cores made using phenolic-isocyanate resin processes
can also be affected, and care must be taken to dry the cores quickly and
immediately after coating.
230 Foseco Ferrous Foundryman’s Handbook

The binder

The function of the binder is to bond together the filler particles and to
provide adhesion to the mould or core. The binder often interacts with the
whole of the coating composition and therefore cannot be considered in
isolation.

The rheology control system

This provides the suspension system that prevents the filler from separating
out during storage of the coating over extended periods. It ensures that the
coating is homogeneous and ready for application with the minimum of
agitation. It also controls the flow properties of the coating and is designed
to suit the application method that is used. Coatings applied by dipping are
different to those designed to be applied by brush, spray or overpouring
and it is the rheology control system that affects whether the coating can be
applied easily and evenly during use.

Application methods for coatings

The main methods of coating moulds and cores are:

Dipping
Brushing
Swabbing
Spraying
Overpouring/flowcoating.

Each method needs different properties in the coating.

Dipping

This method involves submerging the core in the liquid coating, allowing it
to pick up a suitable layer of the coating. It is the fastest method of application
and is used by the majority of high production foundries. The core is dipped
manually or mechanically and when it is removed from the coating bath, it
must be allowed to drain to remove excess coating. The success of the
operation depends on the consistency of the coating and its rheological
properties. It is essential to ensure that a uniform thickness of coating is
applied, that there is adequate wetting of the sand by the coating to give
good adhesion and that excess coating drains freely from pockets in the
core. Failure to do this will affect the dimensional accuracy of the resulting
casting.
 Coatings for moulds and cores 231

The type of liquid flow that is suitable for dipping is known as a

pseudoplastic rheology. A pseudoplastic liquid experiences an immediate
decrease in viscosity upon the application of shear. The original viscosity is
regained as soon as the shear is removed. As the core is submerged in the
coating, the shear involved reduces the viscosity to a low level, thereby
ensuring effective coating of all areas of the core. Coating viscosity increases
rapidly on removal of the core and drainage of excess coating ceases when
the downward coating film weight is counteracted by the upward resistance
of the gel. With a correctly designed coating, the necessary thickness is
deposited in a very even manner without sagging, excessive dripping or
tear formation.

Brushing

This is a simple application technique and one or more layers may be applied.
The correct flow behaviour needed is known as thixotropic rheology (well
known to users of household paints). The viscosity decreases slowly upon
the action of shear and regains more quickly, but not immediately, on shear
removal. The coating can be of almost jelly-like consistency in the can but
will flow readily when brushed. It gives extremely good suspension of the
filler but allows brush marks to flow out after application, leaving a smooth,
even layer of coating on the core or mould.

Swabbing

This is similar to brushing, except that the implement used has long, soft
bristles which tend to adhere to each other. It is easier to penetrate re-
entrant areas of the mould with these long bristles and to apply a thick,
smooth coating with the minimum of dripping. There is less tendency to
mould damage or residual brush marks in the finished coating than with
brush application. Both brushing and swabbing are rather slow and require
a skilled operator to achieve the best results, but developments in coating
rheology have significantly reduced the skill factor needed.

Spraying

This is a much faster technique than brushing but penetration and coverage
of the sand is not as good as with brushing. Skill is required to ensure that
deep pockets and re-entrant angles are fully coated. It is best used on moulds
and cores having rather simple shapes.
Generally, sprayed coatings are of much lower viscosity than brushing or
dipping varieties. They are applied under pressure from a spray gun delivering
a spray of finely divided droplets on to the surface to be coated. A degree of
232 Foseco Ferrous Foundryman’s Handbook

thixotropy is desirable to prevent the coating from sagging and tearing.

Correctly designed spray systems are available that give very good results,
but older equipment may give problems as only very low viscosity coatings
can be handled with these spray guns.

Overpouring/flow-coating

Overpouring, also known as flow-coating, is particularly suitable for larger

cores and moulds, it consists of a pumping system and a specially designed
nozzle. The coating is continuously pumped through the nozzle which directs
it over the core, the excess runs back into a reservoir to be recirculated
through the nozzle once more. Overpouring also requires special equipment
for handling large moulds and cores. The rheology of the coating must be
designed to allow an even layer build-up independent of the amount of
overpoured coating used. The gel system should allow an even deposition
of coating without the risk of residual runs and tear marks.

Mixing and testing suspensions

The suspension agents and rheology control agents used in coatings give
the slurry some unusual properties which are difficult to measure in terms
of simple viscosity. The viscosity of such slurries varies according to the
shear forces applied. Users of household non-drip paints will be familiar
with the way that the paint flows easily during brushing, then gels into a
more viscous liquid as soon as the brush is removed. With coatings, a similar
effect is seen. After shearing with a paddle mixer, for example, the slurry
will take a little time (seconds or minutes) to recover its gel structure. This
makes it difficult for the user, with only simple equipment, to check the
consistency of the coating.
The usual method of checking coating consistency is by means of a Baumé
reading, but the Baumé hydrometer must be carefully used to give consistent,
meaningful readings.

The hydrometer must be clean.

Allow the mixed coating to settle for 5 minutes after mixing.
Gently immerse the hydrometer, supporting it manually until it finds the
correct level (remembering that any shearing force on the slurry will alter
its rheology).
Wash the hydrometer immediately after use, since dried-on coating is
difficult to remove.
Always use the same type of hydrometer, coating suppliers will usually
provide a suitable type.

Another method of testing coatings, particularly the less viscous slurries

 Coatings for moulds and cores 233

used for overpouring or spraying, is to measure the time taken for the
slurry to pass through a ‘flow cup’. There are a number of standard flow
cups available, the 4 mm DIN cup is commonly used for coatings.
Neither of these test methods will measure the true coating properties of
any slurry and, for dip-coats in particular, some foundries use a small test
core which is weighed, dipped in the coating, dried, then weighed again to
measure the ‘dry weight pick-up’.

Equipment for mixing and applying coatings

Coatings applied by brush or swab are often used straight from the container
in which they are supplied with perhaps a small adjustment in water or
carrier liquid made and manually stirred in. For dip coating, spraying and
overpouring it is usual to have a separate mixing tank in which the coating
is adjusted to the required consistency and from which the dip tank, spray
container or holding tank for flow coating can be topped-up when needed.
Transfer between supplier’s drums or containers to the mixing tank and
from mixing tank to the working tank or vessel may be done by using a
double diaphragm slurry pump.
Dip tanks should be fitted with a paddle stirrer to maintain the coating
in suspension without excessive movement, which could affect the rheology
of the coating. The temperature of the mixed coating should be maintained
reasonably constant.
Equipment for the application of coatings by the overpour or flowcoat
system consists of a method of pumping a continuous flow of coating over
the core, allowing the surplus to run off the core to a catchment tray from
which it is returned to the storage tank. Small portable systems may be
used, in which a portable catchment tray is placed under the work and the
coating pumped over the core through a coating gun by a diaphragm pump
or by displacement with low pressure compressed air. Larger cores and
moulds are usually coated at a fixed station having a coating storage tank of
about 100 litres capacity allowing about 4 minutes uninterrupted coating
time. Cranes or hoists and specially designed racks may be needed to support
the work while it is being coated, with moulds being angled over the catchment
tray in such a way that no pools of coating are left on the work.

Drying ovens
Cores coated with water based coatings must be thoroughly dried before
casting. High production cores are usually passed through an oven on a
conveyor. Warm air, infra-red and de-humidifying ovens may be used, in all
cases good air circulation is needed to carry away the moisture laden air
from the core surface. Air temperatures up to 100°C may be used. Large
cores and moulds, which cannot be moved to an oven, may be dried using
portable warm air blowers, or a gas torch.
234 Foseco Ferrous Foundryman’s Handbook

Coatings for iron and steel foundries

Coatings can conveniently be divided into two types:

Coatings for high production cores

Coatings for jobbing cores and moulds.

Foseco supplies both types and following the merger of SMC with Foseco in
1998, the Foseco range of coatings has been supplemented by the SEMCO
and TENO range of coatings.

Coatings for high production foundries

The RHEOTEC series of coatings

RHEOTEC coatings are water based coatings designed for dipping of cores
which are then dried in a warm air oven. The coatings can be supplied as
powders, creams or ‘ready-for-use’ slurries. They help to provide a clean ‘as
cast’ surface finish on the internal passages of iron castings.
High production iron foundries use cores bonded with resin, usually
phenolic-urethane or hot box. At the temperature of iron casting, the cores
are prone to expansion defects which cause ‘finning’ or ‘veining’ defects in
the castings. Veining defects arise from metal entering cracks in the core
surface which form as a result of the thermal stresses generated by the
expansion of silica sand during casting. In severe cases, the effect can lead
to metal penetration rather than simply a vein defect.
Phenolic-urethane cold-box cores are more prone to veining defects than
cores made by hot box, shell, ECOLOTEC, etc., due to the high packing
density of the sand and the inherently low hot strength of the urethane
binder. Castings suffering from veining require additional cleaning and
grinding. If the veining or metal penetration occurs inside a water jacket or
other internal cavity, it is impossible to remove and the performance of the
casting, such as a cylinder head, block or manifold will suffer.
Iron castings made with cold box cores may also be affected by ‘lustrous
carbon’ defects. The phenolic-urethane corebinder, at molten iron temperature,
pyrolyses to form lustrous carbon on the molten metal front which affects
its flow properties and may cause a ‘wrinkled’ surface on the casting which
is unsightly and may be harmful. Core coatings can be designed to reduce
veining and lustrous carbon as well as metal penetration and burn-on.

RHEOTEC XL anti-veining water based core coatings

The Foseco RHEOTEC XL coatings are particularly effective in reducing or

 Coatings for moulds and cores 235

even eliminating veining. They are used on cores for engine blocks, cylinder
heads, ventilated disc brakes, manifolds, domestic boilers etc. where remedial
cleaning is difficult or impossible. The coatings exhibit the following features:

High insulation (Fig. 16.1 )

Excellent application characteristics
Controlled penetration of refractory fillers into the core substrate
High hot strength, giving a stable coating layer at casting temperature.
Core surtace temperature (°C)

1000

800
∝· β phase
600 change

400
RHEOTEC 541 XL
Conventional Coating
200

0
0 100 200 300 400
Time after pouring (seconds)

Figure 16.1 Illustration of the superior insulating behaviour of RHEOTEC XL coatings.

The reduced thermal shock experienced by the core when coated with
RHEOTEC XL delays the α–β phase change of the silica sand, reducing the
veining tendency. A single dip coat of RHEOTEC XL is sufficient. The coating
is supplied either as a cream concentrate or in a ready-for-use consistency.
The coating is applied at a suitable dilution to give a thickness of around
0.2–0.3 mm of dry coating layer. This is achieved at an application Baumé of
38–42.

RHEOTEC RDT fast-drying coatings

Water based coatings must be dried before the coated core or mould can be
cast. Drying stoves use a combination of gas or electric heating (sometimes
infra-red) with forced air movement sometimes with dehumidification to
achieve the fastest drying at minimum fuel cost.
Refinement of the RHEOTEC system has led to the evolution of the RDT
range of fast-drying RHEOTEC coatings. The technology behind these coatings
is based on reducing the degree of ‘wicking’ of the coating into the core
substrate. Drying times are significantly reduced compared with conventional
coatings (Fig. 16.2).
236 Foseco Ferrous Foundryman’s Handbook

100
90 45
40
80 Drying
% Water remaining
35
30
70 time in 25
60 minutes 20
15
50 10
5
40 0
30
20 Rheotec RDT Conventional
10
0
0 10 20 30 40 50
Time (minutes)

Figure 16.2 Drying comparison, RHEOTEC RDT versus conventional coating.

Typical water based dip coatings

RHEOTEC 102 A general purpose, low cost coating for non-critical

applications.
RHEOTEC 204 Water based talc/graphite coating giving exceptional
as cast surface finish, particularly good on cylinder
heads and applications where veining is not a
particular concern.
RHEOTEC 460 Based on alumino silicate and iron oxide fillers, red
in colour. Resistant to metal penetration and burn-
on, good veining resistance. Suppresses lustrous
carbon defects.
RHEOTEC 460 RDT A faster drying version of 460. A special gel/surfactant
system ensures optimum ‘pick-up’ with little pene-
tration of water into the core surface, making it ideal
for intricate cold box cores which suffer from strength
problems when coated with water based products.
RHEOTEC 541XL High performance water based dip coating. The
special refractory technology provides superior anti-
veining characteristics and extremely low levels of
retained sand/coating residues after shake-out. Used
in the production of complex iron castings such as
cylinder heads and blocks, brake components and
hydraulic castings where casting surface finish must
be of the highest standard.

Solvent based dip coating

TENOSIL 301MPX An alumino silicate solvent based coating used for
thin section, complex urethane cold box cores where a
water-based coating may cause problems due to
strength reduction.
 Coatings for moulds and cores 237

Coatings for jobbing moulds and cores

The HOLCOTE and SEMCO series of coatings

This range of coatings are ready-for-use water based coatings use a rheology
control system designed for flow coating, spraying and brushing application
for use on chemically bonded sand moulds and cores. Typical products in
the range include:

SEMCO zir 35PX Zircon based coating designed particularly for flow
coating.

HOLCOTE 110 Zircon based and suitable for steel castings and heavy
section iron castings where organic sand binders such as furane or phenolic-
urethane resins are employed. HOLCOTE is applied by brushing, swabbing,
spraying and overpouring. Control of penetration into the sand can be
obtained by adding up to 10% by weight of water. HOLCOTE coatings are
thixotropic, the normally jelly-like material becomes mobile when shear
forces are applied, as in stirring, brushing or spraying. After a short interval
it will then recover its original gel properties. This means that HOLCOTE
110 will cling without dripping to a brush or swab, will liquefy, spread and
penetrate when worked onto the mould or core surface and then level out,
flattening out brush marks. HOLCOTE 110 is not suitable for dip applications
and like other water based coatings is not recommended for silicate bonded
moulds or cores. The applied coating can be cured by air drying, oven or
infra-red drying. It can also be rapidly dried using a gas torch without any
fear of blistering, lifting or cracking.

HOLCOTE 300 RDT High performance zircon based coatings for steel and
heavy iron castings. Can be applied by all methods, but especially dipping
and overpouring. It has an enhanced drying capability.

Application guide: Dipping 75–85° Baumé

Overpouring 73–84° Baumé
Spraying 85–95° Baumé
Brushing 85–100° Baumé

It will air dry in time but more positive methods are preferred. Even strong
torching will not blister, crack or spall the coating.

Spirit based coatings

In cases where oven-drying of coated cores or moulds is not practicable,
usually because the moulds or cores are too large to move, spirit based
238 Foseco Ferrous Foundryman’s Handbook

coatings are often used. After coating the mould or core, the coating is
ignited and the carrier liquid burned off. Isopropanol is the most commonly
used spirit carrier liquid, but ethanol and methanol may sometimes be
added to improve the burning characteristics. The properties of carrier liquids
are given in Table 16.1.
Isopropanol is chosen because of its low toxicity, low smell and low
vapour pressure. Methanol is little used because of its potentially hazardous
properties.
Spirit based coatings are usually supplied as ready-to-use coatings or
heavy creams. Since spirit based coatings are used mostly on large moulds
and cores for heavy castings, the fillers must be highly refractory and often
contain a high proportion of zircon and other dense materials. This makes
suspension difficult and sedimentation was always a problem, but develop-
ments in suspension and rheology control have improved greatly in recent
years and sedimentation is no longer a serious problem.

Storage of spirit based coatings

Because they are required to burn out readily in use, spirit based pastes and
slurries and the solvents used to dilute them are highly flammable and
must be stored in flame-proof buildings according to local regulations such
as the Petroleum Regulations in force in the UK.

Foseco spirit based coatings: ISOMOL and

TENO product ranges

These coatings are used almost exclusively in jobbing foundries making

iron and steel castings and can be used on any substrate i.e. CO2-silicate,
ester-silicate, furan etc. Application is by brush, spray or flow coating; curing
is by air-drying or preferably by flaming-off.

TENO Coating ZN 75 a ‘sulphur block’ coating for ductile iron castings

to prevent reversion of nodules to flake graphite
at the surface of the casting. Designed for flow
coating.
TENO Coating 70PX based on alumino silicate for iron castings,
designed for flow coating.
ISOMOL 210 based on zircon/graphite for heavy iron castings.
ISOMOL 300 based on zircon for steel castings, designed for
brush application.
ISOMOL 380 FL  based on zircon for steel and heavy section iron,

TENO ZBB P16  specifically designed for flow coating.
ISOMOL 580 
 based on magnesite for manganese steel castings.
TENOSIL MOPX4 
 Coatings for moulds and cores 239

Evaporative polystyrene process coatings

In the evaporative casting, or lost foam casting process, a pattern of expanded

polystyrene (EPS) is moulded to the required shape and attached to a running
system also made of polystyrene. The assembly is coated with a refractory
coating and after drying, is immersed in loose, unbonded sand which is vibrated
to compact the sand around the pattern. Metal is poured into the pattern
which is vaporised by the heat and replaced by the metal. The EPS shape is
reproduced exactly by the metal, forming the casting (see Chapter 15).
The refractory coating plays an essential role in the process. If patterns
are cast without a coating then sand burn-on occurs and severe erosion can
take place near ingates and other areas where metal flow changes direction
or speed. The coating must have the following properties:

It must provide an adequate refractory barrier to the molten metal.

It must have permeability, to allow liquids and vapours from the

polystyrene to escape during mould filling.
It should have good strip and release from the finished casting.

In addition, the dried coating should have good strength, since it will then
stiffen the pattern and reduce the tendency for it to distort during sand fill.
Unlike other casting processes, the surface smoothness of the casting is not
determined by the smoothness of the coating. It is the surface quality of the
polystyrene pattern which dictates the casting surface finish, because the
coating penetrates any crevices present in the pattern surface, causing the
casting to replicate the pattern surface exactly, including any defects that
may be present.
With iron castings, large volumes of gas are generated from the high
temperature degradation of the pattern, so the permeability of the coating
must be high, otherwise dangerous blow-back of metal up the sprue may
occur. Care must be taken to ensure that only coatings designed specifically
for the lost foam process are used. The STYROMOL and SEMCOPERM
range of coatings has been developed specially for use with lost foam. All
the coatings are water based and formulated for application by dipping.
Grades are available for aluminium, grey and ductile irons and for special
irons (p. 220). Patterns must be coated inside and out, since no cores are
used in the process. Care must be taken to dry the patterns thoroughly after
dipping, the drying temperature should not exceed 45°C to avoid thermal
distortion of the foam pattern. Special de-humidifying ovens, with good air
circulation are frequently used to ensure complete drying of the coating,
both inside and out. Microwave heating has also been used to dry difficult
inside surfaces, but it is necessary to remove most of the water by conventional
warm air, then finish by microwave. Use of microwave heating alone will
cause the water in the coating to boil before it dries, and this may cause
bubbles or even cause part of the coating to flake away.
240 Foseco Ferrous Foundryman’s Handbook

SEMCOPERM M66 has been developed for coating EPS patterns used for
making iron castings by the lost foam process. It is particularly suitable for
vacuum assisted casting. The permeability is such that metal pull-through
is unlikely to take place. The coating is supplied as a heavy slurry at 95–100°
Baumé. It must be diluted with water to 55–65° Baumé for application by
dipping or overpour methods.

Coatings for the full-mould process

This process uses polystyrene patterns to make large one-off castings. The
patterns are usually machined from large blocks of expanded polystyrene.
After coating, the patterns are surrounded with a self-setting resin bonded
sand to form a rigid mould. The process is widely used to make large, grey
iron press-tool castings, sometimes many tonnes in weight, having metal
sections as much as 100–150 mm thick. The coating in such cases must be
capable of being applied in a thick layer, without cracking or peeling, to
provide an adequate barrier to the metal. Foseco supplies STYROMOL 702FM
coating, developed specially for this purpose. Spirit based coatings must
not be used for this application, since burning off the coating would result
in loss of the pattern before the sand mould was formed.
STYROMOL 702FM is a dense thixotropic water based slurry containing
a blend of graphite and refractories, which can be applied by spray or
brush. To facilitate application, a small amount of water can be added to
give a Baumé of 90–100°. This will give a coating thickness of 1.5–3 mm. The
most effective method of application is via a spray system. Once applied,
the coating must be completely dried prior to ram-up and casting. The
coating must be air dried using warm air circulation (but the temperature
must not exceed 45°C or there is a risk that the polystyrene pattern may be
damaged). It may take 2–3 days before all the moisture is removed.

The TRIBONOL process

Green sand moulding is the most widely used moulding process for the
production of a wide variety of high production castings. Modern green
sand moulding machines allow production rates of up to 300 moulds per
hour, with ever increasing complexity of casting design. However, even
with an optimised sand system and a high quality moulding machine,
problems of poor surface finish due to metal penetration and burn-on persist
in many cases. Casting surface finish problems from green sand moulds
have traditionally been dealt with in a number of ways:

Use of a high quality facing sand

Use of a liquid based coating
 Coatings for moulds and cores 241

Replacing problem areas of the mould face by a cored surface

Increased shotblast times.

Use of a liquid coating is often favoured, as it is easy to implement and

involves low capital cost. However, liquid coatings are difficult to apply
and slow down the production rate. Often they generate other problems
such as reduced mould permeability and wet patches which can give rise to
gas blow holes.
The TRIBONOL Process was developed to overcome the problems of wet
coatings applied to greensand moulds. It has two components:

A dry, free flowing, highly refractory zircon-based powder coating called

TRIBONOL
A special electrostatic delivery system for applying the TRIBONOL coating
to a green sand mould.

Two types of delivery system are available, a manually operated ‘hand gun’
and a fully automated ‘multi-gun’ system for high productivity.
The TRIBONOL coating is given a frictional electric charge as it passes
through the application gun. Since the charge is generated by friction, there
is no electrical supply or high voltage involved. As the charged powder
leaves the gun it is attracted to the green sand surface, which is at earth
potential, and adheres to the surface. The spray pattern is not as directional
as a liquid spray, because the electrostatic attraction effectively coats shadowed
areas and deep pockets, depositing a very uniform layer of coating on the
mould. Being completely solvent-free, there is no need to dry the coating, so
that moulds can be immediately cored-up and rapid mould closure is possible
with no loss of productivity.
The TRIBONOL Process is particularly suitable for repetition iron foundries,
such as foundries producing engine blocks, cylinder heads, brake drums
and other automotive components.
TRIBONOL ZF is an anti-pinholing coating designed to reduce nitrogen
pinholing arising from the contamination of the moulding green sand with
high nitrogen hot box and shell cores.

Miscellaneous coatings

HARDCOTE bond supplement

HARDCOTE 4 non-refractory, liquid dressing for hardening the face of

greensand, silicate or resin bonded moulds and cores. Its function is to act
as a supplementary bond holding mould faces, edges etc. in place in situations
where friability might otherwise affect mould integrity. It is best applied by
spraying but brush or swab may also be used.
242 Foseco Ferrous Foundryman’s Handbook

HARDCOTE 4 should be left to dry in the air for around 10 minutes (for
silicate or resin bonded moulds) and about 20 minutes for green sand. The
moulds must not be closed before the coatings have dried.

Dressings to promote metallurgical changes

TELLURIT

These coatings contain metallic tellurium which acts as a chill-promoting

medium for cast iron. They are used for producing a wear resistant surface
layer on grey cast irons. The effect is localised to where the coating is applied,
and a chilled layer about 3 mm deep is produced. TELLURIT can also be
used on chills to enhance the chilling effect.

TELLURIT 2 is a paste for dilution with water to form a dressing of paint-

like consistency. Coating is usually applied to moulds and cores which are
still warm from drying or baking. The coating must be thoroughly dried
before a second coat is applied.

TELLURIT 50 is diluted with isopropanol before use and either air dried or
burned off before the mould is closed.

Tellurium vapour can be toxic and care must be taken to ensure that
operators are not exposed to vapour either during drying the coating or
during casting the moulds.

MOLDCOTE 50 is a flammable bismuth-containing, paste mould dressing

for localised densening of cast iron. At molten iron temperatures bismuth
dissolves in the iron and locally alters the solidification characteristics. Its
effect on carbide stabilisation is less severe than tellurium resulting in a
densening effect and avoiding chilling or retention of massive carbides. It
can eliminate the use of metal chills and is useful for picking out isolated
bosses, cylinder bores and other heat centres affected by open grain where
conventional feed is difficult to apply. The paste is diluted with isopropyl
alcohol and applied by brush to the area required. The dressing may be air
dried or burned. Treated castings can be re-melted without fear of bismuth
build-up since it is not retained during melting.

Other special dressings

CHILCOTE A range of dressings for loose chills and cast-in metal inserts
which form parts of castings.

CHILCOTE 4 is a refractory, self-drying dressing for external chills. It provides

 Coatings for moulds and cores 243

a permeable refractory layer on the chill face which permits lateral movement
of any evolved gas, prevents welding and resists damping-back in closed
green sand moulds awaiting casting. The chills are protected, preserved
and more easily cleaned for reuse. It is used for coating external metal chills
used in aluminium and zinc alloy castings and also on small chills for
copper base alloys and cast iron.

CHILCOTE 10 is a fusion promoter. In the grey iron industry, it is sometimes

desirable to include denseners and inserts which, after pouring, become
part of the casting. Application of CHILCOTE 10 to the shot-blasted or
pickled chill before inserting in the mould, ensures maximum fusion between
metals.

SPUNCOTE 10 This is a specialist water based slurry coating, containing

alumina as the refractory, formulated to provide a permeable coating with
very low gas evolution for use in the centrifugal casting process for the
manufacture of pipes and liners. The coating also assists traction on the
coated face, allowing even flow of metal during the casting process and
unrestricted extraction of the casting.

Coatings for foundry tools

Foundry tools used in the melting and handling of aluminium alloys (and,
to a lesser extent, copper-base alloys) must be coated with refractory. The
coating prevents the danger of iron contamination arising from the use of
unprotected tools. HOLCOTE 110 is suitable. Plungers, skimmers, tongs
etc. are cleaned and heated to 80–100°C and plunged into the HOLCOTE
water based coating. The treatment may be repeated several times daily.
Iron and steel ladles and shanks are given three or four coatings once or
twice daily before use. Several thin coatings give better protection than a
single thick one.
The HOLCOTE coating must be thoroughly dried before being brought
into contact with the liquid metal. This may be done by placing them near
to the furnace for a period, then into the furnace flame immediately before
being used.
FRACTON dressings are designed for the protection of troughs, launders,
refractories etc., from attack by molten metal.

FRACTON 4 dressing provides a highly refractory top dressing to refractory

work that is not wetted by molten metal or most slags, drosses and fluxes.
The underlying material, brickwork, crucible or tool, is therefore protected
and preserved. Potential build-up material does not stick but readily falls
away. Skulls drop out cleanly and ladle and crucible cleaning is reduced to
a minimum.
244 Foseco Ferrous Foundryman’s Handbook

FRACTON 100A dressing is designed for the protection of metal launders,

spouts, pig moulds etc. from attack by molten metal. The principal application
is to cast iron launders used to convey molten metal for pipe-spinning
processes. It may also be used to protect metal moulds etc. against attack by
molten copper and copper alloys, nickel, aluminium etc. Its application to
steel is limited by its carbonaceous nature.
Chapter 17
Filtration and the running and
gating of iron castings

Introduction
The running and gating system carries out the following functions:

Controls the flow of metal into the mould cavity at the rate needed to
avoid cold metal defects in the casting.
Avoids turbulence of metal entering the mould.
Prevents slag and dross present in the iron from entering the mould.
Avoids high velocity impingement of the metal stream onto cores or
mould surfaces.
Encourages thermal gradients within the casting which help to produce
sound castings.
Enables the casting to be separated from the running/gating system easily.

It is not possible to achieve all these requirements at the same time and
some compromise is always necessary.
When considering running systems for iron castings, it is necessary to
distinguish between grey iron and ductile iron. While some furnace slag
may be present in liquid grey iron, it is not a dross-forming alloy so is not
subject to inclusions due to oxidation of metal within the running system.
Ductile iron, on the other hand, contains magnesium silicate and sulphide
dross arising during treatment with magnesium. Moreover, residual
magnesium in the treated liquid metal can oxidise when exposed to air to
form more dross. Running system design must take this into account. The
widespread use of ceramic foam filters in iron casting has enabled running
system design to be simplified.

Conventional running systems without filters

The elements of a running/gating system for a horizontally parted mould
are shown in Fig. 17.1.

Pouring bush The use of a properly designed pouring bush is recommended

246 Foseco Ferrous Foundryman’s Handbook

Pouring basin
Feeder head

Sprue

Riser or vent

Runner

Ingate Casting

Figure 17.1 The basic components of a running system.

on all but the smallest of castings. The pouring bush should be designed in
such a way so that the pourer can fill the sprue quickly and so maintain a
near constant head of metal throughout the pour and retain most of the slag
and dross within the bush. An off-set design incorporating a weir achieves
this objective, Fig. 17.2. The pouring bush should be rectangular in shape so
that the upward circulation during pouring will assist in dross removal.
The exit from the pouring bush should be radiused and match up with the
sprue entrance.
Pouring bush

Mould

Sprue

Figure 17.2 A properly designed pouring bush.

The practice of pouring directly down the sprue or the use of conical
shaped bushes which direct flow straight down the sprue is discouraged as
 Filtration and the running and gating of iron castings 247

not only will air and dross be entrained and carried down into the system,
but also the high velocity of the metal stream will result in excessive turbulence
in the gating system.

Sprue The metal stream exiting the bush narrows in diameter as it falls and
its velocity increases. To avoid air aspiration, the sprue should taper with
the smaller area at the bottom, but in mechanised green sand moulding
with horizontal parting, this is not possible since the sprue pattern must be
tapered with the larger area at the bottom to allow the pattern to be drawn
from the mould. Refractory ‘strainer cores’ may be sited at the base of the
sprue to restrict the flow of metal, allowing the sprue to fill quickly and
minimise air entrainment.

Sprue base Because stream velocity is at its maximum at the bottom of the
sprue it is important that a sprue base be used to cushion the stream and
allow the flow to change from vertical to horizontal with a minimum of
turbulence. Recommended sizes of the sprue base are, a diameter two–
three times the sprue exit diameter and depth equal to twice the depth of
the runner bar.

Gating ratio This is the relationship between the cross-sectional area of the
sprue, runner and gate. The system may be ‘pressurised’ or ‘unpressurised’.
A pressurised system is one in which the gates control the flow. A gating
ratio of [Link].7 is pressurised and is suitable for grey iron. An unpressurised
system having gating ratio [Link] is controlled by the sprue and is suitable for
ductile iron since the reduced turbulence in the runners and gates limits the
formation of dross.

Runners Runner cross-sections used in iron casting are usually rectangular

(with some taper to allow for moulding), with width to depth ratio of 1:2,
gates are taken from the bottom of the runner. It is presumed that the tall
runner allows slag and dross to collect in the upper part of the runner. The
distance between sprue and the first gate should be maximised for effective
inclusion removal. The runner should extend beyond the last gate so that
the first cold, slag-rich metal is trapped at the end of the runner.

Gates Ingates should ideally enter the mould cavity at the lowest possible
level to avoid turbulence associated with the falling metal stream but practical
moulding considerations often do not allow this. For grey iron castings, the
ingates are usually thin and wide with a width to height ratio of about 1:4.
The level of iron in the runner rises rapidly and is well above the top of the
gate before iron flows through the gate so minimising entry of slag into the
mould cavity. This shape of gate is easy to break so that grey iron castings
are easily separated from the runners. The gate is usually notched close to
the casting to break cleanly. One disadvantage of using gates to pressurise
the system is that the velocity of the iron is high as it enters the mould and
may cause erosion if the jet of metal impinges on core or mould wall.
248 Foseco Ferrous Foundryman’s Handbook

Feeder head This provides a reservoir of molten metal to compensate for any
metal shrinkage occurring during solidification of the casting.

Riser An opening leading from the mould cavity which relieves air pressure
in the mould cavity as it fills with metal. It also acts as a flow-off, allowing
cold or dirty metal to be removed from the mould cavity. If an open topped
feeder head is used, a riser is not necessary.

Gating vertically parted moulds

Figure 17.3 shows examples of vertically parted running systems. Ideally

the system shown in (a) may be considered best since each mould cavity is
filled uniformly from below, whereas the top-gated system (b) allows metal
to drop down the height of the mould with the possibility of turbulence and
erosion.

(a) (b) (c)

Figure 17.3 Examples of vertically parted moulding systems: (a) A sprue/runner

controlled system. (b) A runner/gate controlled system. (c) A multilevel system.
(from Elliott, R., Cast Iron Technology, 1988, Butterworth-Heinemann, reproduced
by permission of the publishers.)

The effect of ingate size on filling time

In a pressurised running system where the gates control the metal flow,
commonly used for grey iron castings, it is possible to calculate the total
ingate area needed to fill a casting in a certain time. Grey iron at normal
pouring temperatures is so fluid that small variations in temperature and
carbon equivalent have little effect on the fluidity. The factors that control
filling time are

the ingate area

the head of metal, for a bottom gated casting (Fig. 17.4a) this will be H at
the start of pour and h at the finish;
 Filtration and the running and gating of iron castings 249

the shape of the running system, in most cases the metal pours down the
sprue, turns 90° into the runner bar and a further 90° into the ingates,
slowing at each turn.

Top gated system

H h
M= h

(a)
c

(b)
Bottom gated system Side gated system

h M= h– c h
M= h–
p2
2 2c

c p
c

(c) (d)

Figure 17.4 (a) The head of metal varies from H at the start of pour to h at the
end. M is: (b) √h for a top gated system; (c) √(h – c/2) for a bottom gated system;
(d) √(h – p2/2c) for a side gated system.

These factors are taken into account in the formula

A = 8.12 × W
t×M
where A is total ingate area in cm2
t is pouring time in seconds, for casting and risers only
W is the weight of the casting in kg
M is related to the metal head (Fig. 17.4 b,c,d)

Example: A bottom gated casting of 25 kg is required to be poured in

15 seconds, h is 20 cm, c is 10 cm (Fig. 17.4)
M = √(20 – 10/2) = √15 = 3.87

total gate area needed = 8.12 × 25 = 3.5 cm 2

15 × 3.87
This simple formula is remarkably accurate and useful as an initial guide.
250 Foseco Ferrous Foundryman’s Handbook

Running and gating of iron castings is still the subject of controversy and
although the above principles are widely accepted, rules are frequently
broken and successful results still obtained. This is possible through improved
melting and ladle practice which produces cleaner metal, improvements in
the strength of green sand moulds, better cores and core coatings which
resist erosion more than before. Above all, the widespread use of ceramic
metal filters in running systems has not only eliminated most inclusions
from the metal but has allowed more attention to be paid to increasing the
utilisation of sand moulds and improving the yield of castings.

Filtration of iron castings

Filters were originally introduced to prevent non-metallic inclusions in the
liquid metal from entering the casting. While this is still their main function,
they are also used to simplify running systems allowing more castings to be
made in a mould and improving the yield of castings.

Inclusions in iron castings

The occurrence of non-metallic inclusions in castings is one of the most

widespread causes of casting defects encountered by the foundryman. The
presence of these inclusions has a deleterious effect on cast surface finish,
mechanical properties, machining characteristics and pressure tightness and
can lead to the scrapping of castings.
There are two main categories of inclusions:
1. Inclusions which are generated outside the mould and carried in with
the metal stream. The most common are:

Melting furnace slag

Ladle and launder refractories
Ladle slag
Flux residues
Desulphurisation slag
Alloying, nodularisation and inoculation reaction residues
Oxidation products
Contaminants and foreign objects.
2. Inclusions which are generated inside the mould. These include:

Loose sand
Mould and core erosion products
Loose mould and core coating products
Oxidation products generated in the running system
Undissolved in-mould inoculants.
 Filtration and the running and gating of iron castings 251

Traditionally, the incidence of inclusions has been controlled by:

The design of the ladle, e.g. tea-pot ladles

Design of the pouring basin
Gating system design, including slag traps spinners and whirl gates
Use of strainer cores (although these really only act as flow restricters to
choke the base of the down-sprue to allow the sprue to fill quickly so that
slag flotation occurs).

The use of filters is not a substitute for good melting and ladle practice but
it can revolutionise running and gating practice.

Types of filter
For iron filtration there are three types of filter

Filter cloths
Ceramic foam filters
Cellular filters.

Filter cloths are made of woven refractory glass cloths. Although they reduce
turbulence, they have a low filtering efficiency and a small open area (typically
25%) which acts as a severe choke in the running system.
Ceramic foam filters, Fig. 17.5 are the most efficient metal filters, they
have an open pore, reticulated structure with a very high volume of porosity
(over 90%) and a very high surface area to trap inclusions. The metal takes
a tortuous path through the filter effecting the removal of very small inclusions
by attraction and absorption to the internal ceramic pore surfaces.
Ceramic cellular filters have a ‘honeycomb’ structure with square section
passages (Fig 17.6) and, since the ceramic walls are thin, can have up to 75%
open area.
Ceramic metal filters work in several ways:

Coarse inclusions such as sand grains, large pieces of slag and dross
films are trapped on the front face of the filter.
After some metal has passed through the filter, a ‘cake’ of material forms
on its entry face which filters out finer particles. As the ‘cake’ builds up,
it reduces the flow of metal through the filter so that there is a limit to the
volume of metal that a particular size of filter can pass.
In addition to the physical filtration effect, there is a chemical attraction
between the inclusions and the ceramic of the filter causing small inclusions
to be trapped on the internal ceramic pore surfaces.
Finally, the smooth, non-turbulent metal flow through the filter reduces
the exposure of fresh metal to air, limiting oxide film formation.

Ceramic filters were first developed for the aluminium casting industry, for
use at temperatures up to about 900°C. Later, higher duty ceramics and
252 Foseco Ferrous Foundryman’s Handbook

Figure 17.5 Ceramic foam filters.

Figure 17.6 Cellular ceramic filters.

improved manufacturing techniques allowed their use to be extended to a

maximum temperature of about 1500°C, so that they could be used for
copper based castings and in iron foundries. The development of ceramic
materials suitable for the filtration of steel castings proved more difficult,
mainly because the rather low superheat associated with carbon and low
alloy steels caused freeze-off problems. This has now been overcome and
ceramic filters are widely used in steel sand foundries and investment casting
foundries for metal temperatures up to 1700°C. See Chapter 18.
 Filtration and the running and gating of iron castings 253

Since the early 1980s, growth in the use of filters in the foundry industry
has been spectacular. Metal filtration has been responsible for major
improvements in the general quality of castings as well as significant increases
in yield.

The benefits of filtration of iron castings

The obvious reason for filtering iron castings is to remove particulate

inclusions. Other benefits can then be obtained such as increased yield,
better machinability and better properties. The types of inclusions vary in
different types of iron.

Grey iron

The commonly found inclusions in grey iron are:

(1) Sand inclusions from loose moulding sand or sand erosion. Loose
sand lying in the running system will be trapped by an efficient filter
but sand lying in the mould cavity will not be affected by a filter. A
filter is not a substitute for proper sand practice.
(2) Slag arising from the cupola or furnace, ladle treatment slag, refractories
or ladle surface dross from oxidation of the metal. Casting defects
from slag material often are revealed only after machining, they are
frequently accompanied by gas holes. The slag material usually consists
of calcium silicates and oxides of iron, manganese, aluminium or
other metals. Cupola slag is rich in silica and contains CaO from
limestone.
(3) When coreless induction furnaces are used, oxides of silicon,
manganese etc. form and may be stirred into the melt by the turbulence
in the furnace, so that they can be carried over into ladles and castings.
(4) Inoculant particles are sometimes found, particularly when cast mould-
inoculation tablets are used.
(5) Manganese sulphide inclusions form in relatively high sulphur irons
as the metal cools and solidifies. They float to cause cope surface gas
holes but these inclusions are not removed by filtration, since they
form after the metal has passed the filter.

Those grey iron castings which are highly machined such as disc brake
rotors and cylinder blocks, have the greatest risk of having inclusion defects
exposed by machining and they benefit most from filtration.

Malleable iron

Two types of inclusion are often found in malleable iron castings:

(1) Sand inclusions which occur in the same way as in grey iron.
254 Foseco Ferrous Foundryman’s Handbook

(2) Slag, usually lime-silica slag and oxides. Because of the relatively
high furnace and pouring temperatures used, the slag particles remain
liquid and are difficult to filter out. Use of strainer cores is widespread
in malleable foundries to assist the flotation of slag in the sprue and
pouring cup.

Ductile iron

In addition to sand and slag inclusions similar to those found in grey iron,
ductile iron castings suffer from dross arising from the magnesium treatment
process. Dross defects are a major problem in ductile iron casting production.
The dross consists of magnesium silicate films and magnesium sulphide
particles.
Magnesium silicate is formed during the magnesium treatment by reaction
between MgO and SiO2. Where high silicon nodularisers are used such as
MgFeSi, more magnesium silicate is formed than when using pure magnesium
nodularisers. The magnesium sulphide is usually present as clouds of fine
individual particles. The silicate dross films and MgS clouds are usually
found together.
Dross defects in castings are caused by carry-over from the ladle but also
form during the casting process itself, since any exposure of the liquid
metal surface to air will allow formation of dross films. Turbulence in the
casting process increases dross formation.

The severity of dross defects is influenced by:

Magnesium content – as residual magnesium increases, the dross increases.

The initial sulphur content – high initial sulphur increases dross.
The cerium content of the iron – cerium oxidises preferentially, reducing
magnesium silicate formation.
Mould conditions – high moisture content increases dross.
Carbon and aluminium contents – both promote dross if increased.

Scrap castings

In the production of automotive ductile iron castings, scrap from inclusions

is typically around 2% or more. Much of the scrap is only found after
machining the casting. The cost of machining is often much higher than the
cost of the unmachined casting, for example a ductile iron crankshaft
approximately trebles in value after full machining. The additional cost of
filtration thus becomes easy to justify, so the majority of ductile iron castings
are now filtered.

Yield

Inclusions are a cause of lower yield, particularly in ductile and malleable

castings. This is because the running system must be designed to trap and
control slag and sand. Ductile iron running systems are designed for quiet
 Filtration and the running and gating of iron castings 255

mould filling and are long and deep to allow slags to float out. Typically,
20% of the melt output of a repetition ductile iron foundry is used to form
sprue, runners and slag traps. Use of an efficient filter in the running system
results in a valuable yield increase.

Productivity

The smaller running system possible with filters uses less pattern space so
that an extra casting may often be produced in the same pattern area.

Physical properties

Certain properties, particularly fatigue strength, are strongly influenced by

non-metallic inclusions. The effects of large inclusions such as dross films
are severe in both ferritic and pearlitic ductile iron, because notches are
formed at which failure can begin. Dross defects at as-cast surfaces will
reduce fatigue life substantially, reducing the fatigue limit by about 20–30%
(Table 17.1). This is of great significance for castings such as crankshafts
which fail by fatigue.

Table 17.1 The effect of surface imperfections on the fatigue properties of ductile
iron

No. Fatigue Endurance Fatigue Reduction Description of

limit ratio strength in fatigue surface
(N/mm2 ) reduction limit
factor (%)

1 270 0.35 – – Sound, fully nodular

graphite
2 220 0.32 1.23 19 Dross stringer and
isolated areas of flake
graphite
3 220 0.28 1.23 19 Fully nodular graphite
dross defects
4 182 0.24 1.48 33 Fine flake graphite
various dross defects

Machinability

The elimination of inclusions which are hard and abrasive means longer
tool life in machine shops, measurements have shown that tool wear when
turning grey iron is improved by more than 20% by filtration of the castings.
Even greater improvements have been found when turning ductile iron.
The improved surface cleanliness of filtered castings also means that
machining allowances can be reduced, since a smaller cut is needed to
remove surface defects.
256 Foseco Ferrous Foundryman’s Handbook

SEDEX ceramic foam filters

SEDEX foundry filters are ceramic foam filters, with an open-pore reticulated
structure, a very high volume of porosity (over 90%) and a very high surface
area to trap inclusions (Fig. 17.5). This structure provides a highly efficient
inclusion trapping system and gives smooth, non-turbulent mould filling
so that dross generation from re-oxidation is minimised. In order to use
SEDEX correctly it is necessary to understand how the filter works. There
are three mechanisms of inclusion capture and control:

1. Large inclusions and filmy dross inclusions are trapped on the front, or
‘active’ face of the filter. After some time a cake of inclusions forms
which itself collects inclusions and augments the filtering process.
Eventually this cake is so heavy that flow is stopped – the filter has
become blocked. Dirty metal causes early filter blockage, and dross-
forming alloys such as ductile irons are much more prone to this effect.
2. Small inclusions, point inclusions and micro-inclusions which escape
the preliminary filter layer penetrate into the centre of the SEDEX where
the extensive surface area and complex flow of metal ensure their
separation and adhesion onto the ceramic surface.
3. Metal leaving the SEDEX filter does so smoothly, without excessive
turbulence which might create inclusions on the clean side of the filter.

These three highly effective mechanisms mean that the iron is virtually
inclusion-free. Measurements have shown that inclusions as small as 2
micrometres are found trapped in used SEDEX filters.

Filter blockage

Eventually the collected inclusions can block the filter and stop the flow of
metal. To prevent this happening the filter must be chosen correctly, in
particular, the correct size of the filter must be chosen. Flow through a
SEDEX filter, leading to eventual blockage is illustrated schematically in
Fig. 17.7. SEDEX filters are supplied in a range of sizes and porosity to suit
the type and quantity of iron to be filtered (Fig. 17.8).

How to apply SEDEX filters

Figures 17.9 and 17.10 illustrate the use of SEDEX filters in horizontally
parted moulds. The filter area and size is dependent upon the quantity of
iron to be filtered. The choice of filter porosity is influenced by the type of
iron to be filtered (Table 17.2).

The running system should be designed according to Figs 17.9 and 17.10.
 Filtration and the running and gating of iron castings 257

Initial surge Filter blockage

Normal flow

Flow
rate

Time
Filter priming

Figure 17.7 Schematic pattern of flow through a SEDEX filter.

Coarse (10 ppi) Medium (20 ppi) Fine (30 ppi)

Figure 17.8 SEDEX filters of differing porosity.

Use the correct filter print to make the complete entry face of the filter
available for filtration.
For grey iron the filter area should be at least two times and for ductile
iron at least three times the downsprue area.
Position the filter as close as possible to the mould cavity.
The runner bar behind the filter should remain in the drag and be short,
direct and turbulence free.
The use of Foseco filter prints is strongly recommended to ensure correct
filter location (Figs 17.11, 17.12, 17.13 and 17.14).
For gating system calculation purposes the friction factor ξ is usually in
the range 0.2 to 0.6 according to gating system and mould geometry.
The effective pouring height is determined by the relationship between
cope height and ingate level.

SEDEX filter sizes and prints available are shown in Figs 17.11, 17.12, 17.13
and 17.14.
258 Foseco Ferrous Foundryman’s Handbook

Figure 17.9 The use of SEDEX filters in horizontally parted moulds.

• Recommended gating system area ratios

Downsprue : Runner : Ingate(s)

1, 1 1, 2
1, 0 RA IA
DA* : :

22,6 × W
*D A =
ξ×γ×t H

DA : Downsprue area [cm 2]

22,6 : Constant
W : Weight to be pour ed, including feeders (kg)
ξ : Friction factor (0.2 to 0.6 depending on gating system and mould geometry)
γ : Iron density [g/cm 2]
t : Pouring time [s]
H : Effective pouring height [cm]

Figure 17.10 Recommended gating system area ratios.

 Filtration and the running and gating of iron castings 259

Table 17.2 Choice of SEDEX filter porosity

Alloy or process Porosity Filtration capacity

(kg/cm2)

Ductile iron coarse (10 ppi) 1.4 to 2.0 max

Flake graphite iron medium (20 ppi) 4.0
Malleable iron fine (30 ppi) 4.0
Ductile Ni-resist coarse (10 ppi) 0.8 to 1.0 max
Simo coarse (10 ppi) 0.8 to 1.0 max
(Si-Mo ductile)
Inmold process coarse (10 ppi) 0.8 to 1.0 max

Note: The capacity of SEDEX ceramic filters is influenced by a variety of process

factors so the values given above are for guidance only.

Horizontal position with 2 runner bars

2
Type and dimensions Filter area (cm )
2 - 50 × 50 × 22 25,00
2 - 50 × 75 × 22 37,50

Figure 17.11 Filter print for horizontal position with two runner bars.

For the application of SEDEX filters on DISAMATIC moulding lines,

contact Foseco.

Cellular ceramic filters

A cellular ceramic is a refractory body which has been extruded into finely
divided, individual channels called cells. The production method allows
the external geometry of the filter and its internal geometry and density of
the cells to be varied (Fig. 17.6). Foseco supplies CELTEX cellular filters in
certain countries.
Cellular filters divide the metal stream into very small channels. Dross,
slag and sand inclusions are collected on the filter face and along the internal
filter cell walls. Analysis of the inclusions shows high concentrations of Mg,
Al, Si and S in ductile iron applications, and Al and Si in grey iron applications.
These examinations show that the impurities have a high affinity for the
260 Foseco Ferrous Foundryman’s Handbook

Vertical position
Type and dimensions Filter area (cm 2) Type and dimensions Filter Area (cm 2)
4 – 40 × 40 × 15 16,00 4 – 35 × 50 × 22 17,50
4 – 50 × 50 × 15 25,00 4 – 50 × 50 × 22 25,00
4 – 60 × 60 × 15 36,00 4 – 60 × 60 × 22 36,00
4 – 50 × 75 × 22 37,50
4 – 50 × 100 × 22 50,00
4 – 75 × 75 × 22 56,25
4 – 75 × 100 × 22 75,00
4 – 100 × 100 × 22 100,00
4 – 150 × 100 × 22 150,00

Figure 17.12 Filter print for vertical position.

Inclined position
Type and dimensions Filter area (cm 2) Type and dimensions Filter area (cm 2)
5 – 50 × 50 × 22 25,00 5 – 100 × 50 × 22 50,00
5 – 50 × 75 × 22 37,50 5 – 75 × 75 × 22 56,25
5 – 75 × 50 × 22 37,50 5 – 100 × 100 × 22 100,00
5 – 50 × 100 × 22 50,00 5 – 150 × 100 × 22 150,00

Figure 17.13 Filter print for inclined position.

 Filtration and the running and gating of iron castings 261

Vertical position combined with

downsprue base
Type and dimensions Filter area (cm 2)
6 – 35 × 50 × 22 17,50
6 – 50 × 50 × 22 25,00
6 – 50 × 75 × 22 37,50
6 – 75 × 50 × 22 37,50
6 – 75 × 75 × 22 56,25

Figure 17.14 Filter print for vertical position with downsprue base.

ceramic material, allowing particles much smaller than the cell openings to
be trapped. Filtration occurs by two mechanisms, physically removing
particles larger than the cell openings, and chemically attracting particles
smaller than the cell opening.
The cellular filter is less effective than the equivalent SEDEX filter.
The filters are positioned in the gating system so that the metal must flow
through the filter. A filter print is incorporated into the gating system pattern
to form a cavity for the filter. Cellular filters have tightly controlled external
dimensions, so that the filters accurately fit the print.
Choice of filter location follows the same principles as for ceramic foam
filters. Figure 17.15 shows possible filter positions. To function effectively,
the filter should not restrict the metal flow, to achieve this:
Filter frontal area = 4 to 6 times total choke area
This will achieve about the same pouring time as an unfiltered mould.

Flow rates

Metal composition, pouring temperature, metal head and filter position all
have an effect on flow rate. The following data is representative of flow
rates observed in practice.
262 Foseco Ferrous Foundryman’s Handbook

(a) (b)

Figure 17.15 Filter location for cellular ceramic filters. (a) Horizontal placement at
sprue base. (b) Vertical placement at sprue base.

Flow rate of iron through cellular filters

Filter Average flow rates

description
Grey iron Ductile iron
(kg/s) (lb/s) (kg/s) (lb/s)
2.0 × 2.0/100 5–6 10–12 3–4 6–8
3.0 × 3.0/100 10–14 24–30 7–9 15–20

Total poured weight

Non-metallic inclusions, collecting on the front face of a filter, can reduce
the open frontal area of the filter through which metal will pass. In ductile
iron applications the accumulation of dross and slag will eventually block
the filter. The amount of ductile iron that a filter can pass before blocking
depends on the foundry’s practice, the location of the filter and the metal
chemistry. The following guidelines have been developed from field trials.

Capacity of cellular filters

Filter Total flow of ductile iron before blockage

description (kg) (lb)
2.0 × 2.0/100 35–90 75–200
3.0 × 3.0/100 70–270 150–600

Filter blockage has not been observed in grey iron applications and the flow
rate requirements determine the size and number of filters required.
 Filtration and the running and gating of iron castings 263

Using filters to increase yield and number of

castings per flask

Example 1: Grey iron cylinder block casting

The foundry wanted to change from 2 to 4 castings per mould (Fig. 17.16).
Without a filter only two castings were possible because of the need for long
runners to trap inclusions. With two SEDEX filters, a more compact running
system was possible allowing four castings per mould with a 40% productivity
improvement.

TWO UNITS DOCKING

BEFORE USING SYSTEM or SEPARATE
SEDEX FILTERS SYSTEM
SEDEX: NONE SEDEX (20 ppi): 2 pcs
Pouring time: 14-16 sec. Pouring time: 17-18 sec.
Total Pour ed W eight: Total Pouring W eight:
125-130 kg Approx. 300 kg

Result
Difference of defect ratio due to inclusion (dr oss/slag) befor e the use of SEDEX
Filter and after the use of SEDEX filter.

• Repair ratio by welding:

Decreased to 1/5
• Defect ratio after machining:
Decreased to 1/10

By changing 2 Pcs/Flask to 4 Pcs/Flask, even though pouring yield is slightly

reduced due to a larger pouring basin, the productivity is raised by about 40%
– a significant improvement.

Figure 17.16 Use of a SEDEX filter increases the number of cylinder block castings/
flask from 2 to 11.

Example 2: Grey iron boiler casting

Use of SEDEX filters with a compact running system enabled two castings
per mould to be made (Figs 17.17a,b).
264 Foseco Ferrous Foundryman’s Handbook

Chassis 1200 × 850 Moule Inferieur

A
Foseco

COUPE AA A

B
C OUPE BB C
B

D D

C OUPE DD

Oe Projet est la Propriete de

foseco S.A. et ne doit pas etre
reproduit ou diffuse sans son
accord prealable.
C OUPE CC
© Foseco S.A.

(a)

Foseco

3 Sedex Fg
C C
50x50x15 20PPI
Sci = 1550 mm2 C C′ C
w

Sc = 1700 mm2
w

Sc = 1870 mm2
C1 C1 C1
D = 44 mm
en mm2
C2 C2
C2
C = 510 C ′= 680
C1 = 460 C′1= 610 C3 C3 C3
C2 = 390 C′2= 520
C3 = 290 C′3= 390
Sa = 90 mm2

ce Projet est la Propriete de foseco

S.A. et ne doit pas etre reproduit
ou diffuse sans son accord
prealable.
© Foseco S.A.

(b)

Figure 17.17 Use of a SEDEX filter with compact running system to cast 2 boiler
castings/flask: (a) one casting/flask; (b) two castings/flask.
 Filtration and the running and gating of iron castings 265

Using filters in vertically parted moulds

Example 1: Grey iron brake disc casting

Brake disc casting are frequently made on DISAMATIC moulding machines
with vertically parted moulds. The disc castings are subsequently machined
over a large part of their surface, any inclusions revealed by machining lead
to both high tool wear and scrap castings. Filtration is clearly desirable but
there are a number of problems of application:

Mould production rates are up to 350/hour.

Access to the open mould is not easy, filters must be placed by the coresetter
machine to avoid slowing the mould rate.
Space to locate filter prints is limited.

Two systems of placing filters are possible (Fig. 17.18a,b,c).

(a) (b)

(c)
Figure 17.18 Use of a SEDEX filter in vertically parted moulds. (a) unfiltered;
(b) filter placed at base of sprue; (c) filter placed in pouring bush.
266 Foseco Ferrous Foundryman’s Handbook

Figure 17.18a shows the unfiltered running system, as is usual with

DISAMATIC systems, the system is pressurised, being choked at the ingates.
Figure 17.18b shows the SEDEX filter placed in a ‘crush print’ at the base of
the sprue. In this position, the filter can be placed using the coresetter. The
choke is prior to the filter. Figure 17.18c shows the filter placed in the pouring
bush, here it is possible to place it by hand after the mould has been closed.
The sprue base acts as choke.
The pour time is shorter with the SEDEX filter than without.

Running system Pour time (s)

Unfiltered, choke at ingates 6.7

SEDEX filter at base of sprue 5.6
SEDEX filter in pouring bush 6.1

Both methods of using the filter are effective in removing inclusions.

Combined filter, feeder and pouring cup,

the KALPUR direct pouring system
The concept of direct pouring into the top of a mould cavity has long been
recognised as desirable, with the potential benefits of:

Improved yield
Simplified sprue, gating and feeding design
Reduced fettling costs.

Unfortunately, it was frequently found to introduce defects due to the

turbulent flow of the metal in all but the simplest of castings. In addition,
the impingement of high velocity metal streams caused erosion of moulds
or cores. These objections can be overcome by pouring the metal through a
ceramic foam filter situated at the base of an insulating pouring/feeding
sleeve, the KALPUR unit. Clean metal, free from turbulence and oxide, fills
the mould cavity and readily feeds the casting through the filter. Directional
solidification and casting soundness is promoted and gates and sprues made
unnecessary. The impingement problem is reduced because the metal velocity
is reduced as it passes through the filter (Fig. 17.19).

Application to horizontally parted moulds

For manual moulding and simple moulding machines, the open pouring
cup shape of Fig. 17.20 can be used; an example of its use is shown in Fig.
17.21. In horizontally parted automatic moulding lines, the KALPUR type
shown in Fig. 17.22 can be used as in Fig. 17.23.
 Filtration and the running and gating of iron castings 267

Figure 17.19 A schematic view of the cleaning and flow-smoothing effect of pouring
directly through a KALPUR unit.

Figure 17.24 shows a ductile iron vice base, casting weight 26 kg made on
a 20/24 Hunter moulding machine.
268 Foseco Ferrous Foundryman’s Handbook

φ 152

φ 120

φ 125 180
φ 90 150

Figure 17.20 Open KALPUR units for manual moulding. The units are supplied in
a range of sizes.

Figure 17.21 Use of the open cup KALPUR unit.

φ Da

φ du
φ Du

Figure 17.22 KALPUR unit for horizontally parted automatic moulding lines.
 Filtration and the running and gating of iron castings 269

Schnitt AA φ 30

Floating sleeve pattern

150
φ 50 SEDEX-filter

Cope
KSE 5/8/L10

20
φ 55

10

36
Drag

Figure 17.23 Use of the KALPUR unit in horizontally parted automatic moulding
lines.

Figure 17.24 Ductile iron vice base made on a Hunter machine using a KALPUR
unit.
270 Foseco Ferrous Foundryman’s Handbook

Before After
KALPUR KALPUR

Pour weight 34.5 kg 27.7 kg

Pouring time 14 s 8s
Yield 69.3% 86.4%
Total scrap 20% 0.9%
Grinding time 2 min 1.5 min

Application of direct pouring to vertically parted moulds

Increasingly complicated grey and ductile iron automotive castings are being
produced in vertical moulds for hydraulic, brake, suspension and transmission
systems. Components such as conrods, hubs, brake drums, flywheels, brake
discs, brake brackets and steering knuckles are now common production
castings in vertical moulds. Since many of these are safety castings, they
must usually be filtered. For ductile iron, feeders must also be incorporated
so that sound castings can be obtained.
Figure 17.25a shows a conventional pattern layout for a ductile iron car
hub casting made with three castings on the pattern plate. The running

(a)

(b)
Figure 17.25 Use of the KALPUR unit in vertically parted automatic moulding
lines: (a) conventional layout; (b) using KALPUR.
 Filtration and the running and gating of iron castings 271

system allows mould erosion due to high metal velocity. Figure 17.25b shows
a direct pouring system using a KALPUR direct pour unit at the top, and
below it, a KALMIN S feeder sleeve ensuring correct feeding of the lower
two castings. Casting yield increased from 61% to 78% and productivity by
33% while casting quality was improved.
Figure 17.26 shows a ductile iron differential housing, casting weight 30
kg made on a DISAMATIC 2070 moulding machine.

Figure 17.26 Ductile iron differential housing made on a DISAMATIC 2070 moulding
machine with a KALPUR unit.

Before After
KALPUR KALPUR

Pour weight 67 kg 44.5 kg

Moulding rate 215/h 230/h
Yield 44.1% 66.6%
Total scrap 8.5% 1.6%
Slag scrap 2.3% 0.02%
Grinding time 2.67 min 1.83 min

KALPUR direct pouring filter/feeder systems reduce mould and core erosion
and ensure that there is hotter metal in the feeders than in the castings,
giving ideal directional solidification from the casting to the feeder.
Chapter 18
Filtration and the running and
gating of steel castings

Introduction
The running and gating system carries out the following functions:

Controls the flow of metal into the mould cavity at the rate needed to
avoid cold metal defects in the casting
Avoids turbulence of metal entering the mould
Prevents slag and other inclusions from entering the mould
Avoids high velocity impingement of the metal stream onto cores or
mould surfaces
Encourages thermal gradients within the casting which help to produce
sound castings
Enables the casting to be separated from the running/gating system easily.

Controlling the flow of metal

Ideally the gating system should control the flow of metal into the mould.
If lip pouring ladles are used, this can be readily achieved since the pourer
is able to match the requirements of the gating system by altering the tilt of
the ladle. It is more difficult with bottom pour ladles. The flow rate from a
bottom pour nozzle/stopper rod system is determined by the nozzle size
and the metal height in the ladle according to the formula:

4 D( √ M 1 – √( M 1 – ∆M ))
t=
d 2 √2 gρπ

where t = pouring time (s)

D = mean ladle diameter (cm)
M1 = initial weight of metal (kg)
∆M = weight poured (kg)
d = nozzle diameter (cm)
g = acceleration due to gravity (981 cm/s2)
ρ = density of liquid steel (7.7 g/ml)
 Filtration and the running and gating of steel castings 273

(From C.S. Blackburn and B. Blair; Gating system design for bottom pour
ladles, 35th SCRATA Conference, May 1992)
The flow rate varies considerably depending on the level of metal in the
ladle. This is illustrated in Fig. 18.1 which shows the variation in pouring
time of seven 750 kg castings from a ladle initially containing 5750 kg of
steel. A gating system designed to accept the initial pouring rate of 750 kg
in 18.23 seconds will not be suitable to accept the last 750 kg which is
discharged from the ladle in 44.95 seconds, 2.5 times longer!

122 cm dia

Weight in Pouring
ladle (kg) time (sec.)
5750
5000 18.23
4250 19.68
3500 21.48
2750 23.96
2000 27.46
91 cm

5750 kg 1250 33.37

500 44.95

103 cm dia

3.8 cm dia

Figure 18.1 The variation in pouring time of seven 750 kg castings poured from a
bottom-pour ladle initially containing 5750 kg of steel.
(From Blackburn C.S. and Blair B. 35th SCRATA Conference, May 1992, courtesy
CDC.)

The nozzle/stopper rod system of a bottom pour ladle is an excellent

‘on/off’ valve but ideally should not be used as a flow control valve. Attempts
to use it to control flow result in breakup of the metal stream with consequent
risk of reoxidation of the steel and possible erosion of the stopper rod itself.
There is, however, no alternative if the gating system is to control the flow
rate. A compromise must be reached for each cast.
274 Foseco Ferrous Foundryman’s Handbook

Conventional running systems without filters

The elements of a running/gating system for a horizontally parted mould
are shown in Fig. 17.1.

Pouring bush The use of conical shaped bushes which direct flow straight
down the sprue is discouraged as not only will air and dross be entrained
and carried down into the system, but also the high velocity of the metal
stream will result in excessive turbulence in the gating system. Incorrect
alignment between nozzle and pouring cup will also cause metal splashing
(Fig. 18.2).

Figure 18.2 Simulation of metal splashing due to incorrect alignment of nozzle

and pouring cup.
(From Blackburn C.S. and Blair B. 35th SCRATA Conference, May 1992,
Courtesy CDC.)

A pouring bush designed as in Fig. 18.3 provides a larger target area, a

shape which minimises splashing and sufficient volume to accommodate
the maximum flow of metal obtained by opening the stopper rod fully thus
reducing the need to throttle. The exit from the pouring bush should be
radiused and match up with the sprue entrance.
 Filtration and the running and gating of steel castings 275

Figure 18.3 Simulation showing the elimination of metal splashing due to optimisation
of pouring cup design.
(From Blackburn C.S. and Blair B. 35th SCRATA Conference, May 1992,
Courtesy CDC.)

Sprue The sprue should ideally be full of metal throughout the pour to
avoid the possibility of slag being drawn into the mould cavity rather than
being retained in the pouring cup. The metal stream exiting the bush narrows
in diameter as it falls and its velocity increases. To avoid air aspiration, the
sprue should taper with the smaller area at the bottom.

Sprue base Because stream velocity is at its maximum at the bottom of the
sprue it is important that a sprue base be used to cushion the stream and
allow the flow to change from vertical to horizontal with a minimum of
turbulence. Recommended sizes of the sprue base are, a diameter two–
three times the sprue exit diameter and depth at least equal to the depth of
the runner bar (Fig. 18.4).

Runner bar Runner cross-sections used in steel casting may be square,

rectangular (with some taper to allow for moulding), or circular (if refractory
hollow-ware is used). It is presumed that a tall runner allows slag and dross
276 Foseco Ferrous Foundryman’s Handbook

Figure 18.4 Flat bottomed sump located at sprue base minimises turbulent
metal flow.
(From Blackburn C.S. and Blair B. 35th SCRATA Conference, May 1992,
Courtesy CDC.)

to collect in the upper part of the runner. The distance between sprue and
the first gate should be maximised for effective inclusion removal. The
runner should extend beyond the last gate so that the first cold, slag-rich
metal is trapped at the end of the runner.

Ingates The ingate should introduce clean metal into the mould cavity
with the minimum of turbulence. Low turbulence is best achieved by ensuring
that the gating system is unpressurised by sizing the gates to have a total
cross-section at least as large as that of the runner bar. The gating system
must always be full which is achieved by taking ingates of the top of the
runner bar and gating into the lowest part of the mould cavity. The ingate
should also be sized so that its modulus is smaller than that of the section
of the casting into which it enters. This prevents the formation of a small
shrinkage cavity at the ingate/casting interface. If this cannot be achieved,
the metal should be introduced into a feeder head.

Feeder head This provides a reservoir of molten metal to compensate for

any metal shrinkage occurring during solidification of the casting.
 Filtration and the running and gating of steel castings 277

Riser An opening leading from the mould cavity which relieves air pressure
in the mould cavity as it fills with metal. It also acts as a flow-off, allowing
cold or dirty metal to be removed from the mould cavity. If an open topped
feeder head is used, a riser is not necessary.

Gating ratio The gating system should be unpressurised with sprue : runner:
ingate areas 1 : 2 : 2. If a bottom pour ladle is used, then the sprue size
should be related to the nozzle size, with the nozzle bore controlling the
flow although, as mentioned above, this is generally not possible because of
the variable flow from a bottom pour ladle as the level in the ladle falls.
Gating of steel castings is not a precise science and each foundry develops
its own preferred methods.

The use of ceramic foam filters

The primary function of the gating system is to introduce metal quickly into
the mould without turbulence or non-metallic inclusions. Conventional gating
design is always a compromise. When effective ceramic foam filters suitable
for steel were developed, they led to immediate improvements in the quality
of steel castings since, not only do filters allow inclusions in the steel to be
eliminated more certainly than in the past, but they also reduce turbulence
in the metal stream allowing the two main functions of a gating system to
be readily achieved in a way that allows significant yield increases.

Inclusions in steel castings

The cleanliness of steel is, in the main, defined in terms of non-metallic
inclusions. Improved properties of castings result from reducing the number
of inclusions. Inclusions are introduced into steels during every stage of
processing, from melting, trimming additions, deoxidation, tapping and
teeming, as well as being generated within the mould itself. Inclusions are
generally termed:

Exogenous – which arise from external sources and are typically particles of
sand, refractory, moulding materials, melting and ladle slags and agglomerates
of any of these.

Indigenous – which originate from chemical reaction between elements within

the steel itself during melting and deoxidation, and reoxidation, which occurs
during tapping and teeming. Such inclusions are essentially silicates, oxides,
nitrides and sulphides, or more often complexes of these.
278 Foseco Ferrous Foundryman’s Handbook

It has been estimated that 1 tonne of steel contains between 1012 and 1015
indigenous oxide inclusions. Each inclusion has associated with it a localised
stress field, with inclusions of the order of 20 microns capable of nucleating
fatigue cracks, and effecting reduction in fracture toughness and ductility.
A study of macro-inclusions in steel castings from 14 foundries in the
USA found:

83% were reoxidation products which could be up to 5–10 mm in size

14% were from mould materials (sand and coating)
2% were slag
1% were refractory
1% were deoxidation products, always small in size, typically less than
15 µm
(Svoboda J.M. et al. Trans. AFS 95. 187–202 (1987))

Reoxidation occurs during pouring from the ladle and within the running
system as a result of turbulence, illustrating the importance of designing
running and gating systems with care.

Effect of inclusions

Exogenous inclusions: Usually found in areas of rapid solidification, on upper

cast surfaces or trapped under cores or adjacent to vertical walls. They can
be surface or subsurface, revealed on machining and may also be associated
with ‘gassy’ cavities. The size of these inclusions can vary from fine micro-
particles to gross macro inclusions up to about 75 mm diameter and 10 mm
thick. Exogenous inclusions give:

Impaired surface finish.

Poor machinability.
Reduced mechanical properties.

Indigenous inclusions: are very varied and are predominantly of the following
types:

Silicates – Usually glasslike inclusions which are relatively hard. Can have
an adverse effect on fatigue and impact properties. Silicates have a deleterious
effect on machinability by drastically reducing tool life.

Alumina – Primary source is from the deoxidation practice where aluminium

is used as the principal deoxidiser. In general appearance the alumina crystals
have a branched dendritic form. Alumina affects toughness and ductility,
reduces the ‘polishability’ of surfaces and is abrasive to cutting tools. It also
reduces fatigue strength. Aluminium is probably the most common deoxidant
 Filtration and the running and gating of steel castings 279

used in the steel industry, because of its strong deoxidising power and
relatively low cost. It is normally added to the ladle on tapping.

Zirconia and calcium oxide – if Zr or Ca-containing deoxidant is used.

Sulphides – These are commonly manganese sulphide which can have

different shapes, dependent on the conditions under which they form. Their
morphology is dependent on the oxygen level associated with the steel and
so is controlled by the amount of deoxidant, i.e. aluminium in the steel.

STELEX ZR ceramic foam filters

STELEX ZR filters (Fig. 18.5) are zirconia based foam filters designed to
withstand the high temperatures encountered in steel casting. In sand moulds
the filter is positioned in the running system using special filter prints designed
by Foseco. STELEX ZR filters can also be used in precision casting processes
such as investment casting where they are applied in specially designed
pouring cups (Fig. 18.6). Ceramic foam filters are the most efficient metal
filters, having an open pore, reticulated structure with a very high volume
of porosity (over 90%) and a very high surface area to trap inclusions. The
metal takes a tortuous path through the filter effecting the removal of very

Figure 18.5 STELEX ZR ceramic foam filters.

280 Foseco Ferrous Foundryman’s Handbook

Figure 18.6 STELEX ZR filter used in a ceramic pouring cup for investment casting.

small inclusions by attraction and absorption to the internal ceramic pore

spaces.
Ceramic metal filters work in several ways:

Coarse inclusions such as sand grains, large pieces of slag and dross
films are trapped on the front face of the filter.
After some metal has passed through the filter, a ‘cake’ of material forms
on its entry face which filters out finer particles. As the ‘cake’ builds up,
it reduces the flow of metal through the filter so that there is a limit to the
volume of metal that a particular size of filter can pass.
In addition to the physical filtration effect, there is a chemical attraction
between the inclusions and the ceramic of the filter causing small inclusions
to be trapped on the internal ceramic pore surfaces.
Finally, the smooth, non-turbulent metal flow through the filter reduces
the exposure of fresh metal to air, limiting oxidation and reoxidation of
the steel.

Ceramic filters were first developed for the aluminium casting industry, for
 Filtration and the running and gating of steel castings 281

use at temperatures up to about 900°C. Later, higher duty ceramics and

improved manufacturing techniques allowed their use to be extended to a
maximum temperature of about 1500°C, so that they could be used for
copper based castings and in iron foundries. The development of ceramic
materials suitable for the filtration of steel castings proved more difficult,
mainly because the rather low superheat associated with carbon and low
alloy steels caused freeze-off problems. This has been overcome with STELEX
ZR filters which are widely used in steel sand foundries and investment
casting foundries for metal temperatures up to 1700°C.
Since their development in the early 1980s, growth in the use of filters in
the foundry industry has been spectacular. Metal filtration has been
responsible for major improvements in the general quality of castings as
well as significant increases in yield.

The effectiveness of STELEX ZR filters

STELEX ZR ceramic foam filters remove macro inclusions such as ladle

slag, sand and furnace refractory eliminating the need for costly, time
consuming defect removal and welding repair. Work-in-progress associated
with castings awaiting rectification can be reduced significantly, resulting
in on-time deliveries. Micro inclusions such as deoxidation and reoxidation
products are removed by mechanical means at the surface of the STELEX
ZR filter or by adherence to the internal filter structure. Mechanical properties
and machinability of the castings are consequently enhanced and the need
for excessive machining allowances can be eradicated. Figure 18.7 shows
how filtration improves the cleanliness of an aluminium-killed steel.

Machinability tests

A programme of work has been undertaken at The Casting Development

Centre to evaluate different methods for improving the machinability of
casting. This work included a controlled experiment to measure the effect of
ceramic foam filters. A standard test, ISO 3685 – tool life testing for single
point turning tools, was used. Test bars of 100 mm diameter × 500 mm
length were produced in carbon and low alloy steel using both filtered and
unfiltered systems as shown in Fig. 18.8. The machinability of the test bars
was compared by turning them using silicon carbide tipped tools at a cut
depth of 2.5 mm until a flank wear of 381 µm was exhibited on the tool. The
results are summarised in Table 18.1. This work demonstrates that the use
of STELEX ZR filters can significantly increase both tool life and machine
productivity. These results have been substantiated by production foundries,
where improvements in the machinability of castings produced with STELEX
ZR filters and also a reduction in tool wear and machine downtime have
been achieved.
282 Foseco Ferrous Foundryman’s Handbook

Number of oxide inclusions (No./0.15 sq. mm.) 30

20

Unfiltered

10

Filtered

0
1.3 2.6 3.9 5.2 6.5 7.8 9.1 10.4
Size ( µ m) of Inclusions

Figure 18.7 Histogram comparing the cleanliness of an unfiltered and filtered

Al-killed steel (sulphides not included in total number of inclusions).

Machinability of
cast steels
Test specimen
150 mm dia.

150 mm

500 mm
Metallographic
sample
25

10 mm

100 mm Dia.

Figure 18.8 Machinability test piece.

 Filtration and the running and gating of steel castings 283

Table 18.1 The effect of filtration of steel on machinability

Steel Brinell Cutting Cutting Metal Extension in

grade hardness speed time removed tool life due
(m/min.) (min.) (mm2) to filtration

BS3100 unfiltered 150 150 18.8 1 726 000

A1 250 4.9 765 625
(carbon steel)
filtered 150 150 32.6 3 056 250 1.73 times
250 13.0 2 031 250 2.67 times
BS3100 unfiltered 280 150 13.2 1 331 250
BT2 250 2.0 312 510
(low alloy steel)
filtered 280 150 17.3 1 621 875 1.22 times
250 3.3 515 625 1.65 times

Flow control

A further function of STELEX ZR ceramic foam filters is to control metal

flow and to facilitate non-turbulent filling of the mould. Many STELEX ZR
filter users find this particular aspect of the filter’s performance to be as
beneficial as that of filtration; see Fig. 18.9. Non-turbulent flow reduces the
risk of reoxidation of the metal as it enters the mould cavity, reduces lapping
defects, promotes better surface finish, reduces air entrapment and offers
the potential for faster pouring times.

The application of STELEX ZR filters to steel castings

For the successful application of STELEX ZR to steel castings, five factors

need to be considered.

Support The filter must be adequately supported in the running

system.
Priming The application technique and metal temperature at
pouring must be suitable for good priming of the filter.

Figure 18.9 STELEX ZR filters control metal flow, reducing turbulence.

284 Foseco Ferrous Foundryman’s Handbook

Filter capacity The filter must be large enough to allow the casting
cavity and feeders to fill before a filter blockage
mechanism takes effect.
Metal flow rate The filter must be large enough to allow the casting to
fill in the optimum or required time.
Running system The running system should be unpressurised

Support The print the filter is positioned in must:

Adequately support the filter to ensure unnecessary stresses are not

incurred that may damage the filter.
Prevent the possibility of metal bypassing the filter.
Be of a correct size to give a snug fit for the filter, while allowing for the
minor dimensional tolerances relating to STELEX ZR.

A range of optimum resin filter prints and drawings are available for most
sizes of square filters. The STLX FPI and FP7 prints should be used wher-
ever possible and can be considered ‘universal’, a pictorial representation of
these prints is shown in Fig. 18.10. Other filter prints should only be used
where there are specific space restrictions and should be designed by Foseco.

Priming

It is important to ensure that the ladle temperature when the metal is poured
onto the filter is at least 80°C above the liquidus temperature of the alloy.
Typical minimum pouring temperatures are 1580°C for carbon steels and
1520°C for stainless steels. The filter should be positioned in a horizontal
position at the base of the sprue, as characterised by the STLX FPI and STLX
FP7 prints. With high alloy and stainless steel the level of superheat at
pouring is usually high enough to ensure good priming. For carbon and
low alloy steels extra temperature control and superheat may be required.
It is preferable to use one large filter than multiple small filters, this is
particularly important with carbon and low alloy steels.

Filter capacity

The amount of metal that can be passed through a STELEX filter depends
on many factors, including pouring temperature, alloy composition, dirtiness
of the metal and metallostatic pressure on the filter. An accurate prediction
of filter capacity for a specific alloy in a specific foundry is difficult to make.
If the metal is a high alloy or if a bottom pour ladle is used, high capacities
will be achieved.
The deoxidation practice has a big impact on filter capacity. The use of
calcium silicide and aluminium can be considered to be clean practice.
Titanium and zirconium will dirty the metal and reduce the capacity of the
filter. Tables 18.2 and 18.3 provide a guide to capacities that can be achieved.
 Filtration and the running and gating of steel castings 285

STLX FP1

STLX FP7

Figure 18.10 Recommended STELEX filter prints.

286 Foseco Ferrous Foundryman’s Handbook

Metal flow rate

The same factors that affect filter capacity affect the metal flow rate through
the filter. Tables 18.2 and 18.3 provide a guide to flow rates that can be
achieved. Naturally, if the filter starts to become blocked, the flow rate will
be reduced.

Running system design

The running system should be unpressurised after the filter and as simple
as possible, long running systems and other methods of inclusion and
flow control are not required and should be avoided. The filter area
needs to be at least 4.5 times the sprue base area. The following ratios are
recommended.

Sprue area Filter area Runner area Ingate area

1 4.5 1.1 1.2

Minimum

Determination of the metal capacity and flow rates for STELEX filters

Selection of the correct STELEX filter for a given application is dependent

on several factors:

Alloy carbon/low alloy or high alloy/stainless

 factors which define
Ladle practice lip pour or bottom pour  fluidity/cleanliness

Deoxidation practice Zr/Ti or CaSi/Al 

The amount of metal which must pass the filter

The mould fill time or metal flow rate required

Use Tables 18.2 and 18.3 as a guide to the selection of the correct filter. Note
that if it is intended to apply more than one filter to a casting, special
assistance from Foseco should be sought.

KALPUR ST direct pour unit

The KALPUR ST direct pour unit combines a filter and feeder sleeve into a
single unit, Fig. 18.11. This unit replaces an existing feeder on a casting and
then serves as a combined sprue, filter and feeder. Metal is introduced into
Table 18.2 Typical capacity of STELEX filters for carbon/low alloy steels

Pouring method Lip pour Bottom pour

Deoxidation practice Zr/Ti CaSi/Al Zr/Ti CaSi/Al
STELEX type
Capacity Flow rate Capacity Flow rate Capacity Flow rate Capacity Flow rate
kg kg/sec kg kg/sec kg kg/sec kg kg/sec

50 × 50 × 20 35 2.5 53 2.5 46 3.1 68 3.1

55 × 55 × 25 45 3.0 68 3.0 59 3.8 88 3.8
75 × 75 × 25 85 5.6 128 5.6 111 7.0 166 7.0
100 × l00 × 25 150 10.0 225 10.0 195 12.5 293 12.5
125 × 125 × 30 235 15.6 353 15.6 306 19.5 458 19.5
150 × 150 × 30 340 22.5 510 22.5 442 28.1 663 28.1
50 dia. × 25 30 2.0 45 2.0 39 2.5 59 2.5
60 dia. × 25 45 2.9 68 2.9 59 3.6 88 3.6
70 dia. × 25 60 3.8 90 3.8 78 4.8 117 4.8
90 dia. × 25 95 6.4 143 6.4 124 8.0 185 8.0
100 dia. × 25 120 7.8 180 7.8 156 9.8 234 9.8
125 dia. × 30 185 12.2 278 12.2 241 15.3 361 15.3
150 dia. × 30 265 17.7 398 17.7 345 22.1 517 22.1
Table 18.3 Typical capacity of STELEX filters for high alloy/stainless steels

Pouring method Lip pour Bottom pour

Deoxidation practice Zr/Ti CaSi/Al Zr/Ti CaSi/Al
STELEX type
Capacity Flow rate Capacity Flow rate Capacity Flow rate Capacity Flow rate
kg kg/sec kg kg/sec kg kg/sec kg kg/sec

50 × 50 × 20 46 3.5 69 3.5 61 4.4 102 4.4

55 × 55 × 25 59 4.2 88 4.2 79 5.3 132 5.3
75 × 75 × 25 111 7.8 166 7.8 149 9.8 249 9.8
100 × l00 × 25 195 14.0 292 14.0 263 17.5 439 17.5
125 × 125 × 30 306 21.8 459 21.8 412 27.3 687 27.3
150 × 150 × 30 442 31.5 663 31.5 597 39.4 995 39.4
50 dia. × 25 39 2.8 58 2.8 53 3.5 88 3.5
60 dia. × 25 59 4.1 88 4.1 79 5.1 132 5.1
70 dia. × 25 78 5.3 117 5.3 105 6.7 176 6.7
90 dia. × 25 124 9.0 186 9.0 167 11.2 278 11.2
100 dia. × 25 156 10.9 234 10.9 211 13.7 351 13.7
125 dia. × 30 241 17.1 361 17.1 325 21.4 541 21.4
150 dia. × 30 345 24.8 517 24.8 465 31.0 775 31.0
 Filtration and the running and gating of steel castings 289

Figure 18.11 KALPUR ST direct pour units.

the mould cavity by pouring directly through the KALPUR ST unit, as

illustrated in Fig. 18.12.
The sleeve performs initially as a pouring cup while the ceramic foam
filter absorbs the impact energy of the metal stream, preventing sand erosion
as the mould fills. As pouring progresses, the filter prevents non-metallic
inclusions from entering the mould cavity and ensures non-turbulent flow.
On completion of pouring, the KALPUR ST unit acts as an efficient
feeder. The metal in the KALPUR ST unit is the last to be poured and,
being the hottest metal in the casting system, provides ideal conditions for
directional solidification. Benefits of using the KALPUR ST direct pour system
include:

Elimination of the runner system, increasing yield, reducing cleaning

and allowing more castings per flask
Efficient filtration
Low metal velocity in the mould, reducing erosion of sand and coating
Low turbulence of metal in mould reducing reoxidation
Efficient feeding through directional solidification
Lower casting temperature.

Computer solidification simulation

Extensive studies have been conducted to evaluate the thermal gradient

290 Foseco Ferrous Foundryman’s Handbook

Figure 18.12 The KALPUR ST unit is placed as close to the casting cavity as
possible to preserve laminar flow from the filter.

differences between conventional systems and direct pour practice. The

examples shown in Fig. 18.13 illustrate the benefits obtained by the application
of a KALPUR ST unit to a valve casting.

Productivity improvements

The KALPUR ST unit can be applied to a range of casting types. Impellers,

for example, are usually cast with heavy running systems up into the centre
of the casting body. A KALPUR ST unit can be applied as a central feeder,
(see Fig. 18.14) and the metal poured directly into the casting. More castings
can thereby be obtained from a ladle of metal. The promotion of excellent
directional solidification has also resulted in the reduction in size or total
removal of feeders on the perimeter of the casting (Fig. 18.15).

Selecting the proper size and positioning the KALPUR unit

There are two basic designs of unit, one for side feeder application and one
for top feeder application, Fig. 18.16a, b. KALPUR ST units are supplied
in a range of sizes (Fig. 18.11) having capacity from around 30 kg to over
250 kg of carbon or low alloy steel and more for high alloy steel (Table 18.4).
Metal temperature, ladle practice (bottom versus lip pour), melting practice
(cleanliness), metal deoxidation, and steel composition all determine the
amount of metal that can be poured before filter blockage that would prevent
proper pouring or adequate feeding after the mould is filled.
Factors resulting in low capacities and flow rates:
 Filtration and the running and gating of steel castings 291

Temperature Distribution in
the Casting at the End of
Filling … Conventional Gating

(a)

Temperature Distribution in
the Casting at the End of
Filling … Direct Pour

(b)

Figure 18.13 (a) Temperature distribution within a valve casting immediately after
filling through a conventional gating system. Large areas are much colder than the
last metal poured. (b) Temperature distribution within a valve casting immediately
after filling through a KALPUR unit shows higher overall temperature, illustrating
reduced heat loss when direct pour units are used. (This figure is reproduced in
colour plate section.)
292 Foseco Ferrous Foundryman’s Handbook

Figure 18.14 KALPUR unit used as a central feeder.

Conventional method

New KALPUR ST direct pour system

Figure 18.15 The KALPUR ST direct pour system eliminates the conventional
running system and reduces feeder requirements.
 Filtration and the running and gating of steel castings 293

(a) (b)

Figure 18.16 KALPUR ST unit: (a) for side feeder application; (b) for top feeder
application.

Table 18.4 Capacity and flow rates of KALPUR ST units

Capacity range (kg) Nominal flow rate (kg/s)

(at 300 mm metallostatic pressure)
KALPUR High level of Low level of Carbon Stainless
size deoxidation deoxidation steel steel
products products

KALPUR 50 30 90 2.0 3.0

KALPUR 75 65 200 4.4 6.6
KALPUR 100 120 355 7.8 11.7
KALPUR 125 185 550 12.2 18.4
KALPUR 150 265 795 17.7 26.5

Deoxidation with Zr, Ti

Molten metal containing large quantities of inclusion material
Low metal pouring temperatures
Low metallostatic pressure on filter.

A minimum pouring temperature of 1580°C for carbon steel and 1520°C for
stainless steel is recommended to ensure priming.
294 Foseco Ferrous Foundryman’s Handbook

The KALPUR ST system acts as an insulating feeder sleeve of similar

size. The presence of a filter in a unit does not affect the feed efficiency. An
efficient anti-piping compound should be applied to the KALPUR ST
immediately after pouring is finished. Recommended materials are:

KALPUR ST 50 and 75 FERRUX 16

KALPUR ST 100, 125, 150 FERRUX 707F

The KALPUR ST system can incorporate a neck-down feature for reduced

fettling and the possibility of knock-off.
To select the proper KALPUR ST unit for a given application, two criteria
must be met: feeding capacity and filtration performance. First, one must
determine which unit sleeve is large enough to satisfy the feeding requirement.
Next, the filter in that unit must be evaluated for filtration capacity and
flow rate. If the filtration capacity or flow rate is inadequate for the mould
weight, cleanliness of the metal, or maximum pour time, a bigger KALPUR
unit having a larger filter should be selected.
The KALPUR unit is rammed into position. It should be placed as close
to the casting as possible and ideally should be located no more than
150 mm above the impact area (Fig. 18.12).

Cost savings through the use of

STELEX and KALPUR
An analysis of steel casting cleaning costs has been undertaken by the Institut
fur Giessereitechnik, Dusseldorf covering a number of steel foundries
representing 10% of German steel casting production. The work showed
that the cleaning costs associated with a casting can amount to 40% of its
total manufacturing cost. Cleaning costs were further broken down as follows:

Reworking of casting defects 37%

(inclusions, penetration, shrinkage)
Removal of feeding and gating systems 27%
Removal of rough surfaces 20%
(fins, flash and burn-on)

Others 16%

Of these costs
52% were considered as avoidable
31% were considered as possibly avoidable
17% were defined as unavoidable
 Filtration and the running and gating of steel castings 295

The use of STELEX ZR filters and KALPUR ST units enables major reductions
to be made in the avoidable cleaning costs through simplification of gating
and feeding systems and the removal of inclusions.
Chapter 19
Feeding of castings

Introduction
During the cooling and solidification of most metals and alloys, there is a
reduction in the metal volume known as shrinkage. Unless measures are
taken which recognise this phenomenon, the solidified casting will exhibit
gross shrinkage porosity which can make it unsuitable for the purpose for
which it was designed. To some extent, grey and ductile cast irons are
exceptions, because the graphite formed on solidification expands and can
compensate for the metal shrinkage. However, even with these alloys,
measures may need to be taken to avoid shrinkage porosity.
To avoid shrinkage porosity, it is necessary to ensure that there is a sufficient
supply of additional molten metal, available as the casting is solidifying, to
fill the cavities that would otherwise form. This is known as ‘feeding the
casting’ and the reservoir that supplies the feed metal is known as a ‘feeder’,
‘feeder head’ or a ‘riser’. The feeder must be designed so that the feed metal
is liquid at the time that it is needed, which means that the feeder must
freeze later than the casting itself. The feeder must also contain sufficient
volume of metal, liquid at the time it is required, to satisfy the shrinkage
demands of the casting. Finally, since liquid metal from the feeder cannot
reach for an indefinite distance into the casting, it follows that one feeder
may only be capable of feeding part of the whole casting. The feeding
distance must therefore be calculated to determine the number of feeders
required to feed any given casting.
The application of the theory of heat transfer and solidification allows
the calculation of minimum feeder dimensions for castings which ensures
sound castings and maximum metal utilisation.

Natural feeders
Feeders moulded in the same material that forms the mould for the casting,
usually sand, are known as natural feeders. As soon as the mould and
feeder have been filled with molten metal, heat is lost through the feeder
top and side surfaces and solidification of the feeder commences. A correctly
dimensioned feeder in a sand mould has a characteristic solidification pattern:
that for steel is shown in Fig. 19.1 the shrinkage cavity is in the form of a
cone, the volume of which represents only about 14% of the original volume
 Feeding of castings 297

of the feeder, and some of this volume has been used to feed the feeder
itself, so that in practice only about 10% of the original feeder volume is
available to feed the casting. The remainder has to be removed from the
casting as residual feeder metal and can only be used for re-melting.

Figure 19.1 Solidification pattern of a feeder for a steel casting (schematic).

Aided feeders – feeding systems

If by the use of ‘feeding systems’ the rate of heat loss from the feeder can be
slowed down relative to the casting, then the solidification of the feeder will
be delayed and the volume of feed metal available will be increased. The
time by which solidification is delayed is a measure of the efficiency of the
feeding aid. The shape of the characteristic, conical, feeder shrinkage cavity
will also change and in the ideal case, where all the heat from the feeder is
lost only to the casting, a flat feeder solidification pattern will be obtained
(Fig. 19.2). As much as 76% of an aided feeder is available for feeding the
casting compared with only 10% for a natural sand feeder. This increased
efficiency means greatly reduced feeder dimensions with the following
advantages for the foundry:

A greater number of castings can be produced from the given weight of

liquid metal
Smaller moulds can be used, saving on moulding sand costs
A reduction in the time needed to remove the feeder from the casting is
possible
More castings can be mounted on the pattern plate and thus into the
mould
Less metal is melted to produce a given volume of castings
Maximum casting weight potential is increased
Smaller feeders mean less fettling time and cost.
298 Foseco Ferrous Foundryman’s Handbook

Figure 19.2 Ideal feeder solidification pattern where all the heat from the feeder
has been lost to the casting (schematic).

Feeding systems

Side wall feeding aids are used to line the walls of the feeder cavity and so
reduce the heat loss into the moulding material. For optimum feeding
performance, it is also necessary to use top surface feeding aids. These are
normally supplied in powder form and are referred to as anti-piping or hot-
topping compounds. Figure 19.3 illustrates how the use of side wall and top
surface feeding aids extends the solidification times. Now, however, suitable
insulating discs (lids) are increasingly being used in place of hot-topping
compounds for environmental reasons. For mass production castings 90%
of feeders are closed or ‘blind’.

Solidification time
Sand wall KALMINEX sleeve
open top open top

29.2 39.8
minutes minutes

Sand wall KALMINEX sleeve

FERRUX cover FERRUX cover

44.9 73.1
minutes minutes

Figure 19.3 Extension of solidification times with side wall and top surface feeding
products for a steel cylinder 250 mm dia. and 200 mm high.
 Feeding of castings 299

Calculating the number of feeders – feeding distance

A compact casting can usually be fed by a single feeder. In many castings of

complex design the shape is easily divided into obvious natural zones for
feeding, each centred on a heavy casting section separated from the remainder
of the casting by more restricted members. Each individual casting section
can then be fed by a separately calculated feeder and the casting shape
becomes the main factor which determines the number of feeders required.
In the case of many extended castings however, for example the rim of a
gear wheel blank, the feeding range is the factor which limits the function
of each feeder and the distance that a feeder can feed, the ‘feeding distance’,
must be calculated.
The feeding distance from the outer edge of a feeder into a casting section
consists of two components:

The end effect (E), produced by the rapid cooling caused by the presence
of edges and corners.
An effect (A), produced because the proximity of the feeder retards freezing
of the adjacent part of the casting (Fig. 19.4).

Where a casting requires more than one feeder the distance between feeders
is measured from the edge of the feeder, not from its centre; and when the
feeder is surrounded by a feeder sleeve the distance between feeders is
measured from the outside diameter of the sleeve.
The effective distance between feeders can be increased by locating a
chill against the casting mid-way between the two feeders (X1) and the
natural end effect can be increased by locating a chill against the natural
end (X). Chills should be of square or rectangular section with the thickness
approximately half the thickness of the section being chilled.
There are therefore four possible situations:

Sections with natural end effect only (A+E)

Sections with natural end effect plus end chill (A+E+X)
Sections with no end effect (A)
Sections with no end effect plus chill (A+X1)

Figure 19.4 shows the basis for calculating feeding distance in steel castings
and all other ferrous alloys which freeze white, e.g. malleable and high
alloy irons.

Ductile and grey iron castings

Alloy composition, casting section, mould materials and mould hardness

all play a part in determining the actual feeding distance. The following
tables are guidelines for green sand moulds having mould hardness 90° B
scale, variations from these conditions will result in other feeding distances.
300 Foseco Ferrous Foundryman’s Handbook

2.5t 2t 2t 2t 2t 2.5t

A A E

5t 5t 5t 5t

X1 + A A+E+X
(a)

1.5t 1.5t 6 t

A A+E
6 t+t 6 t+t

X1 + A A+E+X
(b)

Figure 19.4 Feeding distance in steel castings. (a) Plate (width: thickness >5:1.
(b) Bar (width: thickness < 5:1).

Table 19.1 Feeding distance factor (FD) for ductile iron castings

Carbon equivalent (%) Feeding distance factor (FD)

4.1 6.0
4.2 6.5
4.4 7.0
4.6 9.0

Non-ferrous castings

Table 19.3 gives a feeding distance factor for some of the non-ferrous alloys;
this factor is used in the calculation of an approximate feeding distance.
 Feeding of castings 301

Table 19.2 Feeding distance factor (FD) for grey iron castings

Carbon equivalent (%) Feeding distance factor (FD)

3.0 6.8
3.4 7.7
3.9 8.8
4.3 10.0

Table 19.3 Feeding distance factor for non-ferrous alloys

Casting alloy Feeding distance factor (FD)

Al (99.99%) 2.50
Al 4.5% Cu 1.50
Al 7% Si 1.50
Al 12% Si 2.50
Cu pure 2.00
Cu 30% Ni 0.50
Brass 1.25
Al Bronze 1.25
Ni Al Bronze 0.50
Sn Bronze 0.75

The feeding distance in Fig. 19.4 is calculated using the feeding distance
factor (FD) in the following manner:

Bars (width : thickness <5:1) Plates (width : thickness >5 : 1)

A = 1.5t × FD 2t × FD
E = (6√t – 1.5t) × FD 2.5t × FD
X = t × FD 0.5t × FD
X1 = (6√t – 0.5t) × FD 3t × FD

The calculation of feeder dimensions

The majority of foundrymen, even today, decide on feeder dimensions on
the basis of experience. However, the application of calculation based on
established theory and experimental data ensures the most efficient design
of natural and aided feeders. In this section some guidance is given for
calculating feeder dimensions from first principles whereas on page 334
there is a description of the various aids such as tables, nomograms, and
computer programs developed by Foseco to make the determination of
feeder dimensions much easier.

The modulus concept

Although this concept has some shortcomings it is, with the exception of
302 Foseco Ferrous Foundryman’s Handbook

computer programs, the most widely used acceptable and accurate method
for calculating feeder dimensions.
The solidification time of a casting section is given by Chvorinov’s rule:

t c = kVc2 /A c2 = kM c2 (1)

where
tc is the solidification time of the casting section;
Vc is the volume of the casting section;
Ac is the surface area of the casting section actually in direct contact
with the material of the mould;
k is a constant which is governed by the units of measurement
being used, the thermal characteristics of the mould material and
the properties of the alloy being cast.
Mc is the ratio of the volume of the casting section to its cooling surface area
and is known as the casting’s Geometric Modulus. It is expressed in units of
length:
Mc = Vc/Ac (2)
The Modulus formulae for some common casting shapes are shown in
Fig. 19.5.
Foundrymen, for the purpose of feeder size determination, are generally
not directly interested in the exact solidification time but only that the feeder
solidifies over a longer time than the casting. Having calculated the modulus
of the casting section therefore, the modulus of the feeder is calculated as
MF = 1.2 × Mc (3)
where MF is the modulus of the feeder required to feed a casting section
having a modulus of Mc. This equation applies to natural feeders for most
alloys. For grey and ductile iron castings because there is a graphite expansion
phase during solidification the safety factor of 1.2 is considerably reduced.

The modulus extension factor (MEF)

The object of using feeding aids is to slow down the rate of heat loss from
the surface of the feeder. It is possible to calculate how the improved thermal
properties of the feeding aid compared with sand can reduce the feeder
size.
From equation (1) the solidification time for a feeder is expressed as

t F = kM F2 (4)

The constant k is composed of two parts:

the thermal characteristics of the mould material surrounding the whole

of the feeder the properties of the metal within the feeder.
 Feeding of castings 303

Shape Dimensions Modulus

L
(a) Cube L
L
6

L Side = L
DH
2D + 4H
(b) Cylinder Note: If H = D the modulus is
D Diameter = D D
H Height = H 6
T
D DT
(c) Disc 2D + 4T
Diameter = D
Thickness = T

T
TWL
(d) Bar or plate L 2 (TW + W L + LT )
Length = L Width = W
Thickness = T
D
4
(e) Endless cylinder Note: Because radial heat
(ends terminated flow is faster than that from
D a flat surface, calculated
by another
part of casting) moduli for endless
cylinders may be reduced
Diameter = D by multiplying by 0.85

(f) Endless plate

(terminated on all T T
sides by another 2
part of casting) Thickness = T

(g) Endless bar W

(ends TW
terminated by T
2(W + T )
another part of Thickness = T
casting) Width = W

D1
D2
(h) Endless hollow D1 – D 2
cylinder OD = D1 =T
T 4 2
Dia. core = D2
Wall thickness = T

D1
T D2 (D 1 – D 2 )
(i) Annulus 2(D 1 – D 2 ) + H
H
OD = D1 TH
=
Dia. core = D2 2(T + H )

Figure 19.5 Modulus formulae for some common shapes.

304 Foseco Ferrous Foundryman’s Handbook

k can therefore be reduced to its two constituent components so that

k = cf 2
where
c is a constant depending on the properties of the metal being cast;
f 2 is a constant depending on the properties of the mould material.
There is no significance in the fact that this constant is expressed as a square
other than mathematical convenience. Equation (4) can now be rewritten
tF = C( fMF)2 (5)
The expression (fMF) is known as the Apparent Modulus and f as the Modulus
Extension Factor of the mould material surrounding all of the feeder’s surfaces.
This approach provides a quantitative means of evaluating and comparing
different feeding aids which should be designed to have the maximum
possible Modulus Extension Factor compatible with the other properties
necessary for a satisfactory product. In practice it is not customary to determine
the absolute values of the constants relating solidification time to the modulus
as these are seldom of interest.
Of greater concern is the improvement which can be expected from a
variety of feeder lining materials when compared with the results obtained
from the same size of feeder lined with the conventional moulding material
– sand. For this purpose the Modulus Extension Factor (f) for sand is equated
to unity and this serves as a basis for comparing other materials.

Example

Using this information it is possible to consider as an example a sound steel

casting fed by one cylindrical feeder moulded in sand with a radius (r) of
16 cm and a height (h) of 32 cm. Because the feeder is attached to the steel
casting, the bottom circular face of the feeder is a non-cooling face and the
Geometric Modulus (MF) of the feeder moulded in sand therefore is

πr 2 h
M F (sand) = (6)
2 π rh + π r 2
because in the example chosen h = 32 cm; r = 16 cm and h = 2r
MF (sand) = 2r/5 = 6.4 cm
If in place of sand, a feeding aid system with a Modulus Extension Factor (f)
of, for example 1.6 were to be used to line the feeder cavity and to cover the
top surface of the molten steel in the feeder, then the aided feeder would
remain liquid for the same time as the sand lined feeder if the Apparent
Modulus of the aided feeder were to be equal to the Geometric Modulus of
the sand lined feeder, i.e. for the equal solidification times:
MF (sand) = fMF (aided) (7)
 Feeding of castings 305

so that the Geometric Modulus of the aided feeder would be

MF (aided) = MF (sand)/f (8)
or in the case chosen as the example
MF (aided) = 6.4/1.6 = 4.0 cm
a cylinder where h = 2r having a Geometric Modulus of 4.0 cm would be
20 cm diameter × 20 cm high.
The use of the feeding aid system quoted in this example therefore
represents a reduction of approximately 75% in the feeder volume needed
to achieve the same solidification time. The values for Modulus Extension
Factors vary widely according to the type of feeding aid, the size and shape
of the feeder sleeve and even the shape of the casting being fed. Foseco
publishes tables showing the dimensions of sleeves available together with
the Geometric and approximate Apparent Modulus of each sleeve.

Determination of feeding requirements

Steel, malleable iron, white irons, light alloys and

copper based alloy castings
1. Calculate casting and feeder modulus
(a) Divide the casting into sections and determine the important volume-
to-surface ratios according to Fig. 19.5.

Modulus = Volume
Cooling surface
M c = Vc /A c (9)
(b) Determine the required feeder modulus (Mf) using the factor 1.2
MF = Mc × 1.2 (10)
(c) Provisional determination of the feeder from the feeder sleeve tables
published by Foseco Companies.

2. Calculate the feed volume requirement

Feeders, whose dimensions obtained above satisfy modulus requirements
do not necessarily always satisfy the total feed metal demand of the casting
section. This must always be checked and if the feeder is found to contain
insufficient available feed metal the feeder dimensions must be increased.
Generally it is preferable to retain the feeder diameter dictated by modulus
considerations and increase the feeder height. Sometimes it is more convenient
to increase the feeder diameter in place of or as well as height. Never reduce
the feeder diameter below that necessary to meet modulus requirements.
The following data is necessary.
306 Foseco Ferrous Foundryman’s Handbook

(a) The proportion of feed metal available from the feeder which meets
modulus requirements (C%). Safe values although not necessarily
the most efficient are:
33–50% if a Foseco sleeve is being used
14–16% if it is a live natural feeder (i.e. one through which metal
has to flow before it reaches the casting cavity in the
mould)
10% for other natural feeders.
(b) The shrinkage of the alloy to be cast (S%).
The shrinkage values for the principal casting alloys are given in
Table 19.4.
(c) The weight of metal in the feeder under consideration (Wf).
The total weight of casting (Wc) which can be fed from a feeder of
weight WF is
Wc = C/100 × 100/S × WF (11)
If the total weight of casting section (WT) which requires feeding is greater
than Wc then increase the dimensions of the feeder until

Table 19.4 Shrinkage of principal casting alloys

Casting alloy Shrinkage (%)

Carbon steel 6.0

Alloyed steel 9.0
High alloy steel 10.0
Malleable iron 5.0
Al 8.0
AlCu4Ni2Mg 5.3
AlSi12 3.5
AlSi5Cu2Mg 4.2
AlSi9Mg 3.4
AlSi5Cu1 4.9
AlSi5Cu2 5.2
AlCu4 8.8
AlSi10 5.0
AlSi7NiMg 4.5
AlMg5Si 6.7
AlSi7Cu2Mg 6.5
AlCu5 6.0
AlMg1Si 4.7
AlZn5Mg 4.7
Cu (pure) 4.0
Brass 6.5
Bronze 7.5
Al Bronze 4.0
Sn Bronze 4.5
 Feeding of castings 307

WT = W c
i.e.
WF = WT × 100/C × S/100 (12)

3. Calculate the dimensions of the feeder neck

(a) Top feeders
No calculation of feeder neck dimensions are required. If possible,
feeder sleeves should be used with breaker cores.
(b) Side feeders
The required feeder neck dimensions are obtained from the calculation
of the neck modulus (MN) by applying the ratios:
Mc : MN : MF = 1.0 : 1.1 : 1.2
then use either the endless bar equation (Fig. 19.5) or the diagram in
Fig. 19.6.

Grey and ductile irons

1. Calculate the casting modulus
Divide the casting into sections and determine the important volume-to-
cooling surface area ratios according to Fig. 19.5.

Modulus = Volume
Cooling surface
M c = Vc /A c

2. Calculate the feeder modulus

(a) The graphite expansion which occurs during solidification of these
alloys means that grey/ductile iron castings do not shrink for the full
time during which liquid metal is present. The shrinking time (ST) is
only a proportion of the total solidification time. This proportion,
expressed as a percentage, is determined from the central and left-
hand sides of Fig. 19.7, which is used in the following manner:
(b) Using the known carbon content, move parallel to the carbon line to
the appropriate (Si + P) content at point A. Draw a line vertically
until it intersects the casting modulus line at point B. Extend a line
horizontally to the left until it intersects at point D with the line
representing the estimated temperature of the iron in the mould.
Read off shrinking time (ST) as a percentage of solidification time.
Effective feeder modulus is determined by:
MF = Mc × 1.2√ST/100
(c) Provisional determination of the feeder from feeder sleeve tables
published by Foseco companies.
308 Foseco Ferrous Foundryman’s Handbook

a (cm)
1 2 5 10 20 50 100
100 100

20 18
,0
16 4
50 50

1
12 .0
10 0
9. .0
8 0
7. .0
20

6 .0
20

5
4. .0
5
4 5
b (cm)

b (cm)
3. .0
10

3 5
10

2. 0
2. .8
1
1. .4
6

5 5
1 2
1. 0

M (cm)
1. 9
0. .8

5
0 7

1:
0. .6

4
1:
0 .5

2 2
0

: 3
1
a /b

2 1
1: 1
5 1 2 5 10 20 50 100
1,
1: a (cm)
1
1:

r
h
b b

a Bar a

M = a⋅b M= a⋅b M= r⋅h

2(a + b ) 2(a + b ) 2(r + h )

r a
b
h
b
a
Annulus Feeder neck
M = r⋅h M = a⋅b M = a⋅b
2(r + h ) 2(a + b ) 2(a + b )

Figure 19.6 Determination of feeder neck dimensions.

3. Calculate the feed volume requirement

Feeders, whose dimensions (obtained from the previous pages) satisfy
modulus requirements, do not necessarily always satisfy the total feed metal
 Feeding of castings 309

Iron temperature in mould (°C) Iron temperature in mould (°C)

1200 1400 3 2 1200 1300 1400 1500 4

Modulus (cm)
re
mo
0 or
3 .
D 2.5 B
C
2.0
1.5
1.0
0.75
0.50

0.10
0 20 40 60 80 100 +3 +10 –1 –3 –5 –7
Shrinking time in % Sc 1 Expansion (%) Shrinkage (%)
5
of solidification time
1.
1. 5

10

4
00

1.
Si + P (%)

1. .90

05
9
1

2
0.
85
0.
80

1
2.8 3.0 3.2 3.4 3.6 3.8 4.0
Carbon (%)
Estimation of shrinkage and shrinking time
from Analysis, casting modulus and metal
temperature in the mould.
Example: 3.35% C, 2.5% Si + P
Casting modulus: 2.0 cm
Casting temperature: 1300°C
Shrinkage: 1.6%
Shrinking time 55.0%

Figure 19.7 Determination of shrinkage time and shrinkage for grey and ductile
iron castings.

demand of the casting section. This must always be checked and if the
feeder is found to contain insufficient available feed metal the feeder
dimensions must be increased. Generally it is preferable to retain the feeder
diameter dictated by modulus considerations and increase the feeder height.
Sometimes it is more convenient to increase the feeder diameter in place of
or as well as height. Never reduce the feeder diameter below that necessary
to meet modulus requirements.
The following data is necessary:

(a) The proportion of feed metal available from the feeder which meets
modulus requirements (C%). Safe values although not necessarily
the most effective are:
33–50% if a Foseco sleeve is being used;
310 Foseco Ferrous Foundryman’s Handbook

14–16% if it is a live natural feeder (i.e. one through which metal

has to flow before it reaches the casting cavity in the mould);
10% for other natural feeders.
(b) The shrinkage of the alloy to be cast. This is determined using the
central and right-hand portion of Fig. 19.7. From point B draw a
horizontal line until it meets the temperature of the metal in the
mould (this must be estimated from the pouring temperature and is
usually about 50°C less than the pouring temperature). Read expansion
or shrinkage (S) expressed as a percentage on the horizontal axis.
(c) Mould wall movement (M%) depends on the hardness of the mould
and can vary from zero for hard silicate or resin moulds to 2% for
green sand moulds with a mould hardness of 85° (B scale). Add the
mould wall movement from the expansion to give a final value of S.
If this is positive or zero the casting will not need feeding.
(d) The weight of metal in the feeder under consideration (WF).
The total weight of casting (Wc) which can be fed from a feeder of
weight WF is
Wc = C/100 × 100/S × WF (11)
If the total weight of casting section (Wt) which requires feeding is
greater than Wc then increase the dimensions of the feeder until
W T = Wc
i.e.
WF = WT × 100/C × S/100 (12)

4. Calculate the dimensions of the feeder neck

(a) Top feeders
No calculation of feeder neck dimensions are required. If possible
feeder sleeves should be used with breaker cores.
(b) Side feeders
The required feeder-neck dimensions are obtained from the calculation
of the neck modulus (MN) by applying the ratios:
Mc : MN : MF = 1.0 : 1.1 √ST/100 : 1.2 √ST/100
Then using either the endless bar equation (Fig. 19.5g) or the diagram
in Fig. 19.6.

Foseco feeding systems

Introduction

Foseco provides complete feeding systems for foundries, comprising

Sleeves – insulating, exothermic and exothermic-insulating for all metals

 Feeding of castings 311

Breaker cores – to aid removal of the feeder from the casting

Sleeve/filter units
Application technology – to suit the particular moulds and moulding
machine used
Aids to the calculation of feeder requirements.

Range of feeder products

The Foseco range of products comprises:

Product Type Application Remarks

name Alloy

KALMIN S Insulating Al, iron, extend solidification time by 2.0–2.2

steel use for hand moulding
KALMINEX Exothermic- iron, steel extend solidification time by 2.5–2.7
2000 Insulating
FEEDEX HD Highly iron, steel small riser-to-casting contact area
exothermic use where application area is
limited or where modulus is the
paramount factor
KALMINEX Insulating and iron, steel for large diameter feeders
exothermic
KALBORD Insulating iron, steel in form of jointed mats for large
copper-base feeders
KALPAD Insulating iron, steel prefabricated boards or shapes
copper-base for padding
KAPEX Insulating Al, Cu-base prefabricated feeder lids, replace
iron, steel hot-topping compounds
KALPUR Insulating non-ferrous mass production foundries
filter-feeder iron, steel insert sleeves
units for
direct pouring
KALPUR Insulating and non-ferrous jobbing foundries
exothermic- iron, steel ram-up sleeves
insulating filter
feeder units for
direct pouring

KALMIN S, KALMINEX 2000, FEEDEX HD, KALMINEX and KALPUR are

supplied as pre-fabricated sleeves in a wide range of sizes and shapes, with
or without breaker cores.

KALMIN S feeder sleeves

By using a high proportion of light refractory raw materials, a density of

0.45 g/cm3 is achieved, ensuring highly insulating properties. KALMIN S
312 Foseco Ferrous Foundryman’s Handbook

sleeves are particularly suitable for aluminium, copper-base and iron alloy
sand castings since the raw materials used are neutral to both the casting
alloys and to the moulding sand. KALMIN S sleeves are supplied as ‘parallel
conical’, ‘opposite conical’, or cylindrical sleeves for either ram-up application
or insertion into cope or drag (Fig. 19.8, Fig. 19.9, Fig. 19.10, Fig. 19.11). They
can be applied in the following moulding systems:

Moulding method KALMIN S feeder sleeve type

Insert sleeves into the turned over Parallel conical insert sleeves in conjunction
cope mould especially in with Foseco Sleeve Patterns
automatic moulding lines
Ram-up on the pattern plate in a Parallel conical and/or opposite feeder
machine moulding or a hand sleeves, in suitable locating pattern or
moulding operation pin
Insert sleeve in a vertically parted Parallel conical and/or opposite conical
moulding system feeder sleeve in conjunction with
appropriate patterns
Insert sleeves into shell moulds Parallel conical insert and/or opposite
conical feeder sleeve
Application of feeder sleeves in a Parallel conical feeder sleeve as floating
drag mould over a feeder base or sleeve system
bridge core

Parallel conical Opposite conical Cylindrical

insert feeder sleeves feeder sleeves feeder sleeves

Figure 19.8 KALMIN S feeder sleeve types: Parallel conical insert sleeves; Opposite
conical sleeves; Cylindrical sleeves.

KALMIN S feeder sleeves extend the solidification times by a factor of 2.0–

2.2 compared to natural sand feeders of the same size. From these results,
modulus extension factors (MEF) of 1.4–1.5 have been calculated. Though
KALMIN S feeder sleeves can give more than 33% of their feeder volume to
the solidifying casting, it is recommended that a maximum of one-third of
the feed metal volume should be fed into the casting so that the residual
feeder modulus is adequate in relation to the casting modulus at the end of
solidification. For this reason, it is recommended to consider modulus as
well as solidification shrinkage in order to determine the correct feeder.
Foseco provides tables allowing KALMIN S feeders to be selected with the
 Feeding of castings 313

Figure 19.9 Example of sleeve support on a moulding machine pattern plate.

Figure 19.10 A section of an insertable KALMINEX sleeve with breaker core attached
and a typical sleeve pattern.

desired modulus, volume (capacity) and dimensions. The KALMIN S feeder

sleeve material does not contain any component which may contaminate
moulding sand systems and is neutral to the moulding environment.

KALMINEX 2000 feeder sleeves

KALMINEX 2000 is a highly insulating and exothermic feeder sleeve material
in the form of prefabricated sleeves for iron and steel casting in the modulus
range between 1.0 to 3.2 cm. For light and other non-ferrous alloys the
KALMINEX 2000 feeder sleeve material is not recommended. The
manufacturing process specifically developed for this unique feeder sleeve
material not only ensures a low density of 0.59 g/cm and, therefore, a high
grade of insulation but also an additional exothermic reaction peaking at
1600°C. Due to the strength of the KALMINEX 2000 feeder sleeves. they can
often be rammed up directly on the pattern plate on many moulding machines.
314 Foseco Ferrous Foundryman’s Handbook

Figure 19.11 The KALSERT system in operation. KALMINEX sleeves being inserted
into green sand moulds on an automatic moulding line.

KALMINEX 2000 feeder sleeves can be applied in the following moulding

systems:

Application method KALMINEX 2000 feeder shape

Insert sleeves into the turned over Parallel conical insert sleeves in con-
cope mould especially in automatic junction with FOSECO Sleeve Patterns.
moulding lines.
Ram-up on the pattern plate in a Parallel conical insert feeder sleeve,
machine moulding or a hand opposite conical feeder sleeve or
moulding system. cylindrical sleeve.
Insert sleeve in a vertically parted Parallel conical insert and/or opposite
moulding system. conical feeder sleeve.
Insert sleeve into shell moulds. Parallel conical insert and/or opposite
conical feeder sleeve.
Application of feeder sleeves in a Parallel conical feeder sleeve as floating
drag mould over a feeder base or sleeve system.
on a bridge core.

When determining the solidification times with KALMINEX 2000 feeder

sleeves it has been found that they extend the solidification time by a factor
of 2.5 to 2.7 compared to a natural sand feeder of the same size. From these
results modulus extension factors (MEF) have been calculated between 1.58
to 1.64. Foseco provides tables allowing KALMINEX 2000 feeders to be
selected with the desired modulus, volume (capacity) and dimensions.
Under specific practical conditions it has been found that the KALMINEX
 Feeding of castings 315

2000 feeder sleeves can render 64% of their volume to the solidifying casting.
When using feeders with the correct modulus it is necessary to take into
account that the modulus of the residual feeder – if more than 33% of the
feeder volume is fed into the casting – may not be adequate in relation to
the casting modulus towards the end of the solidification. Therefore, it is
essential to calculate shrinkage as well as modulus in order to determine
the correct feeder sleeve.

FEEDEX HD V-Sleeves
FEEDEX HD V Feeder Sleeves are used for iron and steel casting alloys.
FEEDEX HD is a fast-igniting, highly exothermic and pressure resistant
feeder sleeve material. The sleeves possess a small feeder volume, a massive
wall, but only a small riser-to-casting contact area (Fig. 19.12a). They are,
therefore, specially suited for use for ‘spot feeding’ on casting sections which
have a limited feeder sleeve application area. The sleeves are located onto
the pattern plate using special locating pins, the majority are supplied with
shell-moulded breaker cores. Owing to their small aperture, these breaker
cores are not recommended for steel castings, though special larger apertures
can be used.
FEEDEX HD V-sleeves are particularly useful for ductile iron castings
since with its low volume shrinkage of below 3%, a modulus controlled
KALMIN or KALMINEX 2000 feeder will often have more liquid metal
than is necessary. The very high modulus and relatively low volume of
FEEDEX HDV gives improved yield. In many ductile iron applications, the
small breaker core aperture of the feeder means that the feeder is separated
from the casting during the shakeout operation and the cleaning cost is
reduced.
VS spot feeder sleeves without breaker cores but with a sand wedge
between the feeder sleeve and casting are rammed up with the help of a
special spring-loaded location pin, Fig. 19.12b. Moulding pressure squeezes
the VS sleeve down the pin, compacting the sand. The sleeve incorporates
an exothermic locating core to prevent sand entering the feeder cavity during
compaction and to heat the feeder neck.
When used in ductile iron applications, it is important to note that the
high temperature reached in the highly exothermic feeder can cause residual
magnesium in the iron to be oxidised so that there may be a danger of
denodularisation on the casting-feeder interface. To avoid this, residual Mg
should be greater than 0.045%, inoculation practice should be optimised
and thick breaker cores used. Note that when calculating FEEDEX metal
volume, only 50% of the capacity should be assumed since part of the metal
in the feeder will be lamellar due to oxidation of the Mg in the feeder cavity.
The permeability of FEEDEX sleeves is as high as the moulding sand, so
sand system contamination is not a problem.
Figure 19.13a and b shows examples of the use of FEEDEX HD V sleeves
on ductile iron castings.
316 Foseco Ferrous Foundryman’s Handbook

Feedex HD
Kalminex 2000 Feedex HD V 88/40; M = 1, 7
ZP 6/9K/11 V 88/10; M = 1, 7
M = 1, 7

φ 84 φ 56 32 × 58
A = 55 cm2 (100%) A = 25 cm2 (45%) A = 15 cm2 (25%)

(a)

(b)
Figure 19.12 (a) The reduction of the feeder contact area which is possible when
using the FEEDEX HD V-Sleeve with a suitable breaker core. (b) Sectioned spring
loaded steel location pin for ram-up of a V-sleeve without a breaker core.

KALMINEX Feeder Sleeves

KALMINEX exothermic-insulating feeder sleeves are used for all iron

and steel casting alloys. They are supplied with feeder diameters from 80 to
 Feeding of castings 317

(a)

(b)

Figure 19.13 (a) The application of FEEDEX HD V-Sleeves to a ductile iron

turbocharger casting. (b) VS-Spot feeder practice. The point feeding technique makes
possible the production of high value, lightweight castings.

850 mm for the modulus range between 2.4 and 22.0 cm and are suitable for
larger sized castings.
The manufacturing process specifically developed for this exothermic-
insulating product and the selection of specific raw materials give a total
closed pore volume of nearly 50%. The excellent heat insulation resulting
318 Foseco Ferrous Foundryman’s Handbook

from the low density (compared with moulding sand) is enhanced by an

exothermic reaction.
When determining the solidification times with KALMINEX Feeder
Sleeves it has been found that they extend the solidification time by a factor
of 2.0 to 2.4 compared to the natural sand feeders of the same size. From
these results modulus extension factors (MEF) of 1.3 to 1.55 have been
found.
Under practical conditions it has been found that KALMINEX Feeders
when adequately covered with KAPEX lids or a suitable APC (anti-piping
compound) may render up to 64% of their contents into the casting. When
using feeders with the correct modulus it is necessary to take into account
that the modulus of the residual feeder (if more than 33% of the feeder
volume is fed into the casting) may not be adequate in relation to the casting
modulus towards the end of the solidification. Therefore it is essential to
calculate shrinkage as well as modulus when determining the size of the
feeder sleeves.
Foseco provides tables allowing KALMINEX feeders to be selected with
the desired modulus, volume (capacity) and dimensions. Several different
shapes of KALMINEX feeders are available Fig. 19.14a,b,c. Breaker cores
are generally made of chromite sand, though they can be produced in silica
sand.

KALMINEX cylindrical feeder sleeves

KAPEX Lid

φ 0
φ d

Standard Breaker Core

(a)
 Feeding of castings 319

KAPEX Lid

φ Do
φ do

h
T

φN
φ Du

(b)

KAPEX Lid

a b

Standard Breaker Core

(c)

Figure 19.14 (a) KALMINEX cylindrical feeder sleeve with breaker core and KAPEX
lid. (b) KALMINEX TA sleeve. (c) KALMINEX oval sleeve.
320 Foseco Ferrous Foundryman’s Handbook

KALBORD insulating material

Although in theory there is no upper limit of inside diameter for using

prefabricated feeder lining shapes for inside diameters above about 500 mm,
manufacture, transport and storage become increasingly inconvenient.
For this reason Foseco has developed KALBORD flexible insulating material
in the form of jointed mats. They can be easily wrapped around a feeder
pattern or made up into conventional sleeves as required for the production
of insulating feeders for very large steel, iron and copper based alloy castings
(Fig. 19.15).

Thickness
th
id
W

Figure 19.15 KALBORD jointed mats.

KALBORD mats are available with 30 mm and 60 mm thickness in widths

up to 400 mm and lengths 1020 or 1570 mm. Their excellent flexibility
permits the lining of irregular feeder shapes. The mat is most easily separated
or shortened with a saw blade.
Produced from high heat insulating materials, 30 mm mats achieve a 1.2
fold and 60 mm mats a 1.3 fold extension of the modulus. It is recommended
that KALBORD feeders are covered with FERRUX anti-piping powder.

KALPAD prefabricated boards and shapes

KALPAD has been developed by Foseco to provide a light-weight, highly

refractory insulating material to avoid metal padding and to promote
directional solidification. If KALPAD insulating shapes are used the desired
shape of the casting need not be altered. This increases yield and reduces
fettling and machining costs. For this purpose KALPAD is used in copper-
base metal and steel foundries, in particular, however, in malleable iron and
grey iron mass production.
Owing to a special manufacturing process and the use of alumina mineral
fibres KALPAD shapes have a density of 0.45 g/cm3 with more than 60% of
the volume being closed pores which are the reason for the high insulation
and refractoriness. During pouring KALPAD produces only negligible fumes
and behaves neutrally towards moulding materials and casting metals.
When evaluating solidification times on KALPAD padded casting sections
it has been found that they extend the solidification time by a factor of 2.25
to 2.5 compared with conventionally moulded castings. From these results
 Feeding of castings 321

modulus extension factors of 1.5 to 1.58 have been calculated. It is

recommended to use a factor of 1.5 if KALPAD shapes of 20 to 25 mm
thickness are applied. The dimensions of KALPAD boards and shapes are
shown in Fig. 19.16.

0
30
22
500
KALPAD board 1001
Packing unit: 165 pieces, or as required

0
10 Segments 30
22

50 50
550
KALPAD jointed mat 1002
Packing unit: 120 pieces, or as required

300
0
14
45
r = 420

KALPAD pad 1012

Packing unit: 15 pieces

KALPAD 1001 and 1002 are standard types which can be easily sawn to the desired shapes.
KALPAD 1012 is an example of an insulating pad for runner wheels and gear rims.
Shapes made to measure for mass production upon request.

Figure 19.16 KALPAD prefabricated boards and shapes.

322 Foseco Ferrous Foundryman’s Handbook

KAPEX prefabricated feeder lids

KAPEX LD insulating feeder lids (Fig. 19.14a) are an improvement over the
hot-topping powders in foundry use, being dust and fume free and giving
repeatable feeding results. They can be applied to all feeders either exothermic,
insulating or natural. The lids have an insulator density of 0.45 g/cm3 and
are purely insulating. Owing to their neutral behaviour towards moulding
material and casting metal they are used in light metal and copper-base
foundries as well as in high alloy steel foundries. KAPEX LD lids have
replaced over 50% of the hot-topping used in Europe for sleeves of diameter
between 100 and 350 mm and their use is still growing.
For smaller neckdown feeder sleeves, KALMINEX 2000 exothermic KAPEX
lids are also available.

KALPUR filter feeder units

KALPUR filter feeder technology is the shortest way to a perfect casting,
see pp. 266, 289. There are two types of KALPUR filter feeder products; one
for mass production foundries and the other for jobbing foundries. For
mass production, KALPUR insert filter feeder units are inserted and secured
into the mould cavity, created by means of suitable KALPUR insert patterns.
KALPUR KSET filter feeder units with upper filter location are supplied for
horizontally parted automatic moulding machines, Fig. 19.17. For vertically
parted moulding lines, KALPUR units with lower filter position are used
Fig. 19.18.
For jobbing and simple moulding machine application KALPUR ram-up
filter feeders, Fig. 19.19a are used by means of a centering support pattern
in combination with a protective bridge pattern Fig. 19.19b.
KALMINEX 2000 ZTAE KALPUR exothermic units and insulating
KALMIN STP units are available for ram-up applications.
Use of the KALPUR unit eliminates the conventional gating system and
creates the ideal directional solidification condition. The KALPUR filter
feeder units must be selected according to their modulus and flow rate with
a filter type appropriate to the alloy being cast.

Table 19.5 The KALPUR filter feeder product range

Name Material Alloy AppIication* Diameter range (mm)

KALMIN S KSET Insulating NF; iron, steel Horizontal insert 50, 60, 70, 80, 90, 100
KALMIN S KSE Insulating NP; iron, steel Vertical insert 60 and 90
KALMIN 70 STP Insulating NF; iron, steel Jobbing ram-up 50, 70, 90, 125
KALMINEX 2000 Exothermic- Iron and steel Jobbing ram-up 100, 120, 150, 180
ZTAE insulating

*NB for aluminium, iron and steel casting KALPUR types are KALPUR AL; KALPUR FE and KALPUR ST
respectively in which SIVEX; SEDEX/CELTEX and STELEX filters are integrated.
 Feeding of castings 323

(a)

KALPUR Fe
KSET 8/11/L10

20
10
40 25

30
23
20

20

40 40

(b)

Figure 19.17 (a) KALPUR insert filter feeder units with upper filter location for
horizontally parted automatic moulding lines. Without breaker core for direct pouring
through a side feeder. With breaker core for direct pouring through a top feeder.
(b) KALPUR Fe Insert side feeder used for feeding two ductile iron case covers.

Breaker cores
Breaker cores for the reduction of the feeder-to-casting contact area enable
feeders to be broken off or knocked off from many types of castings. In the
case of very tough casting alloys where it is not possible to simply break off
or knock off the feeder, the advantage of using breaker cores lies in the
reduction of fettling and grinding costs for the removal of the feeder.
324 Foseco Ferrous Foundryman’s Handbook

(a)

(b)
Figure 19.18 (a) KALPUR pouring cup filter-feeder units with lower filter position
for vertically parted moulding lines. (b) KALPUR pouring technique used on a
DISAMATIC moulding line for casting ductile iron hydraulic arms.

(a)

Bridging pattern

Centering support

(b)
Figure 19.19 (a) KALPUR ram-up filter feeder series for jobbing and simple moulding
machine applications. (b) Ram-up principle with fixed centering support and a loose
‘bridging’ pattern to give the correct height.
 Feeding of castings 325

Besides the conventional types of breaker cores based on silica sand

(Croning) and chromite sand, special breaker cores with a very small aperture
are also in use in repetition iron foundries. These special breaker cores as
shown in Table 19.5 are made from highly refractory ceramic.

Table 19.5 Application of breaker cores

Breaker core material Casting metal Feeder diameter (mm)

Silica sand Steel 35–120

Silica sand Grey, ductile iron, 35–300
non-ferrous metals
Ceramic Grey, ductile iron, 40–120
non-ferrous metals
Chromite sand Steel 80–500
Chromite sand Grey, ductile iron 200–500

Experience has shown that at least 70% of the breaker core area should be
in contact with the casting, in order to level out the temperatures of the
metal and the breaker core from the superheat upon or before reaching
liquidus.
Some of the standard forms of breaker core available from Foseco are
shown in Fig. 19.20. Foseco feeder sleeves can be ordered with or without
breaker cores attached.

Standard Breaker core

(…/11) Standard Neck-down Breaker core
(…/31)
Standard Breaker core
(…/14)
Ceramic Breaker core
(…/11 K)

Figure 19.20 Standard forms of breaker cores.

The application of feeder sleeves

Ram-up jobbing applications

(1) Manual moulding

Sleeves of the correct dimensions are set on the individual pattern in the
predetermined location and the mould is rammed around the sleeves. The
base of the sleeve should not come into direct contact with the casting but
be set on a sand step at least 10 mm thick or the sleeve should be fitted with
a breaker core.
326 Foseco Ferrous Foundryman’s Handbook

(2) On semi-automatic, cold set and slinger moulding lines

If the pattern plate is accessible to the machine operator, the feeder sleeve is
located by hand on the pattern plate. To avoid damage during machine
moulding, sleeves should be supported by standing them on a pattern dummy
or peg at the correct location and having the correct shape and height.
Figure 19.9 shows one such arrangement.

Insert sleeves

Often on high pressure, squeeze press or air impact moulding lines, pattern
plates are no longer accessible. Foseco recognised these changes in the late
1970s and early 1980s and developed insert sleeve application systems
allowing fully automatic machine users to retain all the advantages of
employing feeder sleeves without slowing down the moulding cycle.
A prefabricated feeder sleeve with strictly controlled dimensional tolerances
is inserted into a cavity formed during the moulding operation by a sleeve
pattern of precise dimensions located on the pattern plate (Figs 19.10,
19.11).
The insert sleeve patterns are fixed by screwing them onto the casting
pattern and they provide the cavity for the insert sleeve. Owing to the
special sealing and wedging system no metal can penetrate behind the
inserted sleeves and these cannot fall out from their seat during closing and
handling of the mould.
The design of the insert patterns also forms highly insulating air chambers
behind the inserted sleeves. This additional insulation increases the moduli
of the insert sleeve feeders as follows:

FEEDEX Insert Sleeves HDP + 5 %

KALMINEX 2000 Insert Sleeves ZP +5%
KALMIN S Insert Sleeves KSP + 4 %

The insert sleeve patterns have a solid aluminium core with mounting thread
and a highly wear resistant resin profile. Insert sleeve patterns are available
corresponding to the various types of insert sleeves with and without breaker
cores. Both insert sleeves and insert patterns are thus part of an integral
system.

Floating feeder sleeves

This is a relatively simple application technique with low feeder sleeve
application cost since feeder sleeves are simply placed on the drag parting
line. The method is applicable for all moulding machines having a horizontal
mould parting line. No problems are encountered regarding strength, spring
back etc. of the feeder sleeve. On high pressure moulding lines, more economic
and non-polluting insulating KALMIN sleeves can be applied.
 Feeding of castings 327

A two-part sleeve pattern is used with an integrated feeder base and

feeder neck (Fig. 19.21). The drag sleeve pattern is secured on to the drag
pattern plate which creates a suitable location and positioning cavity for the
corresponding feeder sleeve. The feeder sleeve is simply positioned on this
location cavity (Fig. 19.22a). The cavity created by means of the cope sleeve
pattern ensures location of the feeder sleeve while closing the mould. After
pouring, the feeder sleeve floats along with the liquid metal, secures and
seals itself tight into the mould wall cavity created by means of the cope
sleeve pattern (Fig. 19.22b).

φ Do

Cope pattern

H
M

φ DU

A Drag pattern
2° 2°
M

DU2
B
DU1
L

Bottom view

Figure 19.21 Sleeve pattern for a floating sleeve. The cavity created by means of
the cope sleeve pattern ensures location of the feeder sleeve while closing the
mould.

The floating sleeve patterns incorporate maximum feeder neck dimensions

applicable to iron castings. For steel, light alloys and non-ferrous alloys,
neck modulus can be modified to usual casting modulus equal to neck
modulus. For full details, refer to Foseco leaflets.
328 Foseco Ferrous Foundryman’s Handbook

feeder sleeve

cope
ingate
neck feeder base
casting
drag

(a) (b)
Figure 19.22 Floating sleeve functional principle. After pouring, the feeder sleeve
floats along with the liquid metal and seals itself tight into the mould wall cavity.

Shell mould application

Sleeves may also be inserted into shell moulds. The principle is the same as
for green sand moulding, special sleeve patterns are available which form
ridges in the sleeve cavity which grip the inserted sleeve but do not damage
the mould (Figs 19.23 and 19.24).

Vertical mould or DISA application

Insert sleeves can be applied equally to moulds with a vertical parting, such
as those made on the Disamatic moulding machine. The sleeve pattern is
divided, but off centre, one part being slightly smaller than the other. The
two parts are mounted on opposite sides of the Disamatic pattern plates
with the sleeve located in the larger cavity and held in place by the exact
vertical fit of the sleeve in the mould. When the mould is closed the second
half holds the sleeve fully in position.

Core application

Feeder sleeves may also be inserted into cores. For example, ductile iron
hubs are often fed by one or more side feeders located externally to the
flange but the most efficient feeding method is by means of a sleeve located
in the central core and connected to the casting at the point where the feed
metal is really needed. The sleeve is placed into the core print location and
fits into the cope mould cavity created by the appropriate sleeve pattern
(similar to a floating sleeve pattern); the result is an improvement in yield,
cleaning costs and casting soundness (Fig. 19.25).
 Feeding of castings 329

D0
x top D01

R1
R8
M10

H1
H

60

x bottom
DU H2

DU1

90°

y bottom
120°

y top

Figure 19.23 Sleeve pattern for shell mould application.

Williams cores

The purpose of a Williams core is to prevent the top of the sand feeder
solidifying prematurely so that atmospheric puncture can take place on top
of the feed metal to promote adequate feeding of the casting. Williams cores
are supplied in a range of sizes up to 66 mm diameter D, Fig. 19.26, in
FEEDEX exothermic material. KALMIN S and KALMINEX 2000 parallel
conical insert sleeves are manufactured with a Williams Wedge incorporated
into the design (Fig. 19.8).
330 Foseco Ferrous Foundryman’s Handbook

Figure 19.24 A shell mould with feeder sleeves (sectioned for clarity) located in
the cavity formed by the shell mould sleeve pattern.

Figure 19.25 A ductile iron hub casting using a sleeve located in the central core.
The sleeve is placed into the core print location and fits into the cope mould cavity
created by the appropriate sleeve pattern.
 Feeding of castings 331

φD φ D′

H′

φd

φd
φD
φ d′
Shape I (without flange) Shape II (with flange)

Figure 19.26 Williams cores.

FERRUX anti-piping compounds for iron and

steel castings
The FERRUX range includes anti-piping compounds of all types with reactions
in contact with the molten metal which vary from very sensitive, highly
exothermic to purely insulating. Described as examples are three grades of
FERRUX manufactured in the UK which cover the requirements of the
complete range of all ferrous alloys cast in all feeder diameter sizes.
All three grades have an exothermic reaction and one of them, FERRUX
707F, by expanding in use, incorporates the most modern technology. The
examples detailed below therefore should only be considered as typical of
the types of FERRUX grades and the technology which is available.
The anti-piping compound, pre-weighed and bagged, should be added
in the bag to the surface of the metal immediately after pouring has been
completed. It is advisable to design the feeder to pour slightly short so that
a space can be left between the surface of the metal and the top of the
mould. FERRUX will then be contained in this space. The recommended
application rate is a layer which has a thickness equivalent to one tenth of
the diameter or 25 mm whichever is the greater. If after application the
powder is not evenly distributed then the upper surface should be raked
flat; normally this will not be found to be necessary.
Note that, for environmental and practical reasons, KAPEX insulating
lids are replacing hot-topping compounds in many applications, see p. 322.

FERRUX 16
This is a carbon-free, sensitive, fast reacting exothermic anti-piping compound
of high heat output. After the exothermic reaction has ceased, a firm crust
332 Foseco Ferrous Foundryman’s Handbook

remains on top of the feeder. It is particularly recommended for use on

feeders where rapid sculling takes place and where carbon contamination is
to be avoided. Feeders where FERRUX 16 is most often employed are in the
diameter range 25–200 mm.

FERRUX 101
This is an exothermic anti-piping compound of medium sensitivity. It is
ideal for general steel foundry use on feeders of 150 mm diameter and
upwards. It may also be used on iron casting feeders where the crust formed
after the exothermic reaction has ceased, forms a good insulation against
heat losses. The crust can be broken for topping up large castings. The
absence of carbonaceous materials in the product ensures that no carbon
contamination of the feed metal will occur.

FERRUX 707F
This is a medium sensitivity, exothermic anti-piping compound which expands
during its reaction to approximately twice its original volume, to produce a
residue of outstanding thermal insulation. In spite of the exothermic reaction,
FERRUX 707F is virtually fume free and, in addition, because of the expansion
and because of the product’s lower density, the original weight of FERRUX
707F which has to be used for effective thermal insulation is usually only
about half that of non-expanding grades. The low carbon content of this
product will not normally affect metal quality in any significant way. FERRUX
707F is most generally employed for steel and iron feeders of 150 mm diameter
and upwards.

Metal-producing top surface covers

THERMEXO is a powdered, exothermic feeding product which reacts on

contact with the feeder metal to produce liquid iron at a temperature of
about 2000°C. The product is designed for emergencies in case of metal
shortage.
Even in the best foundries, occasionally the weight of metal left in the
ladle is overestimated and a casting is poured short. The addition of a metal
producing compound may save the casting by providing the extra feed
metal necessary. In such cases the foundry has nothing to lose by trying a
metal producing compound and it is for emergency reasons that every steel
foundry should have a stock of THERMEXO.

FEEDOL anti-piping compounds for all non-ferrous alloys

The lower casting temperatures and the differing chemical requirements for
non-ferrous alloys necessitates a completely different range of anti-piping
compounds than that used on ferrous castings. FEEDOL is the name given
 Feeding of castings 333

to Foseco’s range of anti-piping compounds for non-ferrous castings. As an

example, two of the principal FEEDOL grades manufactured in the UK are
described in detail below.

FEEDOL 1
This is a mildly exothermic mixture suitable for all grades of copper and
copper alloys. The formulation does not contain aluminium and there is
therefore no risk of contamination where aluminium is an undesirable
impurity. After the exothermic reaction has ceased, FEEDOL 1 leaves a
powdery residue through which further feeder metal can be poured if
necessary. FEEDOL 1 is useful for feeders up to 200 mm diameter. For very
large copper based alloy castings such as for example, marine propellers,
FERRUX 707F is to be recommended.

FEEDOL 9
This is a very sensitive and strongly exothermic compound recommended
for use with aluminium alloys. After the completion of the exothermic
reactions the residue forms a rigid insulating crust. FEEDOL 9 is recommended
for aluminium alloy feeders of all sizes.

The development of Foseco feeding technology

Fifty years ago, foundries made their own feeder sleeves from sacks of
FEEDEX powder supplied by Foseco. In the 1960s Foseco developed a vacuum
production technique for the large-scale manufacture of insulating KALMIN
and exothermic-insulating KALMINEX sleeves in standardised cylindrical
and oval types. At the end of the 1960s the highly exothermic-insulating
KALMINEX S sleeve was developed for application on high production
moulding machines, placed on the pattern plate and rammed up to form a
side feeder. With the increased use in the 1970s of ductile iron for high
volume castings and the use of automatic moulding machines which did
not allow access to the pattern plate during the moulding cycle, Foseco
developed the insert sleeve concept which is still widely used in many high
production iron foundries. This concept also made possible the broader
application of breaker cores. At the end of the 1990s, more than 60% of all
insert sleeves supplied by Foseco have breaker cores.
In the 1990s the V-sleeve concept of a small feeder volume and a small
feeder application area allowing ‘spot’ feeding of castings was developed.
By the mid-1990s V-sleeve technology was introduced into most mass
production iron foundries. At the end of the 1990s many foundries combine
the insert and V-sleeve technologies on fast moulding machines. This allows
castings requiring a large number of feeders to be made within the short
moulding cycle time available.
The latest VS-Spot feeder is connected to the casting by a very small
feeder neck which can be removed with a hammer blow leaving only a
small stub to be ground off.
334 Foseco Ferrous Foundryman’s Handbook

The KALPUR filter feeder unit allows conventional gating systems to be

eliminated and creates ideal directional solidification of the casting.
Figure 19.27 illustrates the range of Foseco feeder and filter-feeder products
presently available.

KALMINEX cylindrical and oval feedersleeves

KALPUR
Ram up filter-feeder

KALPUR
Insert filter-feeder KALMINEX FEEDEX HD
Insert sleeve V-Feeder, VS-Feeder, VS-Spot feeder

Figure 19.27 Some of the available Foseco feeding and filter-feeding units. (This
figure is reproduced in colour plate section.)

Aids to the calculation of feeder requirements

Tables

Tables have been drawn up which will convert natural feeders to sleeved
feeders for steel castings. No knowledge of methoding is required all that is
necessary to know are answers to the following questions.

(a) What are the dimensions of the natural feeder?

(b) What is the weight of the casting section being fed?
(c) What is the alloy composition?
(d) Is the casting with the natural feeder sound?

The tables will do the rest. The conversion, however, is very primitive for if
the natural feeder is too large then the sleeved feeder will also be too large
and conversely if the natural feeder is too small and causes shrinkage so
will the sleeved feeder.
 Feeding of castings 335

Similarly a simple table has been compiled for ductile iron castings. It is
simple to use and requires no expert knowledge of methoding practice.
Although in most cases the recommendation if followed will give a suitable
feeder sleeve it is not necessarily the optimum size for a given casting
section. The table compiled by Foseco in the UK is shown as Table 19.6.

Table 19.6 Feeding guide for ductile iron castings

Weight of casting Sleeve type Sleeve unit No. Feed metal wt.
section (kg) KSP, ZP, HDP insert tapered in sleeve (kg)

270.0 16/15 KC3830 19.5

180.0 14/15 KC3826 13.9
130.0 12/15 KC3324 9.8
82.0 10/13 KC3168 5.7
60.0 9/12 KC3596 4.8
37.0 8/11 KC3164 3.0
26.0 7/10 KC3160 2.2
14.0 6/9 KC3156 1.4
8.9 5/8 KC3152 0.92
6.8 4/95 KC3148 0.77
4.5 4/7 KC3144 0.55
1.7 3.5/5 KC3998 0.23

Nomograms
A series of nomograms which relate the casting modulus, which has to be
calculated, and the weight of the casting section to a suitable size of feeder
sleeve has been developed. Two examples are shown in Fig. 19.28. Such
nomograms have distinct disadvantages, they do not take into account many
of the variables commonly found in steel foundries, they are however a
significant step forward for feeder recommendations can be made without
the need to know the original natural feeder dimensions.

FEEDERCALC
FEEDERCALC is a Foseco copyrighted PC computer program which enables
the foundry engineer to make rapid, accurate calculations of casting weights,
feeder sizes, feeder neck dimensions and feeding distances and to make
cost analyses to quickly determine the most cost-effective feeding system
for any given casting. Versions of FEEDERCALC have been produced for
iron and for steel castings.
The program has been designed by foundrymen for foundrymen and
requires only a basic understanding of Windows. Note that the FEEDERCALC
program does not replace the experienced methods engineer for it relies on
the engineer’s skill to interpret drawings, determine feeder locations and
accommodate the particular production practice of the engineer’s foundry.
The program includes:
336 Foseco Ferrous Foundryman’s Handbook

Feeding guide for steel castings Feeding guide for steel castings
using KALMINEX 2000S using KALMINEX 30 sleeves
10 000
200 4 9000
8000
Weight of casting section (kg)

150 3.5 7000

Weight of casting section (kg)

16/15 6000 Height (in) 12
100 16/15 3 5000 11

)
30

odulus (cm
14/15 24 Diameter (in)
4000
14/15 2.5 18 10
12/15 12 20
50 12/15 3000
40 20 18 9
10/13

Modulus (cm)
30 10/13 2 2000 18 16
8
9/12

Casting m
9/12 16 14
20 8/11 8/11
1500
7
7/10 1.5 1000 14 12
7/10 900 6
10 6/9 6/9 700 10 6
5 5/8 5/8 1 500 12 8
4/7 400 10 6 5
2 Open 300
3.5/5 0.5 8 30 4
200 6 24
Blind 18 3
100 12
50 6 2

Figure 19.28 Examples of nomograms used to determine suitable feeder sleeve

dimensions.

Weight estimation: Combinations of geometric shapes can be easily

manipulated to estimate the weight of complex castings or casting sections.
Feeding distance: Feeding distances can be calculated, with and without
the use of chills. This allows the rapid determination of the number of
feeders required for the casting or casting section.
Feeder size calculation: Accurately calculates the optimum feeder size from
a selection based on the complete range of Foseco feeding products. A
selection can then be made to meet individual production requirements.
Side-neck calculation: Facilitates the calculation of neck dimensions, weight
and fettling areas from the minimum neck modulus supplied in the feeder
size calculation.
Cost analysis: Enables the foundry engineer to select not only the optimum
but also the most cost effective technique for each casting. Cost data
defaults can be used or values specific to the foundry can be entered by
the operator.
Casting information: Information relevant to each casting analysed using
FEEDERCALC can be saved along with the above data, namely Methods
Engineer, Pattern No., Drawing Number etc. Therefore, all records
applicable to the casting can be stored and retrieved using FEEDERCALC.

FEEDERCALC for Windows

(Versions 1.0 or later) requires the following computer hardware to run

effectively.

100% IBM compatible computer (386 processor minimum; 486 or Pentium

recommended)
 Feeding of castings 337

Minimum 4 MB memory
Hard Disk (minimum 2MB available space)
31/2" disk drive
Windows version 3.1 or later
Microsoft Mouse
VGA (or higher) graphics capability
Any printer configured for use with Windows (if hard copy is desired).

The opening Information Screen, (Fig. 19.29) allows the user to save details
of a specific casting method and to select the units and material density to
be used for the calculation. The default language to be used can also be
chosen from English, French, German, Portuguese, Spanish or Italian.

Figure 19.29 FEEDERCALC Information screen.

Weight calculation

Single clicking on this tab displays the Weight Calculation screen shown in
Fig. 19.30. This screen is used to estimate the weight of a casting or casting
section. Weights of complex castings or casting sections are calculated by
breaking the casting down into simple shapes, entering dimensions to calculate
the weight of each shape, and combining all of the weights to give a total
weight for the casting or casting section. (Shape weights may be added to,
or subtracted from, the total.)
338 Foseco Ferrous Foundryman’s Handbook

Figure 19.30 FEEDERCALC weight calculation screen.

Feeding distance

Once the weight of the casting has been estimated, the Feeding Distance
section of the program is used to determine how many feeders are required
to feed the casting (see Fig. 19.31).
In the Iron version, the Feeding Distance program is based on a combination
of the work of Holzmuller and Wlodawer (whose work relates feeding
distance to the amount of graphite present in the alloy), and of the work of
R. Heine and of Wallace, Turnball and Merchant relating feeding distance in
ductile and grey iron castings to both mould hardness and casting section
(modulus). In addition, the calculations in FEEDERCALC have been
augmented with empirical casting data based on practical experience. If
lower soundness standards are acceptable for the particular casting in
question, the user may extend feeding distances shown by the program,
based on specific alloy or mould material being used and the user’s experience.
The program assumes vertical feeding distance to be equal to horizontal
feeding distance. Shrinkage or expansion of the alloy under consideration is
calculated by inputting total Carbon, Si and P. From this data the program
calculates the shrinkage value for the alloy. The mould material pull down
list allows the nearest type of mould material to be selected. This is important
in ductile irons because mould wall movement must be taken into account.
In the Steel version, the Feeding Distance program is based on horizontal
feeding distance path data published by the Steel Founders Society of America
and is recommended for low carbon steel sections down to 25 mm thick (to
A1 radiographic standard) cast in green sand. If lower soundness standards
 Feeding of castings 339

Figure 19.31 FEEDERCALC feeding distance screen.

are acceptable for the casting in question, the user may extend feeding
distances shown by the program, based on the specific alloy or mould material
being used and the user’s experience. The program assumes vertical feeding
distance to be equal to horizontal feeding distance. The user can now design
feeders for the casting by accessing the Feeder Design section of the program.
The mould material is selected from a pull down list and the shrinkage of
the alloy being used calculated from the type of steel used and the pouring
temperature.

Feeder size calculation (Fig. 19.32)

The program determines the optimum size of sleeve required in each Foseco
product line which will meet the requirements for the casting section being
considered. It also calculates the smallest possible sand feeder which will
meet these same conditions, enabling a comparison to be made.
The calculation is based on:

Feeder position
Casting section weight
The section DIS (Diameter of Inscribed Sphere) in the casting section
adjacent to the feeder in question
The number of non-cooling faces
The ingate position and type
340 Foseco Ferrous Foundryman’s Handbook

Figure 19.32 Feeder size calculation.

The safety margin required

Feed demand from lower section
Superheated core or mould section

Once all the required inputs have been entered into the left hand side of the
Feeder Size Calculation screen, the user may select a feeding product from
the ‘Selected Product’ pull down list. Once a product has been selected, the
FEEDERCALC program calculates the optimum feeder size for the casting
under consideration and this information is displayed beneath the product
name. The next size up in the product range is also displayed for comparison
purposes.
The Side Neck Calculation button allows the user to calculate the size of
a neck used with a side feeder. The neck modulus, provided by the feeder
size calculation for the feeder to be used, is automatically transferred to this
screen. The program then calculates the range of values which one dimension
of a rectangular-section neck would have based on this neck modulus. When
the most convenient dimension for the casting being considered is entered,
the second dimension is automatically calculated together with the resulting
contact area.

Cost analysis

Selecting this option allows the cost of a feeding practice for the casting or
castings under consideration to be determined. A comparison of two
 Feeding of castings 341

alternative feeding practices can also be made (for example a practice using
sleeves compared against one using sand risers).
By clicking on the ‘Cost Analysis’ tab, the screen Fig.19.33 appears. Six
items of basic foundry cost data are presented. These are average values
estimated by Foseco (for the country for which the particular program was
designed). The operator may simply accept these (and continue onto the Job
Cost button); or the operator may enter actual costs for the foundry and
save this data set by selecting the ‘New’ or ‘Save’ buttons.

Figure 19.33 Cost analysis screen.

The six basic foundry cost elements are:

Cost of molten metal in the mould (currency unit/weight unit)

This includes costs of raw materials, energy, direct and indirect labour,
melting department overheads and any penalty for losses in providing
liquid metal to the mould cavity, i.e. melting and pouring.
Value accorded to returns (currency unit/weight unit)
The value allocated to the metal in feeders, running systems, and scrap
castings etc. which are returned for re-melting.
Cost of cutting (feeder removal) (currency unit/area)
This includes costs of raw materials, energy, direct and indirect labour,
cleaning department overheads and penalty for losses in cutting feeders
from castings.
Cost of grinding (feeder removal) (currency unit/area)
This includes costs of raw materials, energy, direct and indirect labour,
342 Foseco Ferrous Foundryman’s Handbook

cleaning department overheads and penalty for losses for grinding the
feeder stub etc. in cleaning castings.
Sand density (weight unit/volume unit)
The density of the moulding medium (sand) when compacted in the
mould (typically 1.5 g/cm2 for silica sand).
Cost of moulding sand (currency unit/weight unit)
This includes the costs of raw materials, energy, direct and indirect
labour and overheads associated with preparing the moulding sand
(i.e. sand/binder storage, transport and mixing).

Job costs

The Job Costs button accesses the screen of Fig. 19.34 on which details of the
casting and feeding are entered allowing two practices to be compared. The
final Report screen, Fig.19.35, lists all the costs entered and calculated with
the total costs for each practice displayed. Various ‘what if?’ scenarios may
be investigated by changing data on the Base Costs or Job Costs screen.

Figure 19.34 Job costs screen.

Authorisation

The FEEDERCALC program uses a licensing system to operate the software.

If an attempt is made to copy the files, or run the program without proper
authorisation, the program will not run and an error message pertaining to
 Feeding of castings 343

Figure 19.35 Report screen.

authorisation will be shown on the screen. Authorisation to run the program

is obtained when the Serialization Code for the specific computer on which
the program is to be run is notified to Foseco. An Authorisation Code is then
provided by Foseco enabling the program to run. The program is set to
expire after one year. The user must send Foseco a purchase order to receive
the new Authorisation Code.
Chapter 20
Computer simulation of
casting processes

Introduction
The purpose of computer modelling of any industrial process is to enable
predictions to be made about the effect of adjusting the controls of the
process. The casting process is an ideal candidate for modelling. If the process
of filling a mould with liquid metal and its subsequent solidification could
be accurately and quickly modelled by computer, shrinkage cavities and
other potential defects could be predicted. The effect of changing the gating
system, the position and size of feeders and even the casting design could
be simulated. The casting method could then be optimised before the design
and method are finalised, so avoiding expensive and time consuming foundry
trials.
Software for the numerical simulation of flow and solidification during
casting processes has been available since the mid-1980s. A large number of
commercial software packages are now available and they are improving
all the time. The modelling of heat flow and solidification of castings is now
well advanced. Modelling the filling of castings is more difficult since both
turbulent and quiescent flow in complex shaped cavities may be involved.
The effects of surface oxide films and bubble entrainment are further
complications and it is not easy to check the predictions experimentally in
complex moulds.

Solidification modelling
The aim of solidification modelling is to:

Predict the pattern of solidification, indicating where shrinkage cavities

and associated defects may arise.
Simulate solidification with the casting in various positions, so that the
optimum position may be selected.
Calculate the volumes and weights of all the different materials in the
solid model.
Provide a choice of quality levels, allowing for example the highlighting
or ignoring of micro-porosity.
 Computer simulation of casting processes 345

Perform over a range of metals, including steel, white iron, grey and
ductile iron and non-ferrous metals.

A number of systems are available, they may be divided into two basic
types:

Numerical heat flow simulations

Empirical rule based simulations

Numerical based systems, of which the best known is MAGMAsoft, are

based on thermophysical data: surface tension, specific heat, viscosity, latent
heat, thermal conductivity and heat transfer coefficients of metal, mould
and core materials. Mathematical equations are used to calculate heat flow
and to predict temperatures within the cooling casting. The complexities of
the equations do not allow direct solutions so numerical methods of solving
differential equations, finite difference or finite element techniques, must be
used. Both require considerable computing power. The time and position
where solidification commences is predicted and regions within the casting
identified which are likely to become isolated from feed metal. Turbulence
during mould filling and convection effects during cooling need to be taken
into account.
Empirical rule based systems, such as SOLSTAR, take a model of the
casting and its surrounding moulding media divided into small cubic
elements. Heat flow and solidification are then modelled by applying iterative
rules. In the SOLSTAR method, each element is considered to be at the
centre of a Rubik cube of elements with 26 nearest neighbours. The subsequent
heat exchange calculations are then carried out in 26 directions taking into
account the temperatures and properties of the neighbouring sites. The
temperature of a particular site is thus changed step by step resulting in an
accurate prediction of thermal history. A liquid site will transform into solid
when the site reaches a predetermined value. Metal flow from neighbouring
elements is then mathematically simulated to take up the space vacated by
shrinkage. When there are no liquid metal elements left to fill the shrinkage
cavity a void is created. The system is thus able to predict where shrinkage
defects are likely to occur in the casting. By ‘calibrating’ the rule based
system against experimental results, accurate prediction of shrinkage defects
is achieved.
While the numerical systems are in principle more precise than rule based
systems, at the present time the necessary physical data is not yet available
to make numerical systems completely accurate and some ‘correction factors’
must be introduced.
Rule based systems use standard PC-based computers and are designed
to be used by an average foundry engineer. Simulations of freezing of castings
are achieved in a fraction of the time needed by numerical models. Numerical
systems require more computing power, needing a workstation costing several
times more than a PC and requiring a highly trained computer operator.
Both systems are useful to the practical foundryman, not only cutting out
346 Foseco Ferrous Foundryman’s Handbook

trial and error sampling but increasing metal yield, reducing lead times,
optimising production methods and improving the accuracy of quotations.
The speed and user friendliness of rule based programs makes them
better suited to jobbing foundries where many different castings need to be
processed and computation time is an issue. On the other hand, numerical
programs demand strong computing power and lengthy processing time,
but as better physical data becomes available, they can in principle provide
greater accuracy and a closer representation of what actually happens in the
mould. They allow temperature profiles after solidification to be modelled
so that metallurgical structures can be predicted. They may even take into
account the thermal effects that occur during the filling of the mould. Currently
such programs are perhaps better suited to research and to study highly
engineered, critical components. They are also used for important product
development projects, for example automotive castings to be made in very
large numbers.

Mould filling simulation

Many of the problems associated with the casting process are related to
poor mould filling. Programs have been developed which allow mould
filling to be simulated. The aim is to predict how running and gating design
affects turbulence in the mould which may trap oxidation products and
cause potential defects. MAGMAsoft (and others) have developed such
programs which allow the visualisation and animation of the movement of
the melt surface during filling. Current limitations are system time, it can
take several days to simulate the flow in a mould, and the lack of good data
on surface tension, viscosity etc. Several laboratories around the world are
generating the thermo-physical data needed to improve the simulations.

The SOLSTAR solidification program

SOLSTAR is used by a large number of foundries, service bureaux and
educational establishments. The breakdown of usage (in 1994) is:

Steel foundries 36%

Iron foundries 26%
Non-ferrous foundries 10%
Education & Service 27%

The procedures for carrying out a SOLSTAR analysis are:

(1) Using the casting drawing, determine model scale and element size.
(2) Make the solid model of the casting.
(3) Make the solid model of the proposed production method (feeders,
 Computer simulation of casting processes 347

chills, insulators etc.). Use the program’s own feeder-size calculator

if required.
(4) Carry out thermal analysis to establish the order of solidification.
(5) Carry out solidification simulation to a set quality standard, for the
selected alloy incorporating shrinkage percentage, ingate effects etc.
This results in the model being changed to the predicted final shape
(internal and external) of the casting showing size, shape and location
of shrinkage cavities in casting and feeders.
(6) Examine the predicted shrinkage (the equivalent of non-destructive
testing) by viewing and plotting of 3D ‘X-rays’ and sections of the
model in 2D slices or 3D sections and relating predicted defects to
solidification contours and required quality standards.
(7) If the predicted defects do not meet the required quality standard,
develop an improved production method and repeat the procedures.

These trial-and-error sampling procedures can be carried out very rapidly,

allowing the operator to indulge in any number of ‘what-if’ experiments.

Solid modelling
The first stage in any solidification simulation is to create a three-dimensional
model of the component with its associated method. This will often take the
greatest proportion of time, as much as 70%. The SOLSTAR program has its
own solid modeller/mesh generator capable of modelling the most complex
casting shapes. Depending on the computer hardware specification, models
can contain up to 256 million elements but most models use between 2 and
64 million elements. Figure 20.1 shows a solid model of a 350 kg steel valve
casting containing 40 million elements produced in less than 3 hours.
It is possible to transfer 3D models from any other CAD system using
Stereolithography STL files created by them. These models can then be
manipulated within the solidification software so that the method can be
added.

Thermal analysis
The thermal analysis calculates the simulated heat flow between the elements
of the solid model which gives a ‘thermal picture’ of the conditions prevailing
at a specific point in time. SOLSTAR’s thermal analysis simulates ‘heat
flow’ in 26 directions, with each cuboid element of metal, mould, chill etc.
being the equivalent of the centre block of a ‘Rubik’ cube (27 cubes).
SOLSTAR uses the thermal analysis to store details of the solidification
order of each element of the casting and feeding system. Figure 20.2 shows
a ‘thermal’ illustration of a section through the steel valve casting (in black
and white on p. 349 and reproduced in colour in plate section). This is
produced in colours showing ‘solidification contours’ of the metal from the
348 Foseco Ferrous Foundryman’s Handbook

Figure 20.1 Solid model of a valve casting utilising 40 million elements. (This
figure is reproduced in colour in plate section.)
 Computer simulation of casting processes 349

Figure 20.2 Solidification contours for the lower section of the valve casting. (This
figure is reproduced in colour in plate section.)

end of pouring to the end of solidification. The thermal analysis for the steel
valve involves approximately 200 billion ‘heat exchanges’ between adjacent
elements of the model and was calculated in 22 minutes using a 266 MHz
computer.

Solidification simulation
After the thermal analysis is completed, each metal element of the model is
allocated an order of solidification. SOLSTAR then carries out a solidification
simulation of the metal elements, by solidifying them in the predetermined
order. During this simulation several things are happening:

(1) The solidified elements are assumed to have increased in density,

accompanied by a loss of volume.
(2) This loss (liquid shrinkage) is calculated by multiplying the number
of solidifying elements by the input shrinkage factor for the alloy.
(3) The software calculates (according to the alloy and the required quality
standard) whether this shrinkage will manifest itself in the form of a
cavity and, if so, how big the cavity will be.
(4) The resultant cavity is placed in the remaining liquid of the section of
which it is a part. Where it resides in this remaining liquid depends
on the type of alloy.
350 Foseco Ferrous Foundryman’s Handbook

(5) The program continually checks the linkages between the remaining
liquid metal.
(6) Metal elements are continually ‘flowing’ through the liquid paths to
replace volume loss during solidification, so accurate tracing of these
paths is critical to the program.

At the end of the solidification simulation the model represents the casting
(and feeders) at the ‘shake-out’ stage of production. Figure 20.3 is an ‘X-ray’
plot of the final model showing all the shrinkage cavities predicted to be
outside of the requested quality standard. A Class 2 solidification simulation
for this valve took 26 minutes using a 266MHz computer.

Figure 20.3 An ‘X-ray’ plot showing predicted shrinkage cavities. (This figure is
reproduced in colour in plate section.)

During the solidification simulation, the effect of varying

moulding position,
ingate position,
mould materials, chills, insulating and exothermic materials,

can be modelled, allowing the optimum method of making the casting to be

predicted.
 Computer simulation of casting processes 351

Feeder size and weight calculations

The program can calculate the volume, weight and surface area of each
material in any selected part of the model, so that casting yield is readily
obtained. The program can also calculate feeder sizes for steel and ductile
iron, giving a selection of feeder options (height/diameter ratios, sand or
insulating/exothermic materials). This data allows more accurate estimation
of production costs to be made.

Cost benefits of solidification simulation

Figure 20.4 shows a simplified trialling process for making a cast component,
the cycle may need repeating several times before product of acceptable
quality is made. Solidification simulation software such as SOLSTAR can be
used to electronically sample the method and cut down on the number of
actual trials made (Table 20.1).

component “method” make mould

design

cast metal

reject
casting x-ray for holes

cycle repeated
up to good
10 times casting

Figure 20.4 The trialling process for making a cast component of acceptable
quality.

Table 20.1 Pre- and post SOLSTAR methoding accuracy in a steel foundry

Trial methods Before SOLSTAR Using SOLSTAR

Right first time 50% 99%

Two attempts 85% 100%
Three attempts 98%

(From M. Jolly, Foundryman Jan. 1994 p. 11)

352 Foseco Ferrous Foundryman’s Handbook

Each trial carries a cost in materials and manpower but, more important
still is the effect on lead time for producing a component. For new jobs in a
foundry, where solidification simulation is practised as standard procedure,
components can be modelled and methoded at a rate of 10 each week per
method engineer, and most of these methods will be right first time (Table
20.2).

Table 20.2 Probability of right first time reported by

SOLSTAR users

Steel 99%
Iron 90–99%
Non-ferrous
(short freezing range alloys) 99%
Non-ferrous
(long freezing range alloys) 85–95%

Conclusions
While this chapter has concentrated on the benefits of the SOLSTAR program
to foundries, there are many other rule based programs which can give
similar good results. It is important to remember that programs such as
these, as well as the more comprehensive numerical packages, must be used
by skilled foundrymen who are able to interpret the results to achieve practical
solutions in the special circumstances of their own foundries. It must also
be remembered that, however good the simulation is, no foundry has complete
control over its manufacturing processes so some variation in the end result
is inevitable and safety factors must be built in to the design of the casting
being made.
Index

Accuracy, dimensional, 21 Blackheart malleable cast iron, 25

Acid demand value of sand, 147 Breakdown agents for silicates, 207
measurement of, 149 Brinell hardness of metals, 8
Acid resisting cast irons, 101 Breaker cores, 323
Adhesives for core assembly, 213 BRIX cupola flux, 48
AFS Grain Fineness Number, calculation Bulk densities of common materials, 10
of, 16 Buoyancy forces on cores, 18, 20
Air set sands, see Self-hardening Burn-on of sand, 226
Alcohols as carriers for coatings, 229 Burnt clay, 161
Alkaline phenolic resin binders, 183
Aluminium in grey iron, 39 Calcium bentonite, 154
Amine catalysts for cold box, 192 Calibration of mixers, 171
control, 202 Carbon equivalent value, 32, 37
Annealing ductile iron, 82 Carbon equivalent liquidus value, 33
Angle of repose of silica sand, 148 CARSET ester hardener, 210
Anti-piping compounds, see FERRUX CARSIL sodium silicate binders, 207
AOD refining of steel, 125 CARSIN coal dust replacement, 154
Arc furnace melting of steel, 121 Cast irons:
Areas and volumes of circles, spheres, alloy additions in the ladle, 138
cylinders etc., 6 carbon equivalent value, 32
ASTM specifications for graphite flake carbon equivalent liquidus value, 33
form, 34 casting processes, 28
Atomic weights of metals, 7 compacted graphite, 84
Austempered ductile iron (ADI), 83 corrosion resistant, 101
Austenitic cast iron, 96 ductile, see Ductile iron
Austenitic manganese steel, 114 grey iron, see Grey iron
Automatic pouring boxes, KALTEK heat resisting, 95
lining, 136 inoculation of, see Inoculation
Average grain size, calculation, 15 machinability, 34
malleable, see Malleable iron
melting, 40
BAKFIL, 132 physical properties, 27
Baumé measurement of coatings, 232 density, 27
Bench life of self-hardening sand, 167 electrical resistivity, 28
BENTOKOL sand additive, 154 specific heat capacity, 28
Bentonite, 153 thermal conductivity, 29
Binders, chemical thermal expansion, 28
self-hardening, 167 scrap analysis, 50
resin, 180 specifications of, 30
silicate, 209 wear resistant, 102
triggered hardening, Casting defects:
gas hardened resin, 192 cast iron, 250
heat hardened resin, 186 due to ladles, 137
silicate, CO2 hardened, 204 hydrogen pinholes, 138
Bismuth in coatings, 242 steel, 125, 277
354 Index

Casting processes for cast irons, 28 Computer simulation of casting, 344

Castyral process, 223 Consistency, dimensional, 22
Cathodic protection anodes, 101 Contraction allowances, 11
Catalysts for furane binders, 180 Control of metal composition, 60
CELTEX ceramic filters, 259 Conversion tables:
Ceramic cellular filters, 251 SI, metric and non-metric, 2
Ceramic foam filters, 251 stress values, SI, metric, imperial, 5
Cereal binders for sand, 155 Copper in grey iron, 39
Channel furnace, 54 Cooling of iron in ladles, 131
Chaplets, areas for core support, 18 Cooling green sand, 160
Chelford sand, 148 Core making processes:
Chemically bonded sand, 167 CO2 silicate process, 204
effect of temperature, 172 cold box, 192
ester silicate process, 210 comparison of processes, 200
resin bonded systems, 180 ECOLOTEC, 194
testing, 169 ester cured alkaline phenolic, 197
sodium silicate systems, 204 hot box, 188
CHILCOTE coating for chills, 242 oil sand, 190
Chills, coatings for, 242 shell, 186
Chromite sand, 152 SO2 process, 196
Chromium in grey iron, 39 warm box, 189
Chvorinov’s rule for solidification time, Core print support, 19
302 Cores, buoyancy of, 18, 20
Clamping moulds, 19, 173 Coreless induction furnace:
Clay additives for green sand, 153 iron melting, 55
CO2 gassed resin cores, see ECOLOTEC alloy recovery, 58
CO2 silicate process, 204 charge materials, 58
gas consumption, 205 fume extraction, 60
Coal dust, 154 refractories, 59
Coatings: CORFIX adhesives, 188
for foundry tools, 243 Corrosion resistance of cast irons, 101
for moulds and cores, 226 Corrosion resistant cast irons, 101
brushing, 231 CORSEAL sealants, 213
dipping, 230 Croning resin shell process, 29, 186
drying, 233 Cupola melting of cast irons, 40
for metallurgical change, 242 cold blast, 41
overpouring/flow coating, 232 charge calculation, 50
spraying, 231 emissions, 52
water-based, 234 metallic charge, 48
spirit-based, 237 output, 51
TRIBONOL process, 240 cokeless, 43, 53
Coke: hot blast, 52
consumption in cupolas, 42 CUPOLLOY, 49
quality, 42
Cokeless cupola, 43
Cold box coremaking process, 192 Dead clay, 161
environmental problems, 194, 202 Density:
Cold-setting sand, see Self-setting casting alloys, 9
Colours, standard colours of patterns, common materials, 10
17 metals, 7
Compacted graphite iron, 84 cast irons, 27
SinterCast process, 87 Deoxidation of steel, 125
 Index 355

Desulphurisation of iron, 75 core application, 328

DEXIL silicate breakdown agent, 209 Disa application, 328
Dextrin, 155 floating feeder sleeves, 326
Dicalcium silicate, 209 insert sleeves, 326
Diecasting of cast iron, 29 shell mould, 328
Dimensional tolerances of castings, 22 breaker cores, 323
Dip coating of cores, 230 feeder dimensions, calculation of, 301
Direct pouring of iron castings: feeder neck calculation, 307
KALPUR system, 266 feeders:
Disamatic moulding machine, 165 aided, 297
DISA insert sleeves, 328 natural, 296
DMEA (dimethyl ethyl amine), 192 feeding distance, 299
DMIA (dimethyl isopropyl amine), 192 feeding
Drying ovens for coatings, 233 ductile iron castings, 299
Ductile cast iron, 25, 70 non-ferrous castings, 300
casting, 84 steel castings, 305
heat treatment, 82 Foseco feeding systems, 310
inhibiting elements, 78 FEEDEX HD V-Sleeves, 315
inoculation, 79 FEEDOL anti-piping compound,
melting, 74 332
specifications, 79 FERRUX anti-piping compound,
treatment methods, 75 331
Duplex steel, 117 KALBORD insulating material, 320
Dust control in foundries, 18 KALMIN S feeder sleeves, 311
KALMINEX feeder sleeves, 316
KALMINEX 2000 feeder sleeves,
ECOLOTEC process, 194 313
Electrical resistivity of cast irons, 28 KALPAD boards and shapes, 320
Electric melting of cast iron, 53 KAPEX lids, 318, 322
ESHAMINE cold box resin, 192 THERMEXO metal-producing
ESHAMINE Plus process, 194 cover, 332
ESHANOL furane binders, 180 modulus, 301
Ester-cured alkaline phenolic system, 197 extension factor, 302
Ester silicate process, 210 nomograms for calculation of feeders,
CARSET, 210 335
VELOSET, 212 tables for calculation of feeders, 334
Ethanol, 229 Williams cores, 329
Eutectic arrest of cast iron, 61 FEEDEX HD V-sleeves, 315
Evaporative casting process, see Lost foam FEEDOL anti-piping compound, 332
Exogenous inclusions, 277 FENOTEC ADT1 anti-fusion additive, 178
Exothermic feeder sleeves, see Foseco FENOTEC alkaline phenolic resin,
feeding systems gas cured, 197
Expandable polystyrene, 217 self setting, 183
Ferrite, 26
Ferroalloys in the cupola, 49
FEEDERCALC, 335 Ferro-manganese, 49
Feeding of castings, 296 Ferro-silicon, 49
aids to calculation, 334 FERRUX anti-piping compound, 331
FEEDERCALC, 335 Filling time for moulds, 248
nomograms, 335 Filter/feeder/pouring cup
tables, 334 KALPUR, 266, 322
application of sleeves, 325 KALPUR ST, 289
356 Index

Filtration and filters: calculation of average grain size, 15

iron castings, 245, 250 Graphite flake size and shape, 34
effect on physical properties, 255 undercooled, 36, 63
vertically parted moulds, 265 Graphitisation potential, 62
steel castings, 272, 277 Green sand, 152
flow rates and capacity, 286 additives:
types of filter, 251 BENTOKOL, 154
SEDEX ceramic foam filters, 256 bentonite, 153
STELEX ZR ceramic foam filters, cereals, 155
279 clay, 153
Flake graphite, size and shape, 35 coal dust, 154
Flaskless moulding, 165 dextrin, 155
Floating feeder sleeves, 326 MIXAD, 157
Flow-coating, 232 starch, 155
Flow cup, 233 cooling, 160
Fluxes in cupola melting, 48 control, 161
Forces: moulding machines, 164
buoyancy on cores, 18 properties, 160
opening forces on moulds, 19 sand mill, 157
FOSCAST dam board, 133 system, 157
Foundry layout for self hardening testing, 162
moulds, 173 Grey cast iron, 23
Foundry tools, coating, 243 applications of, 37
Foseco-Morval foam patterns, 217 effect of elements in, 37
FRACTON refractory dressing, 243 inoculation of, 62
Full-Mould process, 29 graphite flake size and shape, 34
coatings for, 240 production of, 37
Furane resins, 180 specifications, 30
Furnaces:
iron melting
cupolas, 40 Haltern sand
electric, 53 HARDCOTE bond supplement, 241
steel melting Heat resistant cast irons, 95
arc, 121 Heat treatment of ductile iron, 82
induction, 123 austempering, 83
FUROTEC furane binders, 180 Heat-triggered binder systems, 184
Hexamine, 186
HOLCOTE coatings, 237
Gas curing of core binders, 192 Hoppers for sand, 158
Gas porosity in steel castings, 127 Hot box process, 188
Gas triggered binder systems, 186 Hot tearing in steel castings, 129
Gating: Hydrogen in steel, 128
iron castings, 247 Hydrogen pinholes in iron, 138
steel castings, 272 Hydrometer, see Baumé measurement
Gating with filters:
iron castings, 256
steel castings, 277 Imperial-metric conversions, 2
Grain Fineness Number (AFS), 16 IMPREX Station, 66
Grain shape of sands, 147 Inclusions:
Grain size: iron castings, 250
calculation of AFS grain fineness steel castings, 127, 277
number, 16 Indigenous inclusions, 277
 Index 357

Induction furnace: steel foundries, 138

iron melting: KALTEK ladle lining system, 141
channel, 54 Latent heat of fusion of metals, 7
coreless, 55 Launders, coatings for, 243
steel melting, 123 Lead in grey iron, 39
Ingot moulds, 96 Levi equation for cupola output, 50
Inhibiting elements, 78 Limestone for cupolas, 48
Inmold treatment, 74, 78 Liquidus temperature for steels, 144
Inoculation of cast iron, 62 Loss on ignition, reclaimed sand, 176
ductile iron, 79 Lost foam process, 29, 216
IMPREX station, 66 coatings, 220, 239
INOPAK mould inoculant, 69 Low alloy steels, 112
INOTAB mould inoculant, 68 Low carbon bead, 220
ladle inoculation, 64 Lustrous carbon, 182, 219, 193
late stream inoculation, 66 LUTRON moulding sand, 163
mould inoculation, 69
wedge chill test, 65
INOCULIN products, 64 Machinability:
INOPAK mould inoculant, 69 cast iron, 34
INOTAB mould inoculant, 68 effect of filtration, 255
Iron-carbon phase diagram, 24 steel, 281
Iron oxide additions to prevent nitrogen MAGMAsoft simulation software, 345
defects, 193 Magnesium silicate inclusions in ductile
Insert sleeves, 326 iron, 254
Inspection of steel castings, 119 Magnesium treatment of cast iron, 71
Isocyanate resins, see Phenolic isocyanate converter, 72, 77
ISOMOL spirit-based coatings, 238 INMOLD, 74
Isopropanol, 229 IMPREX treatment station, 73, 78
NODULANT, 77
sandwich ladle, 71, 77
KALBORD insulating material, 320 tundish cover, 71, 75
KALMIN pouring cups, 188 Malleable cast iron, 23
KALMIN S feeder sleeves, 311 blackheart, 25, 92
KALMINEX feeder sleeves, 316 whiteheart, 24, 90
KALMINEX 2000 feeder sleeves, 313 Manganese in grey iron, 37
KALPAD boards and shapes, 320 Manganese steel, 114
KALPAK ramming material, 143 coating for, 238
KALPUR direct pouring system: Manganese sulphide inclusions, 254
iron, 266 Matchplate moulding, 165
steel, KALPUR ST, 286, 322 MDI (methylene diphenyl diisocyanate),
KALSEAL, 132 182, 192
KALTEK ladle lining system: MEKP (methyl ethyl ketone peroxide),
automatic pouring boxes, 136 196
iron foundries, 132 Melting cast irons, 40
steel foundries, 142 cupola melting, 40
KAPEX lids, 318, 322 electric, 53
channel furnace, 54
coreless induction, 55
LADELLOY, 138 Melting point of metals, 7
Ladles: Melting steel:
iron foundries, 130 AOD refining, 125
KALTEK ladle lining system, 132 arc furnace, 121
358 Index

Melting steel (continued) self-setting, 182

deoxidation, 125 Phosphoric acid, 180
induction furnace, 123 Phosphorus in grey iron, 39
Metal penetration, 226 Physical properties of cast irons, 27
Metals, tables of physical properties, 7 Physical properties of metals, 7
Methanol, 229 Pig iron, 49
Methyl formate hardener, 197 Pinholing in steel, 129
Mills for green sand, 158 PMMA (polymethylmethacrylate), 220
MIXAD additive, 158 POLISET phenolic-isocyanate binder, 182
Mixers for chemically bonded sand, 171 POLITEC cold box resin, 192
Modelling: Polystyrene, see Expandable polystyrene
mould filling, 344, 346 Pouring bushes:
solidification of castings, 344 iron, 246
Molten metal handling: steel, 274
iron, 130 Pouring cups, KALMIN, 188
steel, 138 Pouring temperature for steels, 144
Mould design for self hardening sand, Powder coatings, 228
173 Prints:
Mould filling simulation, 346 core support, 18
Moulding machines, 164 design for filters
disamatic, 166 iron, 259, 260
Mould inoculation of cast iron, 69 steel, 285
Moulds, opening forces on, 19 Pre-coated sand, 186
MSI 90 Stream Inoculator, 66 PROCAL, 132
Properties of metals, 7

Newton, 2
Ni-hard, 103 RADEX insulator for iron ladles, 132
Ni-resist, 96, 101 Radioactivity in zircon sand, 151
Nitrogen: Ready for use coatings, 228
grey iron, 39 Reclamation of chemically bonded sand,
steel, 128 173
No-bake sand, see Self-hardening mechanical attrition, 175
NODULANT, 71, 77 thermal, 177
Nodular iron, see Ductile iron VELOSET system, 179, 213
Novolak resin, 186 wet, 179
Nozzles in bottom pour ladles, 141, 143 Refractories:
induction melting
Oil sand, 190 iron, 59
Olivine sand, 152 steel, 124
Opening forces on moulds, 19 Refractory coating for foundry tools, 243
Overpouring coatings, 232 Release agents for self-hardening sand,
192
Resin binders:
Parting agents, 163 alkaline phenolic, 183
Pascal, 2 furane, 180
Patternmakers’ contraction allowances, gas triggered, 186
11 heat-triggered, 184
Patterns, standard colours, 17 hot box, 188
Pearlite, 26 phenolic-isocyanate, 182
Phenolic-isocyanate binders: self-hardening, 180
cold box, 192 Resistivity of cast irons, 28
 Index 359

Rheology of coatings, 230 epoxy, 197

RHEOTEC coatings, 234 furane, 196
Running systems: Sodium silicate:
iron castings, 245 breakdown agents, 207
steel castings, 272 CO2 process, 204
ester silicate process, 204
see also CARSET, CARSIL,
Sand: SOLOSIL, VELOSET
acid demand, 101, 149 properties of, 204
Chelford, 148 Solidification modelling, 344
chromite, 152 SOLOSIL silicate binder, 209
cooling, 160 SOLSTAR solidification simulation, 345
German, 149 Specific coke consumption in cupolas, 42
grain shape, 147 Specific heat capacity of cast irons, 28
green, see Green sand Specific surface area of silica sand, 148
olivine, 152 Specifications:
reclamation, see Reclamation grey cast iron, 30
safe handling of silica sand, 148 ductile iron, 79
segregation of, 149 malleable iron:
silica, 146 blackheart, 93
grain shape, 147 whiteheart, 91
properties, 147 steel, 109
safe handling, 148 Spheroidal graphite cast iron, 26, 70
thermal characteristics, 150 Sprue, 247
zircon, 151 base, 247, 275
radioactivity of, 151 Stainless steel, 116
Sand reclamation, see Reclamation Starch additive for sand, 155
Sandwich ladle treatment, 71 Steel castings, 108
Scrap iron compositions, 50 carbon steel, 112
Sealants, 213 duplex steel, 117
SEDEX ceramic foam filters for iron, 256 low alloy steel, 112
Segregation of sand, 149 high alloy steel, 114
Self-hardening process, 167 hot tearing, 129
foundry layout, 175 manganese steel, 114
mould design, 173 physical properties, 117
SEMCO coatings, 237 pinholing, 129
SEMCO PERM coatings for Lost Foam, pouring temperature, 144
220, 239 specifications, 109
SEPAROL parting agent for green sand, stainless steel, 116
163 Stress conversion table, 5
Shell sand process, 186 Stress relief annealing, 82
Shrinkage of casting alloys, 13, 306 STRIPCOTE release agent, 192
SI units, 1 Strip time of self-hardening sand, 170
Sieve sizes, comparison, 14 STYROMOL coatings for Lost Foam, 220,
Silicon in grey iron, 37 239
Simulation of casting processes, 344 Sulphides in steel, 126
SinterCast process, 87 Sulphonic acid, 180
Sintering point of silica sand, 151 Sulphur in grey iron, 39
Slag removal, 58
SLAX, 58
Sleeve-filter units, see KALPUR TAK sealant, 215
SO2 process: TEA (triethylamine), 192
360 Index

TELLURIT chill-promoting coating, Urethane binders, see Phenolic-isocyanate

242
Temperature losses in ladles:
iron, 131 Vacuum moulding, 29
steel, 142 VELOSET system, 179, 212
TENO spirit-based coatings, 238 Vermicular graphite, 85
Terminal velocity of sand grains, 18 Volume shrinkage of casting alloys, 13
Thermal conductivity of cast irons, 29
Thermal expansion of cast irons, 28
Thermal reclamation of sand, 177 Warm box core making, 189
THERMEXO metal-producing cover, Wear resistant cast irons, 102
332 Wedge chill test, 65, 92
Thixotropic coatings, 231 Weighting moulds, 19
Tin in grey iron, 39 Wet reclamation, 179
Titanium in steel, 126, 286 Williams cores, 329
Tolerances, dimensional of castings, 22 Whiteheart malleable cast iron, 24, 90
TRIBONOL process, 240 Work time of chemically bonded sand,
Tundish cover treatment, 71 167
Types of cast iron, 23

Zircon sand, 151

Undercooled graphite, 63 radioactivity of, 151