Download the full version of the ebook at ebookfinal.

com

Protection of Electricity Distribution Networks

3rd Edition Juan M. Gers

https://ebookfinal.com/download/protection-of-electricity-
distribution-networks-3rd-edition-juan-m-gers/

OR CLICK BUTTON

DOWNLOAD EBOOK

Download more ebook instantly today at https://ebookfinal.com

Instant digital products (PDF, ePub, MOBI) available
Download now and explore formats that suit you...

Protection of Electricity Distribution Networks 2nd

Edition IEE Power and Energy Series Ted Holmes Juan Gers

https://ebookfinal.com/download/protection-of-electricity-
distribution-networks-2nd-edition-iee-power-and-energy-series-ted-
holmes-juan-gers/
ebookfinal.com

Protection of Electrical Networks Christophe Prévé

https://ebookfinal.com/download/protection-of-electrical-networks-
christophe-preve/

ebookfinal.com

Distribution System Requirements for Fire Protection 4th

Edition Edition American Water Works Association

https://ebookfinal.com/download/distribution-system-requirements-for-
fire-protection-4th-edition-edition-american-water-works-association/

ebookfinal.com

Advanced course of mathematical analysis 3 Juan M. Delgado

Sanchez

https://ebookfinal.com/download/advanced-course-of-mathematical-
analysis-3-juan-m-delgado-sanchez/

ebookfinal.com
Teach Yourself Electricity and Electronics 3rd Edition
Stan Gibilisco

https://ebookfinal.com/download/teach-yourself-electricity-and-
electronics-3rd-edition-stan-gibilisco/

ebookfinal.com

The Complete Lab Manual for Electricity 3rd Edition

Stephen Herman

https://ebookfinal.com/download/the-complete-lab-manual-for-
electricity-3rd-edition-stephen-herman/

ebookfinal.com

Pattern Recognition for Reliability Assessment of Water

Distribution Networks UNESCO IHE PhD Thesis 1st Edition N.
Trifunovic (Author)
https://ebookfinal.com/download/pattern-recognition-for-reliability-
assessment-of-water-distribution-networks-unesco-ihe-phd-thesis-1st-
edition-n-trifunovic-author/
ebookfinal.com

Transmission and Distribution 1st Edition U. A. Bakshi M.

V. Bakshi

https://ebookfinal.com/download/transmission-and-distribution-1st-
edition-u-a-bakshi-m-v-bakshi/

ebookfinal.com

ASFP Fire Protection for Structural Steel in Buildings 3rd

Edition Revised June 2004 Association Of Specialist Fire
Protection Contractors And Manufacturers
https://ebookfinal.com/download/asfp-fire-protection-for-structural-
steel-in-buildings-3rd-edition-revised-june-2004-association-of-
specialist-fire-protection-contractors-and-manufacturers/
ebookfinal.com
 Power and Energy Series 65

Protection of Electricity
Protection of

3rd Edition
Distribution Networks
Protection of Electricity
Distribution Networks
3rd Edition
Electricity
High quality electrical service is everyday more stringent in utilities and
industrial facilities around the world. One of the main players to achieve
Juan M. Gers obtained his BSc in Electrical
Engineering at University of Valle, Colombia, in
Distribution
Networks
this is the protection system, which has to be reliable, fast and with a good 1977. In 1981 he finished an MSc at the University
cost/benefit ratio. This book refers to most aspects of electrical protections, of Salford in England and his PhD in 1998 at the
University of Strathclyde in Scotland. He has been
with emphasis on Distribution Systems. Protection of generation and
working as consulting engineer and university
transmission systems are also treated in the text. References to modern lecturer for over 30 years. Dr. Gers is a Chartered
topics such as the Distributed Generation, Smart Grid and Standard IEC Engineer of the IET and Member of the IEEE.
61850 have been introduced. Written by two well experienced engineers

3rd Edition
Ted Holmes graduated from Queen Mary,
who combine a comprehensive theoretical background with examples and University of London in 1950 and joined the
exercises, this book will allow the reader to easily follow the ideas explored. CEGB, where he worked for over 35 years. A
The book will be valuable to pre and postgraduate students, design, Fellow of the IEE and a Chartered Engineer, he
maintenance and consulting engineers as well as instructors looking for is co-author of the book Electricity Distribution
proper references. Network Design. He passed away in July of 2006
after fighting a long illness.

Juan M. Gers
and Edward J. Holmes
Gers and Holmes

The Institution of Engineering and Technology

www.theiet.org
978-1-84919-223-1

Protection of Electrical Distribution.indd 1 11/05/2011 10:25:46

 IET POWER AND ENERGY SERIES 65

Protection of
Electricity
Distribution
Networks

00_Gers_Prelims_i-xvi 7 May 2011; 17:21:50

Other volumes in this series:

Volume 1 Power circuit breaker theory and design C.H. Flurscheim (Editor)
Volume 4 Industrial microwave heating A.C. Metaxas and R.J. Meredith
Volume 7 Insulators for high voltages J.S.T. Looms
Volume 8 Variable frequency AC motor drive systems D. Finney
Volume 10 SF6 switchgear H.M. Ryan and G.R. Jones
Volume 11 Conduction and induction heating E.J. Davies
Volume 13 Statistical techniques for high voltage engineering W. Hauschild and W. Mosch
Volume 14 Uninterruptible power supplies J. Platts and J.D. St Aubyn (Editors)
Volume 15 Digital protection for power systems A.T. Johns and S.K. Salman
Volume 16 Electricity economics and planning T.W. Berrie
Volume 18 Vacuum switchgear A. Greenwood
Volume 19 Electrical safety: a guide to causes and prevention of hazards
J. Maxwell Adams
Volume 21 Electricity distribution network design, 2nd edition E. Lakervi and E.J. Holmes
Volume 22 Artificial intelligence techniques in power systems K. Warwick, A.O. Ekwue
and R. Aggarwal (Editors)
Volume 24 Power system commissioning and maintenance practice K. Harker
Volume 25 Engineers’ handbook of industrial microwave heating R.J. Meredith
Volume 26 Small electric motors H. Moczala et al.
Volume 27 AC-DC power system analysis J. Arrillaga and B.C. Smith
Volume 29 High voltage direct current transmission, 2nd edition J. Arrillaga
Volume 30 Flexible AC transmission systems (FACTS) Y-H. Song (Editor)
Volume 31 Embedded generation N. Jenkins et al.
Volume 32 High voltage engineering and testing, 2nd edition H.M. Ryan (Editor)
Volume 33 Overvoltage protection of low-voltage systems, revised edition P. Hasse
Volume 34 The lightning flash V. Cooray
Volume 36 Voltage quality in electrical power systems J. Schlabbach et al.
Volume 37 Electrical steels for rotating machines P. Beckley
Volume 38 The electric car: development and future of battery, hybrid and fuel-cell
cars M. Westbrook
Volume 39 Power systems electromagnetic transients simulation J. Arrillaga
and N. Watson
Volume 40 Advances in high voltage engineering M. Haddad and D. Warne
Volume 41 Electrical operation of electrostatic precipitators K. Parker
Volume 43 Thermal power plant simulation and control D. Flynn
Volume 44 Economic evaluation of projects in the electricity supply industry H. Khatib
Volume 45 Propulsion systems for hybrid vehicles J. Miller
Volume 46 Distribution switchgear S. Stewart
Volume 47 Protection of electricity distribution networks, 2nd edition J. Gers
and E. Holmes
Volume 48 Wood pole overhead lines B. Wareing
Volume 49 Electric fuses, 3rd edition A. Wright and G. Newbery
Volume 50 Wind power integration: connection and system operational aspects
B. Fox et al.
Volume 51 Short circuit currents J. Schlabbach
Volume 52 Nuclear power J. Wood
Volume 53 Condition assessment of high voltage insulation in power system
equipment R.E. James and Q. Su
Volume 55 Local energy: distributed generation of heat and power J. Wood
Volume 56 Condition monitoring of rotating electrical machines P. Tavner, L. Ran,
J. Penman and H. Sedding
Volume 57 The control techniques drives and controls handbook, 2nd edition B. Drury
Volume 58 Lightning protection V. Cooray (Editor)

00_Gers_Prelims_i-xvi 7 May 2011; 17:22:18

 Protection of
Electricity
Distribution
Networks
3rd Edition

Juan M. Gers
and Edward J. Holmes
The Institution of Engineering and Technology

00_Gers_Prelims_i-xvi 7 May 2011; 17:22:18

Published by The Institution of Engineering and Technology, London, United Kingdom
The Institution of Engineering and Technology is registered as a Charity in England &
Wales (no. 211014) and Scotland (no. SC038698).
† 1998, 2004 The Institution of Electrical Engineers
† 2011 The Institution of Engineering and Technology

First published 1998 (0 85296 923 6)

Second edition hardback 2004 (0 86341 357 9)
Second edition paperback 2005 (0 86341 537 7)
Third edition 2011

This publication is copyright under the Berne Convention and the Universal Copyright
Convention. All rights reserved. Apart from any fair dealing for the purposes of research
or private study, or criticism or review, as permitted under the Copyright, Designs and
Patents Act 1988, this publication may be reproduced, stored or transmitted, in any
form or by any means, only with the prior permission in writing of the publishers, or in
the case of reprographic reproduction in accordance with the terms of licences issued
by the Copyright Licensing Agency. Enquiries concerning reproduction outside those
terms should be sent to the publisher at the undermentioned address:
The Institution of Engineering and Technology
Michael Faraday House
Six Hills Way, Stevenage
Herts, SG1 2AY, United Kingdom
www.theiet.org
While the author and publisher believe that the information and guidance given in
this work are correct, all parties must rely upon their own skill and judgement when
making use of them. Neither the author nor publisher assumes any liability to
anyone for any loss or damage caused by any error or omission in the work, whether
such an error or omission is the result of negligence or any other cause. Any and all
such liability is disclaimed.
The moral rights of the author to be identified as author of this work have been
asserted by him in accordance with the Copyright, Designs and Patents Act 1988.

British Library Cataloguing in Publication Data

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

ISBN 978-1-84919-223-1 (paperback)

ISBN 978-1-84919-224-8 (PDF)

Typeset in India by MPS Ltd, a Macmillan Company

Printed in the UK by CPI Antony Rowe, Chippenham

00_Gers_Prelims_i-xvi 7 May 2011; 17:22:18

 Contents

Preface and acknowledgements xiii

Preface to second edition xiv
Preface to third edition xv

1 Introduction 1
1.1 General 1
1.2 Basic principles of electrical systems 2
1.3 Protection requirements 2
1.4 Protection zones 5
1.5 Primary and back-up protection 5
1.5.1 Primary protection 6
1.5.2 Back-up protection 6
1.6 Directional protection 7

2 Calculation of short-circuit currents 11

2.1 Modelling for short-circuit current calculations 11
2.1.1 Effect of the system impedance 11
2.1.2 Effect of rotating machinery 13
2.1.3 Types of fault duty 15
2.1.4 Calculation of fault duty values 16
2.2 Methods for calculating short-circuit currents 18
2.2.1 Importance and construction of sequence networks 21
2.2.2 Calculation of asymmetrical faults using symmetrical
components 25
2.2.2.1 Line-to-earth fault 25
2.2.2.2 Line-to-line fault 25
2.2.2.3 Line-to-line-to-earthfault 25
2.2.3 Equivalent impedances for a power system 25
2.3 Supplying the current and voltage signals to protection systems 27
2.4 Calculation of faults by computer 29

3 Classification and function of relays 31

3.1 Classification 31
3.1.1 Construction 31
3.1.2 Incoming signal 31
3.1.3 Function 31
3.1.4 International identification of electrical devices 32

00_Gers_Prelims_i-xvi 7 May 2011; 17:22:18

vi Protection of electricity distribution networks

3.2 Electromechanical relays 33

3.2.1 Attraction relays 33
3.2.2 Relays with moveable coils 34
3.2.3 Induction relays 35
3.3 Evolution of protection relays 38
3.4 Numerical protection 38
3.4.1 General 38
3.4.2 Characteristics of numerical relays 39
3.4.3 Typical architectures of numerical relays 40
3.4.4 Standard functions of numerical relays 41
3.5 Supplies to the relay circuits 42

4 Current and voltage transformers 43

4.1 Voltage transformers 43
4.1.1 Equivalent circuit 43
4.1.2 Errors 44
4.1.3 Burden 45
4.1.4 Selection of VTs 45
4.1.5 Capacitor voltage transformers 45
4.2 Current transformers 49
4.2.1 Equivalent circuit 49
4.2.2 Errors 50
4.2.3 AC saturation 50
4.2.4 Burden 52
4.2.5 Selection of CTs 54
4.2.6 Accuracy classes established by the ANSI standards 55
4.2.7 DC saturation 57
4.2.8 Precautions when working with CTs 58

5 Overcurrent protection 61
5.1 General 61
5.2 Types of overcurrent relay 61
5.2.1 Definite-current relays 61
5.2.2 Definite-time/current or definite-time relays 64
5.2.3 Inverse-time relays 64
5.3 Setting overcurrent relays 64
5.3.1 Setting instantaneous units 65
5.3.2 Coverage of instantaneous units protecting lines between
substations 66
5.3.3 Setting the parameters of time-delay overcurrent relays 67
5.3.3.1 The pick-up setting 68
5.3.3.2 Time dial setting 69
5.3.3.3 Time discrimination margin 70
5.3.3.4 Use of mathematical expressions for the relay
characteristics 70

00_Gers_Prelims_i-xvi 7 May 2011; 17:22:18

 Contents vii

5.4 Constraints of relay coordination 71

5.4.1 Minimum short-circuit levels 71
5.4.2 Thermal limits 71
5.4.3 Pick-up values 73
5.5 Coordination across Dy transformers 84
5.5.1 Three-phase fault 85
5.5.2 Phase-to-phase fault 85
5.5.3 Phase-to-earth fault 86
5.6 Coordination with fuses 94
5.7 Coordination of negative-sequence units 95
5.8 Overcurrent relays with voltage control 95
5.9 Setting overcurrent relays using software techniques 96
5.10 Use of digital logic in numerical relaying 97
5.10.1 General 97
5.10.2 Principles of digital logic 98
5.10.3 Logic schemes 99
5.11 Adaptive protection with group settings change 100
5.12 Exercises 102

6 Fuses, reclosers and sectionalisers 107

6.1 Equipment 107
6.1.1 Reclosers 107
6.1.2 Sectionalisers 111
6.1.3 Fuses 112
6.1.3.1 Selection of nominal current 114
6.1.3.2 Selection of nominal voltage 114
6.1.3.3 Selection of short-circuit capacity 114
6.1.3.4 Fuse notation 114
6.2 Criteria for coordination of time/current devices in distribution
systems 114
6.2.1 Fuse–fuse coordination 115
6.2.2 Recloser–fuse coordination 116
6.2.2.1 Fuse at the source side 116
6.2.2.2 Fuses at the load side 117
6.2.3 Recloser–recloser coordination 118
6.2.3.1 Hydraulic reclosers 118
6.2.3.2 Electronically controlled reclosers 119
6.2.4 Recloser–relay coordination 119
6.2.5 Recloser–sectionaliser coordination 120
6.2.6 Recloser–sectionaliser–fuse coordination 120

7 Directional overcurrent relays 125

7.1 Construction 125
7.2 Principle of operation 125

00_Gers_Prelims_i-xvi 7 May 2011; 17:22:18

viii Protection of electricity distribution networks

7.3 Relay connections 126

7.3.1 30
connection (0
AMT) 126
7.3.2 60
connection (0
AMT) 127
7.3.3 90
connection (30
AMT) 127
7.3.4 90
connection (45
AMT) 128
7.4 Directional earth-fault relays 129
7.5 Coordination of instantaneous units 135
7.6 Setting of time-delay directional overcurrent units 139
7.6.1 Pick-up setting 139
7.6.2 Time dial setting 139
7.6.2.1 Time dial setting by direct method 139
7.6.2.2 Time dial setting considering contact travel 140
7.7 Exercises 144

8 Differential protection 147

8.1 General 147
8.2 Classification of differential protection 149
8.3 Transformer differential protection 149
8.3.1 Basic considerations 150
8.3.2 Selection and connection of CTs 151
8.3.3 Percentage of winding protected by the differential relay
during an earth fault 157
8.3.4 Determination of the slope 159
8.3.5 Distribution of fault current in power transformers 159
8.4 Differential protection for generators and rotating machines 161
8.5 Line differential protection 164
8.6 Busbar differential protection 165
8.6.1 Differential system with multiple restraint 165
8.6.2 High impedance differential system 166
8.6.2.1 Connecting CTs in parallel 166
8.6.2.2 Using common buses for each phase 167
8.7 Exercises 167

9 Distance protection 171

9.1 General 171
9.2 Types of distance relays 171
9.2.1 Impedance relay 173
9.2.2 Directional relay 175
9.2.3 Reactance relay 177
9.2.4 Mho relay 179
9.2.5 Completely polarised mho relay 180
9.2.6 Relays with lens characteristics 182
9.2.7 Relays with polygonal characteristics 182
9.2.8 Relays with combined characteristics 182
9.3 Setting the reach and operating time of distance relays 183

00_Gers_Prelims_i-xvi 7 May 2011; 17:22:19

 Contents ix

9.4 The effect of infeeds on distance relays 186

9.5 The effect of arc resistance on distance protection 190
9.6 Residual compensation 193
9.7 Impedances seen by distance relays 193
9.7.1 Phase units 194
9.7.2 Earth-fault units 194
9.8 Power system oscillations 194
9.9 The effective cover of distance relays 198
9.10 Maximum load check 199
9.10.1 Mho relays 199
9.10.2 Relays with a polygonal characteristic 201
9.11 Drawing relay settings 202
9.12 Intertripping schemes 210
9.12.1 Underreach with direct tripping 212
9.12.2 Permissive underreach intertripping 212
9.12.3 Permissive overreach intertripping 213
9.13 Distance relays on series-compensated lines 214
9.14 Technical considerations of distance protection in
tee circuits 214
9.14.1 Tee connection with infeeds at two terminals 214
9.14.2 Tee connection with infeeds at all three terminals 218
9.15 Use of distance relays for the detection of the loss of excitation
in generators 219
9.16 Exercises 223

10 Protection of industrial systems 225

10.1 Protection devices 225
10.1.1 Overcurrent relays 225
10.1.2 Direct-acting devices in power and moulded-case
circuit breakers 225
10.1.3 Combined thermal relay contactor and fuse 226
10.2 Criteria for setting overcurrent protection devices associated
with motors 226
10.2.1 Thermal relays 226
10.2.1.1 Coil rating 227
10.2.2 Low-voltage breakers 227
10.2.2.1 Nominal current 229
10.2.2.2 Setting of the time band 230
10.2.2.3 Range of short-time unit 230
10.2.2.4 Setting of instantaneous element 230

11 Industrial plant load shedding 239

11.1 Power system operation after loss of generation 239
11.2 Design of an automatic load shedding system 241

00_Gers_Prelims_i-xvi 7 May 2011; 17:22:19

x Protection of electricity distribution networks

11.2.1 Simple machine model 241

11.2.2 Parameters for implementing a load shedding system 241
11.2.2.1 Maximum load to be disconnected 241
11.2.2.2 Starting frequency of the load shedding system 242
11.2.2.3 Minimum permissible frequency 242
11.3 Criteria for setting frequency relays 242
11.3.1 Operating times 243
11.3.2 Determination of the frequency variation 243
11.4 Example of calculating and setting frequency relays
in an industrial plant 243
11.4.1 Calculation of overload 243
11.4.2 Load to be shed 243
11.4.3 Frequency levels 244
11.4.4 Load shedding stages 245
11.4.5 Determination of the frequency relay settings 245
11.4.5.1 First-stage setting 245
11.4.5.2 Second-stage setting 245
11.4.5.3 Third-stage setting 246
11.4.6 Verification of operation 247
11.4.6.1 Modelling load as a constant power 248
11.4.6.2 Modelling load with damping as a result of
the voltage drop (P ¼ kV 2) 249
11.4.6.3 Analysis of voltage with total loss of infeed
supply and operation of the load shedding
system 250

12 Protection schemes and substation design diagrams 251

12.1 Protection schemes 251
12.1.1 Generator protection 251
12.1.1.1 Small generators 251
12.1.1.2 Large generators 252
12.1.2 Motor protection 253
12.1.2.1 Protection of low power motors 254
12.1.2.2 Protection of motors up to 1 000 HP 255
12.1.2.3 Protection of motors greater than 1 000 HP 254
12.1.2.4 Additional protection for synchronous motors
over 1 000 HP 255
12.1.2.5 Protection for the field winding 255
12.1.3 Transformer protection 255
12.1.3.1 MV/LV transformers 256
12.1.3.2 HV/MV/LV transformers and autotransformers 256
12.1.4 Line protection 258
12.1.4.1 Medium voltage lines 259
12.1.4.2 High voltage lines 261

00_Gers_Prelims_i-xvi 7 May 2011; 17:22:19

 Contents xi

12.2 Substation design diagrams 261

12.2.1 Single-line diagrams 263
12.2.2 Substation layout diagrams 263
12.2.3 Diagrams of AC connections 264
12.2.4 Diagrams of DC connections 264
12.2.5 Wiring diagrams 265
12.2.5.1 Multicore cables 267
12.2.6 Logic diagrams 267
12.2.7 Cabling lists 267

13 Communication networks for power systems automation 269

13.1 IEC 61850 overview 270
13.2 Standard documents and features of IEC 61850 271
13.3 System configuration language 276
13.4 Challenges facing the testing of IEC 61850 devices 277
13.5 Configuration and verification of GOOSE messages 279
13.5.1 Configuration of the system 281
13.5.2 System verification test 282
13.6 Substation IT network 282
13.7 Process bus 282

14 Installation, testing and maintenance of protection systems 285

14.1 Installation of protection equipment 285
14.2 Testing protection schemes 287
14.2.1 Factory tests 287
14.2.2 Pre-commissioning tests 287
14.2.2.1 Insulation resistance measurement 287
14.2.2.2 Secondary injection tests 288
14.2.2.3 Current transformer tests 291
14.2.2.4 Primary injection test 291
14.2.3 Periodic maintenance 291
14.2.3.1 Methodology 292
14.2.3.2 Maintenance criteria 293
14.3 Commissioning numerical protection 293
14.3.1 Setting the parameters 294
14.3.2 Performance tests 294

Appendix: Solutions to exercises 299

Bibliography 339
Index 343

00_Gers_Prelims_i-xvi 7 May 2011; 17:22:19

00_Gers_Prelims_i-xvi 7 May 2011; 17:22:19
 Preface and acknowledgements

The quality of electricity supplies is an important factor in the socio-economic

development of any area. Approximately 75% of all customer hours lost are owing
to faults on the distribution networks, and customers rightly expect a high level of
security for their supply. Although this can be achieved by good distribution net-
work design using proven equipment, it is also essential to provide suitable pro-
tection schemes and relay settings to ensure that faults are quickly disconnected to
minimise outage times and improve the continuity of supplies to customers.
It was with this in mind that this book has been produced as a reference guide
for professional engineers and students. It is hoped that the many detailed examples
and exercises throughout the book, which the authors have taken from actual case
studies in the field, will provide worthwhile material for planning and design
engineers and maintenance staff, particularly those engaged in the coordination and
setting of protection on distribution systems.
The book is based on original material by Mr Gers and translated from the
Spanish by Mr Holmes. Subsequently, the authors have expanded the text con-
siderably and added much up-to-date material. It is our view that in the process, the
continuous dialogue between the co-authors from differing backgrounds and
experiences has led to a deeper exploration of the subject matter.
Thanks are due to the University of Valle and the Colombian Institute of
Sciences (COLCIENCIAS) for financial help, the British Council for continuing
support during many years of institutional exchange, C. Delgado for his valuable
comments on the initial manuscripts and our colleagues at GERS Ltda. for their
ideas and assistance, F. Pacheco in producing the diagrams, Professor K.L. Lo at
Strathclyde University for his guidance, and Stephen and Philip Holmes for sharing
so generously their computer expertise. In addition, the authors have been most
grateful for the help received from many sources and the permission, readily given,
by various organisations to include copyright material that is acknowledged in the
text. Finally, we wish to acknowledge the considerable support and understanding
that we have received from our wives Pilar and Maggie throughout the four years of
work on this book.

J.M. Gers E.J. Holmes

Cali, Colombia Stourbridge, England
[email protected] [email protected]

00_Gers_Prelims_i-xvi 7 May 2011; 17:22:19

 Preface to second edition

In the six years since this book was first published, there have been considerable
advances in relay protection design. The development of powerful numerical
algorithms and further improvements in digital technology have greatly extended
the scope of protection systems. Most of the latest types of relays are now multi-
functional devices with control, metering, reporting and alarm functions in addition
to their protection capabilities that normally include several types within the same
device. They also have very good communication facilities that allow them to work
in virtually any automated scheme. Modern relays therefore now offer better pro-
tection coverage and can be programmed to automatically adjust for changes in
power system topologies and different operating conditions due to the multiple
setting groups feature incorporated in most of them.
Chapters 3 and 5 have been considerably extended to include more detail on
numerical relays. Chapter 12, dealing with protection schemes, has been updated to
take account of the new technology available, while the testing procedures covered
in the last chapter now include ample reference to numerical protection.
We have also taken the opportunity of updating sections of the original text
and have added a new chapter on the processing of alarms since the fast and effi-
cient processing of the many alarms that flow from the power system into control
centres has an important bearing on the speed with which system faults are dealt
with. Our thanks are due to our colleague Professor K. L. Lo of the University of
Strathclyde, Glasgow, Scotland, for his help with this material.
Finally, once again we acknowledge the support of our wives Pilar and Maggie
during the work on this edition.
J.M. Gers E.J. Holmes
Weston, Florida Stourbridge
USA England
[email protected] [email protected]

00_Gers_Prelims_i-xvi 7 May 2011; 17:22:19

 Preface to third edition

The fundamentals of protective relays have remained the same for many years, but
software, hardware and communications have experienced outstanding progress.
That has been introduced in the protection systems as a whole, which has impacted
positively the performance of relays and in particular in regards to reliability and
speed of operation.
Distribution systems are receiving great attention in all utilities worldwide
since the responsibility of the quality of service depends greatly on their appro-
priate performance. A very important role in the proper operation of distribution
systems is played by the protective relays that, as always, have to perform reliably,
rapidly and in selective form. Interesting enough is that the price of the relays has
dropped remarkably as it has happened with all computer-based devices. Modern
relays, of course, are structured very much following computer architecture.
One important requirement of relays applied to distribution systems is their
ability to adapt to the prevailing condition of the topology that can vary in multiple
ways due to the ever growing number of devices installed in the system to imple-
ment distribution automation. The use of adaptive criteria is a good solution for
this. It requires an individual study of all the operating scenarios to achieve the
correct setting for each one of the groups that modern relays have. This has made
the setting of relays applied to distribution system more interesting but more
complex and time demanding too.
The third edition does not introduce considerable changes in regards to fun-
damentals but in new technologies implemented. Fewer changes have been intro-
duced in the different chapters to ease the understanding of the material or to
update some topics of the first two editions. The chapter on processing alarms was
removed, and instead a chapter on communication networks for power systems
automation was introduced, which has ample reference to the IEC 61850 Standard.
I regret not having had for the preparation of this new edition our late friend
Ted Holmes who fought for many years a long illness. I still miss the long and
fruitful discussions we had during the preparation of the previous editions. How-
ever, the spirit of the book and its technical level have been rigorously maintained.
Special thanks are given to James Ariza for his cooperation with the chapter on
communication networks, Cesar Gallego for his comments on several topics,
Francisco Pacheco for his care and patience in producing the drawings, Francia
Solis for his help with the overall organization of the material and in general to the
engineers of the GERS group for their invaluable contribution throughout the many
years working in the field of protections.

00_Gers_Prelims_i-xvi 7 May 2011; 17:22:19

xvi Protection of electricity distribution networks

I also extend my gratitude to the IET for their trust and in particular to Lisa
Reading for her enthusiasm with the new edition and to S. Ramya for her valuable
contribution with the revision of the text.
I am always grateful to my wife Pilar for her continuous understanding and to
Maggie, Ted’s wife, who always gave great support to this effort.

J.M. Gers
Weston, Florida
USA
[email protected]

00_Gers_Prelims_i-xvi 7 May 2011; 17:22:20

 Chapter 1
Introduction

1.1 General
With the increasing dependence on electricity supplies, in both developing and
developed countries, the need to achieve an acceptable level of reliability, quality
and safety at an economic price becomes even more important to customers.
A further requirement is the safety of the electricity supply. A priority of any
supply system is that it has been well designed and properly maintained in order to
limit the number of faults that might occur.
Associated with the distribution networks themselves are a number of ancillary
systems to assist in meeting the requirements for safety, reliability and quality of
supply. The most important of these are the protection systems that are installed to
clear faults and limit any damage to distribution equipment. Among the principal
causes of faults are lightning discharges, the deterioration of insulation, vandalism,
and tree branches and animals contacting the electricity circuits. The majority of
faults are of a transient nature and can often be cleared with no loss of supply, or just
the shortest of interruptions, whereas permanent faults can result in longer outages.
To avoid damage, suitable and reliable protection should be installed on all circuits
and electrical equipment. Protective relays initiate the isolation of faulted sections of
the network in order to maintain supplies elsewhere on the system. This then leads to
an improved electricity service with better continuity and quality of supply.
A properly coordinated protection system is vital to ensure that an electricity
distribution network can operate within preset requirements for safety for indivi-
dual items of equipment, staff and public, and the network overall. Automatic
operation is necessary to isolate faults on the networks as quickly as possible in
order to minimise damage. The economic costs and the benefits of a protection
system must be considered in order to arrive at a suitable balance between the
requirements of the scheme and the available financial resources. In addition,
minimising the costs of non-distributed energy is receiving increasing attention.
When providing protective devices on any supply network, the following basic
principles must apply. On the occurrence of a fault or abnormal condition, the
protection system must be capable of detecting it immediately in order to isolate
the affected section, thus permitting the rest of the power system to remain in
service and limiting the possibility of damage to other equipment. Disconnection of
equipment must be restricted to the minimum amount necessary to isolate the fault

CH001 10 May 2011; 14:33:12

2 Protection of electricity distribution networks

from the system. The protection must be sensitive enough to operate when a fault
occurs under minimum fault conditions, yet be stable enough not to operate when
its associated equipment is carrying the maximum rated current, which may be a
short-time value. It must also be fast enough to operate in order to clear the fault
from the system quickly to minimise damage to system components and be reliable
in operation. Back-up protection to cover the possible failure of the main protection
is provided on most circuits in order to improve the reliability of the protection
system. While electromechanical relays can still be found in some utilities, the
tendency is to replace these by microprocessor and numerical relays, particularly in
the more complex protection arrangements.

1.2 Basic principles of electrical systems

The primary aim of any electricity supply system is to meet all customers’
demands for energy. Power generation is carried out wherever it achieves the
most economic selling cost overall. The transmission system is used to transfer
large amounts of energy to major load centres, while distribution systems carry
the energy to the furthest customer, using the most appropriate voltage level.
Where the transport of very large amounts of power over large distances is
involved, an extra high voltage (EHV) system, sometimes termed major or pri-
mary transmission, is required. Such systems operate in the >300-kV range,
typical values being 400, 500 and 765 kV.
High-voltage (HV) networks transport large amounts of power within a particular
region and are operated as either interconnected systems or discrete groups. Below the
transmission system, there can be two or three distribution voltage levels to cater for
the variety of customers and their demands. In general, the medium-voltage (MV)
networks and low-voltage (LV) networks are operated as radial systems.
Figure 1.1 illustrates the interrelation of the various networks. The HV
networks are supplied from EHV/HV substations that themselves are supplied by
inter-regional EHV lines. HV/MV transforming substations situated around each
HV network supply individual MV networks. The HV and MV networks provide
supplies direct to large customers, but the vast majority of customers are connected
at low voltage and supplied via MV/LV distribution substations and their asso-
ciated networks, as shown in Figure 1.2.

1.3 Protection requirements

The protection arrangements for any power system must take into account the
following basic principles:
1. Reliability: The ability of the protection to operate correctly. It has two
elements – dependability, which is the certainty of a correct operation on the
occurrence of a fault, and security, which is the ability to avoid incorrect
operation during faults.

CH001 10 May 2011; 14:33:12

 Introduction 3

EHV

EHV/HV
Major HV
substation

HV/MV

HV network

HV/MV

MV

MV
overhead MV cable
network network
To adjacent
MV
network

Urban distribution

Rural MV/LV
distribution
LV board

LV network

Generator Circuit breaker Feeder Customer

Transformer Disconnector Fuse

Figure 1.1 EHV/HV/MV/LV network arrangements [reproduced from LAKERVI, E.,

and HOLMES, E. J.: ‘‘Electricity distribution network design’’,
Peter Peregrinus, 1995, 2nd Edn.; revised 2003

CH001 10 May 2011; 14:33:12

4 Protection of electricity distribution networks

Generation

EHV
Generation

EHV/HV

HV

HV/MV

MV
Consumption

MV/LV

LV

Consumption

Figure 1.2 Block schematic of transmission and distribution systems [reproduced

from LAKERVI, E., and HOLMES, E. J.: ‘‘Electricity distribution
network design’’, Peter Peregrinus, 1995, 2nd Edn.; revised 2003]

2. Speed: Minimum operating time to clear a fault in order to avoid damage to

equipment.
3. Selectivity: Maintaining continuity of supply by disconnecting the minimum
section of the network necessary to isolate the fault.
4. Cost: Maximum protection at the lowest cost possible.
Since it is practically impossible to satisfy all the above-mentioned points
simultaneously, inevitably a compromise is required to obtain the optimum
protection system.

CH001 10 May 2011; 14:33:12

 Introduction 5

1.4 Protection zones

The general philosophy for the use of relays is to divide the system into separate zones,
which can be individually protected and disconnected on the occurrence of a fault, in
order to permit the rest of the system to continue in service wherever possible.
In general, a power system can be divided into protection zones – generators,
transformers, groups of generator transformers, motors, busbars and lines. Figure 1.3
shows a system with different protection zones. It should be noted that the zones
overlap at some points indicating that, if a fault occurs in these overlap areas, more
than one set of protection relays should operate. The overlap is obtained by connecting
the protection relays to the appropriate current transformers as illustrated in Figure 1.4.

M M
~ ~

Figure 1.3 Protection zones

1.5 Primary and back-up protection

All the elements of the power system must be correctly protected so that the relays
only operate on the occurrence of fault. Some relays, designated as unit type

CH001 10 May 2011; 14:33:12

6 Protection of electricity distribution networks

Busbar protection

Line protection

Figure 1.4 Overlap of protection zones

protection, operate only for faults within their protection zone. Other relays are able
to detect faults both within a particular zone and outside of it, usually in adjacent
zones, and can be used to back up the primary protection as a second line of
defence. It is essential that any fault is isolated, even if the associated main pro-
tection does not operate. Therefore, wherever possible, every element in the power
system should be protected by both primary and back-up relays.

1.5.1 Primary protection

Primary protection should operate every time an element detects a fault on the
power system. The protection element covers one or more components of the power
system, such as electrical machines, lines and busbars. It is possible for a power
system component to have various primary protection devices. However, this does
not imply that they all have to operate for the same fault, and it should be noted that
the primary protection for one item of system equipment might not necessarily be
installed at the same location as the system equipment; in some cases, it can be
sited in an adjacent substation.

1.5.2 Back-up protection

Back-up protection is installed to operate when, for whatever reason, the primary
protection does not work. To achieve this, the back-up protection relay has a sen-
sing element that may or may not be similar to the primary protection, but that also
includes a time-delay facility to slow down the operation of the relay so as to allow
time for the primary protection to operate first. One relay can provide back-up
protection simultaneously to different pieces of system equipment. Equally the
same equipment can have a number of different back-up protection relays, and it is
quite common for a relay to act as primary protection for one piece of equipment
and as back-up for another. It is convenient to introduce here the term ‘redundant’,
which is related to back-up. Redundant is a duplicate of a protective system.
Back-up oversees a protection scheme and its operation necessarily is delayed.

CH001 10 May 2011; 14:33:12

 Introduction 7

1.6 Directional protection

An important characteristic of some types of protection is their capacity to be able

to determine the direction of the flow of power and, by this means, their ability to
inhibit opening of the associated switch when the fault current flows in the opposite
direction to the setting of the relay. Relays provided with this characteristic are
important in protecting mesh networks, or where there are various generation
sources, when fault currents can circulate in both directions around the mesh.
In these cases, directional protection prevents the unnecessary opening of switch-
gear and thus improves the security of the electricity supply. On protection sche-
matic diagrams, the directional protection is usually represented by an arrow
underneath the appropriate symbol, indicating the direction of current flow for
relay operation.
Example 1.1: Using the power system shown in Figure 1.5, examples are given
where there has been incorrect operation of protection and the associated breakers,
leading to the operation of back-up protection to isolate the fault from the system,
followed by an example of correct relay operation, with a final example of unne-
cessary relay operation. The directional protection is indicated by the arrows below
the corresponding breakers.

A B C
G2 G3 G4
2 5 9

G1
1 3 4 6 10

F1 F2 F3

7 8 11

F4

Figure 1.5 Power system for Example 1.1

Table 1.1 shows the breakers that failed to open and those that were tripped by
the primary protection and by the back-up protection.
For fault F1, the protection correctly tripped breaker 4 to open one end of the
faulted feeder. With breaker 3 failing to open, breakers 1 and 2 were tripped by
back-up protection to stop fault current flowing into the fault from generators

CH001 10 May 2011; 14:33:12

8 Protection of electricity distribution networks

Table 1.1 Relay/breaker operations for Example 1.1

Case Breakers Breakers Tripped by Tripped by

that that primary back-up
operated maloperated protection protection
F1 1, 2, 4 3 4 1, 2
F2 3, 5, 8 6 – 3, 5, 8
F3 10 – 10 –
F4 8, 11 8 11 –

G1 and G2. With fault F2, when breaker 6 failed to operate, the directional pro-
tection on breakers 3 and 8 operated to open the incoming feeders from the adjacent
busbars, and the back-up protection on breaker 5 tripped to stop G3 feeding into
the fault.
Fault F3 was correctly cleared by the tripping of feeder breaker 10. Fault F4
was correctly cleared by the operation of breaker 11, so that the tripping of breaker

G1

1
Termoyumbo 115 kV

G2 2 3 4

F1 5 6 29
10
Bajo Anchicayá 115 kV Chipichape 115 kV San Luis 115 kV

9 28
11 12 13 14 7 8

15
20 27
Pailon 115 kV Chipichape 34.5 kV San Luis 34.5 kV

F3
16 21 22 26

17 18 23 24

Tabor 115 kV Diesel 34.5 kV

F2

19 25

G3

Figure 1.6 Schematic diagram for Exercise 1.1

CH001 10 May 2011; 14:33:12

 Introduction 9

8 was incorrect. Any fault current flowing along inter-busbar feeder 7–8 before
breaker 11 opened would have been from 7 to 8. Relay 8 is directional, and
operation should not have been initiated for flows from 7 to 8. Thus, the first two
cases illustrate maloperation from a dependability point of view, with the last one
illustrating maloperation from a security standpoint.

Exercise 1.1
For the power system arrangement shown in Figure 1.6, complete Table 1.2, taking
into account the operation of the circuit breakers as shown for each fault case.
Please note that, as in Example 1.1, some of the circuit breakers that operated may
have done so unnecessarily.

Table 1.2 Relay/breaker operations for Exercise 1.1

Case Breakers Breakers Tripped by Tripped

that that primary by back-up
operated maloperated protection protection
F1 2, 3, 4, 5 2, 5
F2 21, 22, 23, 24, 27
F3 10, 11, 17, 19

CH001 10 May 2011; 14:33:13

CH001 10 May 2011; 14:33:13
 Chapter 2
Calculation of short-circuit currents

The current that flows through an element of a power system is a parameter that
can be used to detect faults, given the large increase in current flow when a
short-circuit occurs. For this reason, a review of the concepts and procedures for
calculating fault currents will be made in this chapter, together with some calcu-
lations illustrating the methods used. Although the use of these short-circuit
calculations in relation to protection settings will be considered in detail, it is
important to bear in mind that these calculations are also required for other appli-
cations, for example, calculating the substation earthing grid, the selection of
conductor sizes and for the specifications of equipment such as power circuit
breakers.

2.1 Modelling for short-circuit current calculations

Electrical faults are characterised by a variation in the magnitude of the short-circuit

current due to the effect of the equivalent system impedance at the fault point
that produces a decaying DC component, and the performance of the rotating
machinery that results in a decaying AC component.

2.1.1 Effect of the system impedance

System currents cannot change instantaneously when a fault occurs due to the
equivalent system resistances and reactances at the fault point, which result in a
decaying DC component. The rate of decay depends on the instantaneous value of
the voltage when the fault occurs and the power factor of the system at the fault
point. To perform the corresponding calculations, the treatment of electrical faults
should be carried out as a function of time, from the start of the event at time t = 0+
until stable conditions are reached, and it is therefore necessary to use differential
equations when calculating these currents. To illustrate the transient nature of the
current, consider an RL circuit as a simplified equivalent of the circuits in elec-
tricity distribution networks. This simplification is important because all the system
equipment must be modelled in some way in order to quantify the transient values
that can occur during the fault condition.

Ch002 10 May 2011; 19:16:54

12 Protection of electricity distribution networks

For the circuit shown in Figure 2.1, the mathematical expression that defines
the behaviour of the current is

di
eðtÞ ¼ L þ RiðtÞ ð2:1Þ
dt
wL

Vmax sin (wt + α) R

Figure 2.1 RL circuit for transient analysis study

This is a differential equation with constant coefficients, of which the solution

is in two parts:

iðtÞ ¼ ih ðtÞ þ ip ðtÞ

where ih(t) ¼ the solution of the homogeneous equation corresponding to the

transient period ip(t) ¼ the solution to the particular equation corresponding to the
steady-state period.
By the use of differential equation theory, which will not be discussed in detail
here, the complete solution can be determined and expressed in the following form:

V max h i
sinðwt þ a  fÞ  sinða  fÞ eðR=LÞ
t
iðtÞ ¼ ð2:2Þ
Z

where
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Z¼ R2 þ w2 L2

where a is the closing angle that defines the point on the source sinusoidal voltage
when the fault occurs, and

wL
f ¼ tan1
R

It can be seen that, in (2.2), the first term varies sinusoidally and the second
term decreases exponentially with a time constant of L/R.The first term corresponds
to the AC component, whereas the second term can be recognised as the DC
component of the current having an initial maximum value when a– f ¼ p/2, and

Ch002 10 May 2011; 19:16:54

 Calculation of short-circuit currents 13

zero value when a ¼ f (see Figure 2.2). It is impossible to predict at what point on
the sinusoidal cycle the fault will be applied and therefore what magnitude the DC
component will reach. If the tripping of the circuit, due to a fault, takes place when
the sinusoidal component is at its negative peak, the DC component reaches its
theoretical maximum value half a cycle later.

(a)

(b)

Figure 2.2 Variation of fault current with time: (a) / f ¼ 0 and

(b) / f ¼ p=2

An approximate formula for calculating the effective value of the total asym-
metrical current, including the AC and DC components, with acceptable accuracy
can be used by assuming that these components are in quadrature with the fol-
lowing expression:
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
I rms:asym: ¼ I 2rms þ I 2DC ð2:3Þ

2.1.2 Effect of rotating machinery

When a fault occurs close to the terminals of rotating machinery, a decaying AC
current is produced, similar in pattern to that flowing when an AC voltage is
applied to an RL circuit as discussed in the previous section. Here, the decaying
pattern is due to the fact that the magnetic flux in the windings of rotating
machinery cannot change instantaneously because of the nature of the magnetic
circuits involved. The reduction in current from its value at the onset, due to the
gradual decrease in the magnetic flux caused by the reduction of the m.m.f. of the

Ch002 10 May 2011; 19:16:54

14 Protection of electricity distribution networks

induction current, can be seen in Figure 2.3. This effect is known as armature
reaction.

ia

ib

ic

Figure 2.3 Transient short-circuit currents in a synchronous generator

The physical situation that is presented to a generator, and which makes the
calculations quite difficult, can be interpreted as a reactance that varies with time.
Notwithstanding this, in the majority of practical applications it is possible to take
account of the variation of reactance in only three stages without producing
significant errors. In Figure 2.4, it is to be noted that the variation of current with
time, I(t), comes close to the three discrete levels of current, I00 , I0 and I – the
subtransient, transient and steady-state currents, respectively. The corresponding
values of direct axis reactance are denoted by X d00, Xd0 and Xd, and the typical
variation with time for each of these is illustrated in Figure 2.5.

Ch002 10 May 2011; 19:16:55

 Calculation of short-circuit currents 15

I″

I(t)
I′

t
t1 t2

Figure 2.4 Variation of current with time during a fault

X(t)
Xd

Xd′

Xd″

t1 t2 t

Figure 2.5 Variation of generator reactance with time during a fault

2.1.3 Types of fault duty

Short-circuit levels vary considerably during a fault, taking into account the rapid
drop of the current due to the armature reaction of the synchronous machines and
the fact that extinction of an electrical arc is never achieved instantaneously.
Therefore, short-circuit currents have to be calculated carefully in order to obtain
the correct value for the respective applications.
The following paragraphs refer to the short-circuit currents that are specifically
used for the selection of interrupting equipment and protection relay settings – the
so-called normal duty rating. ANSI/IEEE Standards C37 and IEC 6090 refer to four

Ch002 10 May 2011; 19:16:55

16 Protection of electricity distribution networks

duty types defined as first cycle or momentary, peak, interrupting or breaking, and
time-delayed or steady-state currents.
First cycle currents, also called momentary currents, are the currents present
one half of a cycle after fault initiation. In European Standards, these values are
00
indicated by I k . These are the currents that are sensed by circuit breaker protection
equipment when a fault occurs and are therefore also called close and latch cur-
rents. They are calculated with DC offset but no AC decrement in the sources, and
using the machine subtransient reactances. Peak currents correspond to the max-
imum currents during the first cycle after the fault occurs and differ from the first
cycle currents that are totally asymmetrical rms currents.
Interrupting currents, also known as contact parting currents, are the values
that have to be cleared by interrupting equipment. In European Standards, these
values are called breaking currents and typically are calculated in the range from
three to five cycles. These currents contain DC offset and some decrement of
the AC current. Time-delayed or steady-state short-circuit currents correspond to
the values obtained between 6 and 30 cycles. These currents should not contain DC
offset, and synchronous and induction contributions should be neglected
and transient reactances or higher values should be used in calculating the currents.
Reactance values to be used for the different duties are reproduced in
Figure 2.6, based on IEEE Standard 399-1990. For each case, asymmetrical or
symmetrical rms values can be defined depending on whether the DC component pffiffiisffi
included or not. The peak values are obtained by multiplying the rms values by 2.
The asymmetrical values are calculated as the square root of the sum of the
squares of the DC component and the rms value of the AC current, i.e.
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
I rms ¼ I 2DC þ I 2AC ð2:4Þ

2.1.4 Calculation of fault duty values

The momentary current is used when specifying the closing current of switchgear.
Typically, the AC and DC components decay to 90% of their initial values after the
first half cycle. From this, the value of the rms current would then be
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
I rms:asym:closing ¼ I 2DC þ I 2AC:rms:sym:
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
!2ffi
u pffiffiffi !
u 0:9 2V 2 0:9V
¼t 00 þ 00 ð2:5Þ
Xd Xd
1:56V
¼ 00 ¼ 1:56I rms:sym:
Xd

Usually a factor of 1.6 is used by manufacturers and in international

standards so that, in general, this value should be used when carrying out similar
calculations.

Ch002 10 May 2011; 19:16:55

 Calculation of short-circuit currents 17

Reactance values for first cycle and interrupting duty calculationsa

Duty System component Reactance value for Reactance value

calculation medium- and high- for low-voltage
voltage calculations calculationsb
per IEEE Standard
C37.010-1979 and
IEEE Standard
C37.5-1979
First cycle Power company supply XS XS
(momentary All turbine generators; all 1.0 Xd″ 1.0 Xd″
calculations) hydrogenerators with amortis-
seur windings; all condensers
Hydrogenerators without 0.75 Xd″ 0.75 Xd′
amor tisseur windings
All synchronous motors 1.0 Xd″ 1.0 Xd″
Induction motors
Above 1 000 hp 1.0 Xd″ 1.0 Xd″
Above 250 hp at 3 600 rev/min 1.0 Xd″ 1.0 Xd″
All others, 50 hp and above 1.2 Xd″ 1.2 Xd″
All smaller than 50 hpf 1.67 Xd″ 1.67 Xd″
Interrupting Power company supply XS N/A
calculations All turbine generators; all 1.0 Xd″ N/A
hydrogenerators with amortis-
seur windings; all condensers
Hydrogenerators without 0.75 Xd′ N/A
amor tisseur windings
All synchronous motors 1.5 Xd″ N/A
Induction motors
Above 1 000 hp 1.5 Xd″ N/A
Above 250 hp at 3 600 rev/min 1.5 Xd″ N/A
All others, 50 hp and above 3.0 Xd″ N/A
All smaller than 50 hp Neglect N/A
a
First cycle duty is the momentary (or close-and-latch) duty for medium-/high-voltage equipment
and is the interrupting duty for low-voltage equipment.
b
Reactance (X) values to be used for low-voltage breaker duty calculations (see IEEE Std C37.13-
1990 and IEEE Std 242-1986).
c
Xd″ of synchronous-rotating machines is the rated-voltage (saturated) direct-axis subtransient
reactance.
d
Xd′ of synchronous-rotating machines is the rated-voltage (saturated) direct-axis transient reactance.
e
Xd″ of induction motors equals 1 divided by per-unit locked-rotor current at rated voltage.
f
For comprehensive multivoltage system calculations, motors less than 50 hp are represented in
medium-/high-voltage short-circuit calculations (see IEEE Std 141-1993, Chapter 4).

Figure 2.6 Reactance values for first cycle and interrupting duty calculations [from
IEEE Standard 399-1990; reproduced by permission of the IEEE]

Ch002 10 May 2011; 19:16:55

18 Protection of electricity distribution networks

The peak value is obtained by arithmetically adding together the AC and DC

components. Itpshould
ffiffiffi be noted that, in this case, the AC component is multiplied
by a factor of 2. Thus,

I peak ¼ I DC þ I AC
pffiffiffi ! pffiffiffi !
0:9 2V 0:9 2V
¼ 00 þ 00 ð2:6Þ
Xd Xd
¼ 2:55I rms:sym:
When considering the specification for the switchgear opening current, the
so-called rms value of interrupting current is used in which, again, the AC and DC
components are taken into account, and therefore,
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
I rms:asym:int: ¼ I 2DC þ I 2AC:rms:sym:int:

Replacing the DC component by its exponential expression gives

rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
pffiffiffi 2 ffi
I rms:asym:int: ¼ 2I rms:sym:int: e ðR=LÞ t
þ I rms:sym:int:
2
ð2:7Þ
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
¼ I rms:sym:int: 2e2ðR=LÞt þ 1

The expression (Irms.asym.int/Irms.sym.int) has been drawn for different values of

X/R, and for different switchgear contact separation times, in ANSI Standard
C37.5-1979. The multiplying factor graphs are reproduced in Figure 2.7.
As an illustration of the validity of the curves for any situation, consider a
circuit breaker with a total contact separation time of two cycles – one cycle due to
the relay and one related to the operation of the circuit breaker mechanism. If the
frequency, f, Hz and the ratio X/R is given as 50, with t ¼ 2 cycles ¼ 0.033s, then
X/R ¼ wL/R ¼ 50. Thus, L/R ¼ 50/w ¼ 50/2 pf ¼ 0.132. Therefore,
I asym: pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffi
¼ 2 eð0:0332Þ=0:132 þ 1 ¼ 1:49
I sym:
as can be seen from Figure 2.7.
In protection coordination studies, rms symmetrical interrupting current values
are normally used when setting the time-delay units of the relays. For setting the
instantaneous elements, the same values should be used but multiplied by a factor
that depends on the application, as will be discussed later on.

2.2 Methods for calculating short-circuit currents

Symmetrical faults, that is, three-phase faults and three-phase-to-earth faults, with
symmetrical impedances to the fault, leave the electrical system balanced and

Ch002 10 May 2011; 19:16:55

 Calculation of short-circuit currents 19

150

140

130

120

110

100

90

4
3
2
1
Ratio X/R

80

70

60

es
50

ycl
40 e, c
im
gt

30
in

rt
20 t- pa
n ta c
Co
10

0
1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7
Multiplying factors

Figure 2.7 Multiplying factors for three-phase and line-to-earth faults (total
current rating basis) [from IEEE Standard C37.5-1979; reproduced
by permission of the IEEE]

therefore can be treated by using a single-phase representation. This symmetry is

lost during asymmetrical faults – line-to-earth, line-to-line and line-to-line-to-earth –
and in these cases a method of analysing the fault that provides a convenient means
of dealing with the asymmetry is required. In 1918 a method of symmetrical
components was proposed in which an unbalanced system of n related phases could
be replaced by a system of n balanced phases that were named the symmetrical
components of the original phases. Although the method can be applied to any
unbalanced polyphase system, the theory is summarised here for the case of an
unbalanced three-phase system.
When considering a three-phase system, each vector quantity, voltage or cur-
rent, is replaced by three components so that a total of nine vectors uniquely
represent the values of the three phases. The three-system balanced phasors are
designated as:

Ch002 10 May 2011; 19:16:55

20 Protection of electricity distribution networks

1. Positive-sequence components, which consist of three phasors of equal mag-

nitude, spaced 120 apart, and rotating in the same direction as the phasors in
the power system under consideration, i.e. the positive direction.
2. Negative-sequence components, which consist of three phasors of equal
magnitude, spaced 120 apart, rotating in the same direction as the positive-
sequence phasors but in the reverse sequence.
3. Zero-sequence components, which consist of three phasors equal in magnitude
and in phase with each other, rotating in the same direction as the positive-
sequence phasors.
With this arrangement, voltage values of any three-phase system, Va, Vb and
Vc, can be represented as follows:

V a ¼ V a0 þ V a1 þ V a2
V b ¼ V b0 þ V b1 þ V b2
V c ¼ V c0 þ V c1 þ V c2

It can be demonstrated that

V b ¼ V a0 þ a2 V a1 þ aV a2
V c ¼ V a0 þ aVa1 þ a2 V a2

where a is a so-called operator that gives a phase shift of 120 clockwise and a
multiplication of unit magnitude, i.e. a ¼ 1ff120 , and a2 similarly gives a phase
shift of 240 , i.e. a2 ¼ 1ff240 .
Therefore, the following matrix relationship can be established:
2 3 2 3 2 3
Va 1 1 1 V a0
4 V b 5 ¼ 4 1 a2 a 5  4 V a1 5
Vc 1 a a2 V a2

Inverting the matrix of coefficients, we get

2 3 2 3 2 3
V a0 1 1 1 Va
1
4 V a1 5 ¼ 4 1 a a2 5  4 V b 5
3
V a2 1 a2 a Vc

From the above matrix, it can be deduced that

1
V a0 ¼ ðV a þ V b þ V c Þ
3
1
V a1 ¼ ðV a þ aVb þ a2 V c Þ
3
1
V a2 ¼ ðV a þ a2 V b þ aV c Þ
3

Ch002 10 May 2011; 19:16:55

 Calculation of short-circuit currents 21

The foregoing procedure can also be applied directly to currents and gives
I a ¼ I a0 þ I a1 þ I a2
I b ¼ I a0 þ a2 I a1 þ aI a2
I c ¼ I a0 þ aI a1 þ a2 I a2
Therefore
1
I a0 ¼ ðI a þ I b þ I c Þ
3
1
I a1 ¼ ðI a þ aI b þ a2 I c Þ ð11:1Þ
3
1
I a2 ¼ ðI a þ a2 I b þ aI c Þ
3
In three-phase systems, the neutral current is equal to In ¼ Ia + Ib + Ic and,
therefore, In ¼ 3Ia0.
By way of illustration, a three-phase unbalanced system is shown in Figure 2.8
together with the associated symmetrical components.

2.2.1 Importance and construction of sequence networks

The impedance of a circuit in which only positive-sequence currents are circulating
is called the positive-sequence impedance and, similarly, those in which only
negative- and zero-sequence currents flow are called the negative- and zero-
sequence impedances. These sequence impedances are designated Z1, Z2 and Z0,
respectively, and are used in calculations involving symmetrical components. Since
generators are designed to supply balanced voltages, the generated voltages are of
positive sequence only. Therefore, the positive-sequence network is composed of
an e.m.f. source in series with the positive-sequence impedance. The negative- and
zero-sequence networks do not contain e.m.f.s but only include impedances to the
flow of negative- and zero-sequence currents, respectively.
The positive- and negative-sequence impedances of overhead line circuits are
identical, as are those of cables, being independent of the phase if the applied
voltages are balanced. The zero-sequence impedances of lines differ from the
positive- and negative-sequence impedances since the magnetic field creating the
positive- and negative-sequence currents is different from that for the zero-
sequence currents. The following ratios may be used in the absence of detailed
information. For a single circuit line, Z0/Z1¼2 when no earth wire is present and 3.5
with an earth wire. For a double circuit line, Z0/Z1 ¼ 5.5. For underground cables,
Z0/Z1 can be taken as 1–1.25 for single-core cables, and 3–5 for three-core cables.
For transformers, the positive- and negative-sequence impedances are equal
because in static circuits these impedances are independent of the phase order,
provided that the applied voltages are balanced. The zero-sequence impedance is
either the same as the other two impedances, or infinite, depending on the
transformer connections. The resistance of the windings is much smaller and can
generally be neglected in short-circuit calculations.

Ch002 10 May 2011; 19:16:55

22 Protection of electricity distribution networks

Vc

Va

Vb

Va = 8.0 0° Vb = 6.0 –90° Vc = 16.0 143.1°

Vc1

Va1
Vb2
Vc2

Va0 Vb0 Vc0

Va2

Vb1

Va1 = 9.8 18.4° Va2 = 4.3 –86.2° Va0 = 2.0 143.1°

Vb1 = 9.8 –101.6° Vb2 = 4.3 33.8° Vb0 = 2.0 143.1°
Vc1 = 9.8 138.4° Vc2 = 4.3 –206.2° Vc0 = 2.0 143.1°

Vc0
Vc2

Vc Vc1
Va1
Va Va2

Va0
Vb
Vb1
Vb0

Vb2

Figure 2.8 Symmetrical components of an unbalanced three-phase system

Ch002 10 May 2011; 19:16:56

 Calculation of short-circuit currents 23

When modelling small generators and motors, it may be necessary to take

resistance into account. However, for most studies, only the reactances of syn-
chronous machines are used. Three values of positive reactance are normally
quoted–subtransient, transient and synchronous reactances, denoted by X d00, Xd0 and
Xd. In fault studies, the subtransient and transient reactances of generators and
motors must be included as appropriate, depending on the machine characteristics
and fault clearance time. The subtransient reactance is the reactance applicable
at the onset of the fault occurrence. Within 0.1 s, the fault level falls to a value
determined by the transient reactance and then decays exponentially to a steady-
state value determined by the synchronous reactance. Typical per-unit reactances
for three-phase synchronous machines are given in Table 2.1.
Table 2.1 Typical per-unit reactances for three-phase synchronous machines

00 0
Type of machine Xd Xd Xd X2 X0
Turbine (2-pole) 0.09 0.15 1.20 0.09 0.03
Generator (4-pole) 0.14 0.22 1.70 0.14 0.07
Salientpole (withdampers) 0.20 0.30 1.25 0.20 0.18
Generator (withoutdampers) 0.28 0.30 1.20 0.35 0.12

In connecting sequence networks together, the reference busbar for the

positive- and negative-sequence networks is the generator neutral that, in these
networks, is at earth potential, so only zero-sequence currents flow through the
impedances between neutral and earth. The reference busbar for zero-sequence net-
works is the earth point of the generator. The current that flows in the impedance Zn
between the neutral and earth is three times the zero-sequence current. Figure 2.9
illustrates the sequence networks for a generator. The zero-sequence network carries
only zero-sequence current in one phase that has an impedance of Z0 ¼ 3Zn + Ze0.
The voltage and current components for each phase are obtained from the equa-
tions given for the sequence networks. The equations for the components of voltage,
corresponding to the phase a of the system, are obtained from the point a on phase a
relative to the reference busbar and can be deduced from Figure 2.9 as follows:
V a1 ¼ Ea  I a1 Z 1
V a2 ¼ I a2 Z 2
V a0 ¼ I a0 Z 0
where
Ea ¼ no-load voltage to earth of the positive-sequence network
Z1 ¼ positive-sequence impedance of the generator
Z2 ¼ negative-sequence impedance of the generator
Z0 ¼ zero-sequence impedance of the generator (Zg0) plus three times the
impedance to earth

Ch002 10 May 2011; 19:16:56

24 Protection of electricity distribution networks

Ia1
a Reference bus

Z1 –
Ea
+
Ea +
Va1
Ec – –
– Eb
Z1 +
+ Z1 Z1
Ib1
c b
a
Ic1
Ia1
(a) (b)

Ia2
a
Reference bus

Z2

Va2
Z2
Z2 Z2
Ib2
c b

Ic2 a
Ia2
(c) (d )
Ia0
a
Ia0 Reference bus
Ib0 = Ia0
Ic0 = Ia0 Ze0

3Zn
Zn Va0
Z0
Ze0 Ze0
Ib0 Ze0
c b
a
Ic0
Ia0
(e) (f)

Figure 2.9 Equivalent sequence networks and current flows for a synchronous
generator: (a) positive-sequence current; (b) positive-sequence
network; (c) negative-sequence current; (d) negative-sequence
network; (e) zero-sequence current; (f) zero-sequence network

Ch002 10 May 2011; 19:16:56

 Calculation of short-circuit currents 25

The above equations can be applied to any generator that carries unbalanced
currents and are the starting point for calculations for any type of fault. The same
approach can be used with equivalent power systems or loaded generators, Ea then
being the voltage behind the reactance before the fault occurs.

2.2.2 Calculation of asymmetrical faults using symmetrical

components
The positive-, negative- and zero-sequence networks, carrying currents I1, I2 and I0,
respectively, are connected together in a particular arrangement to represent a given
unbalanced fault condition. Consequently, in order to calculate fault levels using the
method of symmetrical components, it is essential to determine the individual
sequence impedances and combine these to make up the correct sequence networks.
Then, for each type of fault, the appropriate combination of sequence networks is
formed in order to obtain the relationships between fault currents and voltages.

2.2.2.1 Line-to-earth fault

The conditions for a solid fault from line a to earth are represented by the equations
Ib ¼ 0, Ic ¼ 0 and Va ¼ 0. As in the previous equations, it can easily be deduced that
Ia1 ¼ Ia2 ¼ Ia0 ¼ Ea/(Z1+Z2+Z0). Therefore, the sequence networks will be con-
nected in series, as indicated in Figure 2.10a. The current and voltage conditions
are the same when considering an open-circuit fault in phases b and c, and thus the
treatment and connection of the sequence networks will be similar.

2.2.2.2 Line-to-line fault

The conditions for a solid fault between lines b and c are represented by the
equations Ia ¼ 0, Ib ¼ Ic and Vb ¼ Vc. Equally it can be shown that Ia0 ¼ 0 and
Ia1 ¼ Ea/(Z1 + Z2) ¼Ia2. For this case, with no zero-sequence current, the zero-
sequence network is not involved and the overall sequence network is composed
of the positive- and negative-sequence networks in parallel as indicated in
Figure 2.10b.

2.2.2.3 Line-to-line-to-earthfault
The conditions for a fault between lines b and c and earth are represented by the
equations Ia ¼ 0 and Vb ¼ Vc ¼ 0. From these equations, it can be proved that
Ea
I a1 ¼
Z 1 þ ½Z 0 Z 2 =ðZ 0 þ Z 2 Þ

The three sequence networks are connected in parallel as shown in

Figure 2.10c.

2.2.3 Equivalent impedances for a power system

When it is necessary to study the effect of any change on the power system, the
system must first of all be represented by its corresponding sequence impedances.

Ch002 10 May 2011; 19:16:56

26 Protection of electricity distribution networks

Ea

Va1
Z1

Z2 Va2

Va0
Z0

Ia1

(a)

Ea

Va1 Va2
Z1 Z2

Ia1 Ia2

(b)

Ea

Va1 Va2 Va0

Z1 Z2 Z0

Ia1 Ia2 Ia0

(c)

Figure 2.10 Connection of sequence networks for asymmetrical faults: (a) phase-
to-earth fault; (b) phase-to-phase fault; (c) double-phase-to-earth
fault

Ch002 10 May 2011; 19:16:56

 Calculation of short-circuit currents 27

The equivalent positive- and negative-sequence impedances can be calculated

directly from

V2
Z¼
P

where
Z ¼ equivalent positive- and negative-sequence impedances
V ¼ nominal phase-to-phase voltage
P ¼ three-phase short-circuit power
The equivalent zero sequence of a system can be derived from the expressions
of sequence components referred to for a single-phase fault, i.e. Ia1 ¼ Ia2 ¼ Ia0 ¼
VLN/(Z1 + Z2 + Z0), where VLN is the line-to-neutral voltage.
For lines and cables, the positive- and negative-sequence impedances are equal.
Thus, on the basis that the generator impedances are not significant in most distribution
network fault studies, it may be assumed that overall Z2 ¼ Z1, which simplifies the
calculations. Thus, the above formula reduces to Ia ¼ 3 Ia0 ¼ 3VLN/(2Z1 + Z0), where
VLN is the line-to-neutral voltage and Z0 ¼ (3VLN/Ia)– 2Z1.

2.3 Supplying the current and voltage signals to

protection systems
In the presence of a fault, the current transformers (CTs) circulate current propor-
tional to the fault current to the protection equipment without distinguishing
between the vectorial magnitudes of the sequence components. Therefore, in the
majority of cases, the relays operate on the basis of the corresponding values of
fault current and/or voltages, regardless of the values of the sequence components.
It is very important to emphasise that, given this, the advantage of using symme-
trical components is that they facilitate the calculation of fault levels even though
the relays in the majority of cases do not distinguish between the various values of
the symmetrical components.
In Figure 2.11, the positive- and negative-sequence values of current and
voltage for different faults are shown together with the summated values of current
and voltage. Relays usually only operate using the summated values in the right-
hand columns. However, relays are available that can operate with specific values
of some of the sequence components. In these cases, there must be methods for
obtaining these components, and this is achieved by using filters that produce the
mathematical operations of the resultant equations to resolve the matrix for vol-
tages and for currents. Although these filters can be constructed for electromagnetic
elements, the growth of electronics has led to their being used increasingly in logic
circuits. Among the relays that require this type of filter in order to operate are
those used in negative-sequence and earth fault protection.

Ch002 10 May 2011; 19:16:56

 Fault Positive- Negative- Zero- Fault Fault Positive- Negative- Zero- Fault 28
sequence sequence sequence currents sequence sequence sequence voltage
network network network network network network
current current current voltage voltage voltage
a1 a a1
Zero a1
a, b, c c a, b, c
c1 b1 b c1 b1 fault
a1 a2 a a1 a2 a
c2 c2 b
a, b a, b
c1 b1 b2 c=0 b c1 b1 b2 c a=b
a1 c2 a1 c2 a
b2 c b b2
b, c b, c
c1 b1 a2 a=0 c1 b1 a2 b=c
a1 b=0 a a1 a
b2 a2 b2 a2 c
c, a c, a
c1 b1 c2 c c1 b1 c2 a=c b

Ch002
a1 a2 c2 a0, b0, c0 a a1 a2 c2
a, b, e a, b, e c
c1 b1 b2 c=0 b c1 b1 b2 a0, b0, c0 a=b=0
a1 c2 a=0 a1 c2 a0, b0, c0 a
b2 b2
b, c, c b, c, c
c1 b1 a0, b0, c0 c b c1 b1
a2 a2 b=c=0
a1 a a1
Protection of electricity distribution networks

b2 a2 b2 a2
c, a, e c, a, e b
c1 b1 a0, b0, c0 c b=0 c1 b1 a0, b0, c0 a=c=0
c2 c2

10 May 2011; 19:16:56

a1 a2 a0, b0, c0 a a1 a2 a=0
a, e b1 b2 a, e c b
c1 c2 b=c=0 c1 b1 b2 c2 a , b , c
0 0 0
a1 c2 a1 c2 a
b, e b b, e
c1 b1 a2 b2 a , b , c
0 0 0 a=c=0 c1 b1 a2 b2 a , b , c
0 0 0
c b=0
a1 b2 a1 b2 a0, b0, c0 a
c, e c c, e
c1 b1 c2 a2 a , b , c a=b=0 c1 b1 c2 a2 b
0 0 0 c=0
(a) (b)

Figure 2.11 Currents and voltages for various types of faults: (a) sequence currents for different
types of faults and (b) sequence voltages for different types of faults
 Calculation of short-circuit currents 29

2.4 Calculation of faults by computer

The procedure for calculating fault levels starts by taking the single-line diagram of
the system under analysis; collecting the sequence impedances for all the compo-
nents; calculating the Thevenin equivalent of neighbouring systems; collecting
background data including machine impedances; the length, conductor diameter
and the configuration of the feeders; the values of connections to earth, etc. Having
obtained these values, an updated single-line diagram can be produced, indicating
the positive-sequence impedance values referred to the respective base quantities.
The corresponding positive-, negative- and zero-sequence networks can then be
built up to form the basis of the calculations of the voltage and current under fault
conditions.
Having collected and processed the basic information, the calculation of fault
levels for large power systems is now invariably carried out using computers, given
the vast facilities that are now available, in both hardware and software. For small
systems, however, hand calculations can still be used since short-circuit calcula-
tions do not require an iterative process. A large proportion of the existing pro-
grams have been developed with interactive algorithms whose principal
characteristic is the man–machine dialogue. This is much superior to the batch
process used earlier. The interactive program permits examination of the results as
they are printed on the screen or via the printer and enables the user to select those
results that are important in the study. When investigating electrical faults, this
method speeds up the calculations considerably bearing in mind that, apart from
being able to keep direct control on the performance of the program (i.e. when to
stop/check, interrogate files, print out, etc.), it is possible to alter variables such as
the type of fault being analysed, the faulted busbar and values of impedance.
Modern software packages enable the following features to be carried out:
● Perform calculations not only for the standard fault types, i.e. three-phase, line-
to-line, line-to-line-to-earth and line-to-earth, but also for faults such as those
between systems of different voltages.
● Indicate the fault contributions (sequence and phase values) from the different
elements whether or not they are associated with the faulted node.
● Include pre-fault values.
● Calculate the different duties associated with a fault and handle IEEE and IEC
Standards that can be slightly different, especially in the pre-fault voltage level.
● Calculate simultaneous faults.
● Calculate faults along different line lengths.

Ch002 10 May 2011; 19:16:56

Ch002 10 May 2011; 19:16:56
 Chapter 3
Classification and function of relays

A protection relay is a device that senses any change in the signal it is receiving,
usually from a current and/or voltage source. If the magnitude of the incoming
signal is outside a pre-set value, the relay will carry out a specific operation, gen-
erally to close or open electrical contacts to initiate some further operation, for
example the tripping of a circuit breaker.

3.1 Classification

Protection relays can be classified in accordance with their construction, the

incoming signal and function.

3.1.1 Construction
● electromechanical
● solid state
● microprocessor
● numerical
● non-electrical (thermal, pressure, etc.)

3.1.2 Incoming signal

● current
● voltage
● power
● frequency
● temperature
● pressure
● speed
● others

3.1.3 Function
● overcurrent
● directional overcurrent

CH003 10 May 2011; 16:27:5

32 Protection of electricity distribution networks

● distance
● overvoltage
● differential
● reverse power
● other

3.1.4 International identification of electrical devices

The international classification for the more common relays, which is used in the
following chapters, is given below:
21: distance relay
24: volts/hertz
25: synchronising or synchronism-check device
26: thermal device
27: undervoltage relay
32: reverse-power relay
37: undercurrent or underpower relay
40: relay for field excitation
41: field circuit breaker
43: manual transfer or selector device
46: negative-sequence current relay
47: negative-sequence voltage relay
49: thermal relay
50: instantaneous overcurrent relay
51: time-delay overcurrent relay
52: circuit breaker
55: power factor relay
59: overvoltage relay
60: voltage or current balance relay
62: time-delay relay
63: pressure relay, for flow or level of liquid or gases
64: earth protection relay
67: directional overcurrent relay
68: blocking relay
74: alarm relay
78: out-of-step relay
79: reclosing relay
81: frequency relay
85: carrier or pilot-wire receiver relay
86: lockout relay
87: differential relay
94: auxiliary tripping relay
In some cases, a letter is added to the number associated with the protection in
order to specify its place of location, for example, G for generator and T for

CH003 10 May 2011; 16:27:5

 Classification and function of relays 33

transformer. Non-electrical relays are outside the scope of this book and therefore
are not referred to.

3.2 Electromechanical relays

These relays are constructed with electrical, magnetic and mechanical components
and have an operating coil and various contacts, and are very robust and reliable.
They are also referred to as electromagnetic relays due to their magnetic compo-
nents. Their construction characteristics can be classified in three groups, as
detailed below.

3.2.1 Attraction relays

Attraction relays can be supplied by AC or DC, and operate by the movement of a
piece of metal when it is attracted by the magnetic field produced by a coil. There
are two main types of relays in this class. The attracted armature type, which is
shown in Figure 3.1, consists of a bar or plate of metal that pivots when it is
attracted towards the coil. The armature carries the moving part of the contact that
is closed or opened, according to the design, when the armature is attracted to the
coil. The other type is the piston or solenoid type relay, illustrated in Figure 3.2, in
which a bar or piston is attracted axially within the field of the solenoid. In this
case, the piston also carries the operating contacts.

Fixed contact
Moving contact

Restraining
spring

Armature
Coil

Pivot

Figure 3.1 Armature type relay

CH003 10 May 2011; 16:27:5

Exploring the Variety of Random
Documents with Different Content
the pavement be sloped away from these windows with considerable
pitch, not less-than 1 inch in 1 foot, as otherwise the water is likely
to lay against the window sill or run under it, causing it to rapidly
decay, the capillary attraction of the window frame drawing up the
water.

Fig. 10
29. Window Openings.—Referring to Fig. 9, which shows the
face view of a bay of the wall illustrated in section in Fig. 7, the
details of the several window openings in the walls may be studied.
The basement windows are independent frames with double-hung
sash, a rowlock brick arch supporting the brickwork over the window
head. In the practice and design of window heads for mill buildings,
it is usual to make the radius of the window head equal to the width
of the reveal. In this instance, the distance across the opening is 4
feet 3 inches, and the radius of the arched head is the same
dimension.
The windows throughout the balance of the building are twin
windows, double hung, and the construction of the window frame
and sash is shown in the drawing. This frame is what is known as a
reveal frame, and is built in as the brickwork progresses.
Sometimes the frame is slipped in from the back, as shown in Fig.
10, and when this is the case the work can be carried along without
waiting for the window frames.

Fig. 11
As distinguishable from the reveal frame, there is the plank-
frame construction, which is not built into the brickwork, but is
built up as shown in Fig. 11. When it is desirable to have the central
mullion a, Fig. 9, as narrow as possible, the box construction
indicated on the drawing is done away with and the window is hung
by means of overhead pulleys, the weights operating in the boxes at
the sides.

STANDARD SLOW-BURNING
CONSTRUCTION
30. A type of factory construction more usual than that
previously described is illustrated in Fig. 12. In this illustration, it will
be noticed that the main girders bear on wall pilasters, and the
spandrel filling between the pilasters is kept as thin as possible. The
usual reveal window frame is used, as shown at a, and the soffit of
the arch over the window openings is checked at the head of the
opening to provide a wind and water stop as at b. In this
construction, which is probably the best, though it does not possess
the advantage of giving the maximum amount of window space,
and, consequently, light in the building, a rowlock or bonded brick
arch is used over the window frames. By means of this construction,
either the window frame may be built in place, or the windows may
be slipped in from the back against a rabbet formed in the
brickwork. The arch over the window head is indicated at c.
 Fig. 12
31. Floor Construction.—The floor construction consists of
heavy timber girders, no dimension of which may be less than 6
inches, as otherwise it would not comply with the requirements of
slow-burning construction. The floor planking consists of 3- or 4-inch
tongued-and-grooved spruce, or yellow-pine planking, planed on the
under side, and thoroughly spiked to the girder. Planking of the
former thickness may be used for clear spans as great as 8 feet,
while the latter thickness may be used for up to 10-foot or even 12-
foot spans, if the loads are light. The girders are indicated at d, and
 the floor planking at e. Usually the girders,
in order to obtain the requisite strength, are
made of long-leaf yellow pine. On the top
of the spruce planking is placed a finished
maple floor. This floor is made from either
1-inch maple, which finishes as ⅞ inch, or
1¼-inch maple, which finishes as 1⅛
inches, in thickness. Neponsett sheathing
paper, or deadening felt, is placed between
the spruce planking and the finished maple
flooring for the purpose of preventing dust
from percolating through. This sheathing
Fig. 13 paper or felt is sometimes made waterproof
to prevent leakage due to water used for
fire-extinguishing purposes.
Frequently, the brickwork is corbeled out, as indicated at f, in
order to form a fire-stop between floors, or at least to prevent an
open joint at this place. Where the walls are offsetted, as shown in
Fig. 13, there is no need of corbeling out, for the offset in the
brickwork can be made to form the fire or dust stop.
32. Where heavy yellow-pine girders bear on brick walls, it is
usual to obtain the requisite bearing area by the use of cast-iron
bearing plates, as indicated in Fig. 14 (a), (b), and (c). In (a) is
shown an ordinary flat plate that has an area figured so that the
load on the brickwork will not exceed its ultimate stress, which for
brickwork laid in lime-and-cement mortar is about 150 pounds per
square inch, while for brickwork laid in cement mortar, it is in the
neighborhood of 200 pounds per square inch. This plate is usually
cast with a lug on the back, as at a, to be built in the brickwork, and
dowel-pins, or a lip, as at b, over which the girder is fitted, or
notched. By this means, a tie to the wall is obtained. There is
difficulty, however, in using such a connection, for the carpenters on
the job frequently miscut their beams, so that the notchings or
borings at b do not come where they should, and to remedy the
defect, the notchings, or borings, are cut or gouged out, so that
frequently the pin or lip at b is not brought to bear against the
timber.
(a)

(b )
 ( c)
Fig. 14
A more practicable bearing plate is illustrated in Fig. 14 (b). Here,
instead of providing dowels, or a lip, to set into the girder, the top of
the plate is cast with teeth, as indicated at c. While these teeth tend
to destroy fibers at the bottom of the beam, they nevertheless sink
into the timber, creating great friction, and thus accomplish a tie to
the wall fully as efficient as a dowel-pin, or lip, let into the timber
would be.
Probably the most common form of bearing plate is that
illustrated in Fig. 14 (c), which is known as the Goetz-Mitchell
bearing box. This is usually built flared, as indicated in the
illustration, so that when built into the brickwork it will have a hold
in it, and the timber acts as a tie by being notched over the lip, as at
d in this figure. These Goetz-Mitchell boxes are generally provided
with a plate that sets on top of them, on which the brickwork may
be built, and not infrequently the sides of the boxes are grooved so
that the ends of the girders are ventilated.
 Fig. 15
33. Window Heads.—In Fig. 12 was shown a form of window
head that is the best for strength, but possesses the disadvantage of
lowering the top of the window, thus cutting off light to the room,
which is a serious objection where the room is wide, or where it
depends on the windows in one side for lighting the entire floor
area. In order to keep the window head up near the under side of
the floor construction, an I beam, lintel, or some similar form of
support for the brickwork over the head that takes up little room,
must be employed. A construction using shallow I beams is
illustrated in Fig. 15. Here the window head is directly beneath the
rough flooring; and while the outside face of the window is formed
with an arch, the brickwork above the window head is supported on
shallow I beams. This figure illustrates a section through the wall
extending parallel with the main girders, a bearing being obtained
for the floor planking by bolting to the I beams a bearing strap a.
 This construction would not be permitted in some of the larger
cities, as the building laws require that all steel beams supporting
brickwork must be fireproofed. Consequently, a steel lintel of this
construction would have to be surrounded with concrete, and the
window head dropped somewhat to allow a bearing for the floor
planking, or some other form of construction adopted.

FACTORY BUILDINGS OF
REINFORCED CONCRETE
 Fig. 16
34. Within the last few years, the cost of the best Portland
cement has been so materially reduced that concrete has become an
available material for the construction of factories. Unless used in
great masses, however, it has not the strength to support the
necessary floor loads without the use of steel reinforcement. As
explained in Design of Beams, the fibers on the bottom of all beams
subjected to transverse stress are in tension, and while concrete has
considerable resistance to compression, it offers comparatively little
to tensile stress. It is therefore necessary to reinforce the lower
portion of all beams and floor slabs as indicated at a, Fig. 16.
35. Advantages of Reinforced Concrete.—In Fig. 16, the
details of a typical reinforced-concrete factory building are
illustrated, and a building of this character may be constructed for a
cost of from 10 to 15 per cent. greater than the ordinary slow-
burning type of building. Besides, this construction possesses the
advantage of being practicable for long spans and heavy loads,
whereas in buildings of the slow-burning type, owing to the fact that
the size of the wooden beams is limited to the available commercial
timber, it is frequently impossible to design floors with girders of
large spans for floor loads of over 250 pounds per square foot. While
this is a heavy load, it is too light for some classes of work, such as
occur in printing houses and lithographing establishments where
heavy stones are used and stored. The floor loads in such buildings
sometimes amount to as much as 300 or 400 pounds per square
foot, while it is not unusual to find the load on floors in warehouses
amounting to as much as 500 pounds per square foot.
36. Strength of Concrete Columns With Steel Cores.—In
the building shown in Fig. 17, it will be noticed that the columns are
reduced in size in the lower floors, increased in the middle portion of
the building, and reduced toward the roof. The reduction in the
columns a and b is due to the fact that these columns are reinforced
with a steel core composed of structural shapes riveted together,
angles usually being employed for this purpose. In proportioning
such columns, it is good practice to figure on the ultimate safe unit
compressive stress of the steel without considering the reduction
made by the usual column formula, but to neglect, in the
consideration of the strength of the column, the resistance of the
concrete surrounding the steel core. To illustrate, if the sectional
area of the steel reinforcements in these columns equals 20 square
inches, and a safe unit fiber stress of 16,000 pounds is assumed, the
safe strength of the column will be 320,000 pounds.
 Above the second floor, the
columns are made much larger,
for here there is less steel
reinforcement, and it is
necessary to figure on the safe
bearing strength of the
concrete.
37. Strength of
Reinforced-Concrete
Columns.—In proportioning
reinforced-concrete columns, it
is customary among
conservative engineers to figure
the safe strength of the
concrete-column section at 500
pounds per square inch of
section; that is, if the column is
20 inches square, its area is 400
square inches, and its safe
strength at 500 pounds per
square inch will be 200,000
pounds. In the top floor, it is
seldom advisable to use
concrete columns less than 10
inches square, though at this
dimension they generally
possess several times the
Fig. 17 requisite amount of resistance.
All columns in reinforced
construction generally have embedded in them 3¾-inch to 1-inch
round steel rods, tied together with round iron binders, or bar-iron
straps as indicated in Fig. 16 (b).
38. Floor and Roof Construction.—In considering the floor
and roof construction of buildings built of reinforced concrete, it will
be noted from Fig. 16 that the roof slab is made 3 inches in
thickness. Such a slab made of good concrete, reinforced with ⅜-
inch steel rods, spaced 6 inches from center to center, will carry the
usual roof loads for spans up to 7 feet in the clear.
In forming the gutter for such roofs, as indicated at b, the gusset
is made by filling in with cinder concrete. Usually cast-iron eave
boxes are embedded in the concrete, and these in turn connected
with inside rain conductors.
The beams supporting the roof, when the span is from 12 to 14
feet, are made about 12 inches deep and 8 inches wide, while the
girders, also constructed of reinforced concrete, are usually made
about 3 inches deeper and 11 inches in width.
In order to make the roof impervious to moisture, a covering of
felt and slag is commonly employed. This slag joins the parapet wall
with the usual tin flashing and counter flashing, as at c, though
copper is recommended for best work.
In the floor construction of reinforced-concrete factory buildings,
the slabs forming the floor panels are made not less than 4 inches in
thickness, and seldom over 5 inches, with a 1-inch finish coat of
cement besides, if this character of finish is desired. Such a floor
slab is shown in the construction at d, Fig. 16, while the wooden
floor construction is shown in Fig. 16 (c). Here the structural feature
of the floor is a 4-inch concrete slab upon the top of which is placed
2" × 3" beveled hemlock sleepers, the space between these sleepers
being filled with cinder concrete, and the floor finish obtained by
laying 1-inch tongued-and-grooved maple floorings.
39. Reinforced-Concrete Beams and Girders.—The depth of
the beams and girders in reinforced-concrete construction varies, of
course, with the span and loads to be supported. Their width enters
little into the strength, and they may be made as narrow as possible
in order to cover the reinforcing steel. It is the best practice to make
beams and girders of the same width, for then the process of
forming the molds is greatly simplified and the cost reduced.
 In placing the reinforcement in the concrete, it should always be
at least 2 inches from the outside surface, for a distance less than
this is considered inadequate fireproofing. In order that the
reinforcing metal e, Fig. 16, may enter over the top of the
reinforcing metal at f, it is usual to make the secondary girders, or
beams, 3 inches less in depth than the main girders. To stiffen the
building, brackets are customarily introduced between the column
and girders, as illustrated at g. These brackets tend to greatly
increase the rigidity of the connection and shorten the span of the
girder somewhat.

Fig. 18
40. Construction at Window Heads.—Where it is necessary
to have the window head near the top of the ceiling, reinforced-
concrete construction lends itself readily to the requirements of this
condition, for even where girders are supported over the window
head, the construction may be followed out, as indicated at h, Fig.
16. Where it is desired to have the window head raised still higher, a
construction similar to that shown in Fig. 18 may be used. In this
case, however, care must be taken to have the girders bear on the
piers between the windows, and to have no intermediate beams.
(a)

(b )
 Fig. 19
41. Column Footings.—With factory buildings of more than five
or six stories in height, great pressure is transmitted to the soil from
the base of the bottom column, and as it is necessary with soils of
even fairly good bearing capacity to have footings beneath the piers
supporting columns of from 6 to 10 feet square, adequate means of
providing these footings must be obtained. In Fig. 19 (a) and (b) are
shown two types of footings for concrete columns. In (a) is indicated
a reinforced-concrete column with a steel core. In such an instance,
all the load is transmitted by the steel core through its angle plates
and webbing at the foot to grillage beams. These grillage beams are,
however, not made sufficiently large to transmit the load to the soil,
but merely to distribute the load on the bed of concrete. The spread
portion of the footing is reinforced with steel rods a, a crossed each
way, and longitudinal shear is taken up in the footing by means of
stirrups b b. This is the usual type of footing construction under
reinforced-concrete factory columns.
Where, however, the column is not reinforced with a steel core,
but is merely a pier, footings may be designed as illustrated in Fig.
19 (b). Here the base of the column is enlarged in order to better
distribute the load on the several steps of the footing, and where the
bottom step has a considerable overhang, it is reinforced with steel
rods and stirrups, as indicated.
42. Detail of Slab and Girder Reinforcement.—In the
previous article, the general construction of the floors and column
supports of a factory building was explained. By referring to Fig. 20,
it will be shown how the girders and beams are reinforced with the
steel bars. In this figure, a plan is indicated at (a) and an elevation
at (b). The rod reinforcement of the slab is shown in the plan at a,
a. It will be noticed that over every other beam these rod
reinforcements lap, or break joints, and that some additional tie or
reinforcement is placed over the girders, as indicated by b, b. These
latter rods tend to tie in the floor slabs still more rigidly than can be
accomplished with their individual reinforcement.
 Referring to the elevation (b), it will be noticed that all the
reinforcement of the beams is not usually carried along the lower
portion of the girder for its entire distance, but that some of the
reinforcement is bent up at a point about one-quarter of the span
from the abutment, in the form of a camber rod. By arranging the
reinforcing rods in this manner, an additional stirrup action, or tie, to
the girder supports is provided, and the oblique section made by a
horizontal line passing through these rods tends to provide
additional resistance to the horizontal shear in the beams and also
provide for negative bending moment produced in the beams near
the support. To further provide for this, shear stirrups are placed
closer together, toward the abutments, as indicated at c, c. These
stirrups are ordinarily light pieces of bar iron bent in a U-shape, and
sometimes bent around the rod reinforcement, a detail of this stirrup
being shown in Fig. 20 (c).
 Fig. 20

STEEL-FRAME MILL BUILDINGS

43. There is a type of building which, while not distinctly mill
construction as usually understood, is frequently used for one-story
buildings, such as rolling mills, cement works, machine shops,
foundries, rail yards, and buildings of this class.
The essential feature of these buildings is a steel-roof truss
supported on steel columns, the columns being braced both to the
truss and longitudinally of the building. It is usually the purpose in
the design of such buildings to neglect everything but the necessary
stability and the first cost. The steelwork, consequently, is of the
lightest possible construction, usually designed for a unit fiber stress
of from 18,000 to 20,000 pounds, and the covering of the sides of
the building, together with window details, etc., is made only
sufficiently good to keep out the weather.
44. Material for Roof Covering.—The roof covering of this
class of building is either of slag on 2-inch spruce plank, spiked to
nailing strips bolted on to steel purlins from beneath, with lagscrews,
or of slate laid on 1-inch or 2-inch sheathing boards. Even
galvanized iron is used for the roofing of some of the cheapest class
of buildings, especially those which, owing to the process of
manufacture, are subjected to a high temperature.
45. Construction of Sides of Building.—The sides of these
buildings may be covered with either expanded-metal lath on
metallic furring strips, plastered inside and out with cement mortar
so as to form a fireproof and rigid screen wall about 2 inches in
thickness; or, the walls may be 9-inch or 13-inch brick walls built
part way up the height of the columns and leaving the columns
exposed on the face; or, corrugated galvanized iron lapped 6 inches
and secured either by riveting to metallic supports or nailed to
wooden studding secured to the steel frames. Of these
constructions, probably the first is the most expensive and also the
most satisfactory.
 Fig. 21
46. Partially Supported Steel-Frame Building.— In Fig. 21,
there is designated a type of construction that may be built for about
$1 per square foot of the area covered. This consists of steel I
beams, or angle-and-plate columns, used for column supports
carrying the usual angle iron steel-roof truss. The roof is sheathed
with 2-inch spruce tongued-and-grooved planking, covered with a
good quality of roofing felt and slag, with a stop-gutter a at the
edge. Owing to the fact that the steel columns are supported in a
direction of their minimum radius of gyration by means of the brick
walls, they can be made very light. The building illustrated has what
is known as a saw-tooth roof. By this means, light is obtained on the
side next to an adjacent and higher building by means of a sash b.
This sash is usually made hinged or pivoted, to provide the
necessary ventilation.
 47. In Fig. 22, there is illustrated, diagrammatically, the
framework of a one-story skeleton-construction building. In the
design of all such buildings, where there are no end gable walls, the
several columns and trusses must be braced diagonally, as indicated
at a, a, and frequently it is necessary to introduce a secondary
system of horizontal bracing from one panel point on the lower
chord to another, as indicated at b, b.

Fig. 22
In placing galvanized ironwork on the sides of steel-mill buildings,
it is best to construct the necessary framework between the main
supporting members of the building of light angles, or tees. These
should be furnished punched with ⅜-inch or ⁵/₁₆-inch holes, to
which the galvanized iron may be riveted, it being best to mark the
galvanized iron in the field and punch it there. This may be done
without much difficulty with the usual light gauge used for this
purpose. It is sometimes necessary with this construction to flash
around the window and door heads with IX tin.
 DETAILS OF MILL CONSTRUCTION
AND DESIGN

STRUCTURAL FEATURES

BEAM CONNECTION TO GIRDERS

48. In factory construction, the headroom is seldom available to

support beams on the girders, as indicated in Fig. 23 (a). It is
usually necessary, in order to cheapen the construction of mill
buildings, to keep the distance between the clear headroom and the
finished floor level to the very minimum, and consequently the tops
of the beams are most always brought flush, or nearly so, with the
top of the girder.
A common construction is to use some of the various forms of
wrought-iron hangers, as shown in Fig. 23 (b). The type of hanger
shown is a single stirrup, and is probably the best of any on the
market; where beams enter the girder on both sides, the hanger is
designed double. While it is popularly supposed that this hanger
would readily fail by the bending of the metal at a, it is usually
proportioned to safely carry any reaction imposed under ordinary
floor loads. This hanger is obtained stamped out of steel plate or
formed from bar iron.
(a)

(b )
 ( c)
Fig. 23
49. Where it is not desirable to use wrought-iron or steel
hangers, a simple and inexpensive form of construction may be
adopted as that shown in Fig. 23 (c). Here the beam a is supported
on a wooden strip b, which extends the full length of the girder, and
is bolted near the bottom with through bolts. Such a construction
provides sufficient strength for the support of the average factory
floor, but its strength is difficult to figure with any degree of
certainty, and some surer form of connection is generally considered
preferable. In all instances, it is good practice to tie together the
opposite floor-beams butting on a girder by means of an iron dog, or
tie-plate, c.
50. In Fig. 24 (a), (b), (c), and (d) are indicated other methods
of supporting the secondary floor-beams on main girders in the
construction of factories. In Fig. 24 (a) is shown an I-beam girder
supporting heavy timbers of a floor of slow-burning construction. It
is always necessary in this construction to bring the top edge of the
timbers above the upper flange of the I beam, and to span the
space a thus created with a piece of timber for a tie and for the
support of the floor planking. By providing this space between the
ironwork and the wooden tie, any shrinkage that may occur in the
secondary timbers will not cause the floor to ride on the top of the
steel beam and thus make a ridge evident in the finished floor at this
place. The timbers forming the secondary girders may either be
supported on angle-iron brackets, or on angle irons extending the
entire length of the girder. The latter method is only pursued when it
is necessary to keep the end of the timber a few inches away from
the steel beam, and the angle, consequently, being subjected to a
greater bending moment, must have more resistance by increasing
the width of the section of the bracket.

Fig. 24
Sometimes, the secondary beams are supported on double
stirrup hangers, as shown in Fig. 24 (b). When it is not desired to
use steel beams, resort is frequently had to flitch-plate girders. They
are, however, held in some disfavor by the building departments of
the several cities, who do not consider that the combined strength of
the timber and metal can be taken, and will only permit the strength
of either the timber or metal to be used.
51. The building departments of several of the large cities
stipulate that buildings of the second class, which includes factory
construction, shall not have steel girders that are not fireproofed
supporting brick walls or floors. When this construction is required,
the secondaries must be supported as in Fig. 24 (c). In this view is
two angle brackets riveted or bolted to the steel beam, and
extending through the concrete for the support of the wooden
beams. While there is some danger of heat being transmitted to the
beams through the projecting ends of these brackets, nevertheless it
is considered better construction than that shown in Fig. 24 (d),
where stirrups are used over the concrete fireproofing. In this latter
construction, there is a liability of the stirrup bending at a, a, and
crushing the concrete beneath. Where the reaction from the end of
the girder is great, this undoubtedly is likely to occur, and such
stirrups should be provided with a bearing plate on top of the
concrete, so that their bearing at the edge will be distributed over a
considerable area.

TRAVELING-CRANE LOADS

52. Planning for Traveling Cranes.—In designing factories or

mill buildings in which traveling cranes are to be installed, it is
important to observe that the track of the crane can be properly
supported, and also that there is sufficient headroom under the floor
or roof construction to permit the trolley of the crane and the
traveling mechanism of the crane girder to move underneath.
In Fig. 25, there is shown the upper portion of a steel-mill
building. The columns a support the girder carrying the runway of
the crane. A convenient means of supporting the roof is to splice to
this column a similar column b, which is incorporated in the design
 of the roof truss and rigidly braced
with the truss by means of a knee
brace at c. In the design of such a
building, it is very important to
determine the distances x and y
required by the makers of the
traveling crane. These distances x, y
depend on the size of the crane, that
is, whether it is designed to carry 5,
10, 15, or more tons. Usually from 9
to 12 inches is sufficient for the
measurement x, while the
measurement y varies from 5 to 8
feet.
53. Cranes Supported on
Reinforced-Concrete Walls.—
Frequently, in the latest types of
construction, the runway for the crane
is supported on reinforced-concrete
walls, which construction is shown in
Fig. 26 (a). It will be observed that
the pilasters supporting the crane are
Fig. 25 strongly reinforced in all directions
from which stresses are likely to be
created from the eccentric load
imposed by the crane track.
Where cranes are supported on reinforced-concrete columns, as
in Fig. 26 (b), it would be good practice to put additional rods in the
far side of the column as at a, in order to supply a greater resistance
to bending, and thus counteract the effect of the eccentric load
produced by the reaction from the crane track. Where cranes handle
heavy rails or cumbersome material that might, by swinging, impose
a blow on the reinforced-concrete columns, it is good construction to
protect the edge of the columns with an angle iron as indicated at b.