lEI P0WEk AN0 ENEk0¥ 5EklE5 47

Series Editors: Prof. A.T. !ohns
D.F. Warne
Protection of Electricity
Distribution Networks
2nd Edition
0tber volumes in tbis series:
volume 1 Power circuit breaker tbeory and design C.H. Flurscheim (Editor)
volume 4 lndustrial microwave beating A.C. Metaxas and R.!. Meredith
volume 7 lnsulators for bigb voltages !.S.T. Looms
volume 8 Variable frequency AC motor drive systems D. Finney
volume 10 5F6 switcbgear H.M. Ryan and C.R. !ones
volume 11 Conduction and induction beating E.!. Davies
volume 13 5tatistical tecbniques for bigb voltage engineering W. Hauschild and
W. Mosch
volume 14 Uninterruptable power supplies !. Platts and !.D. St Aubyn (Editors)
volume 15 0igital protection for power systems A.T. !ohns and S.K. Salman
volume 16 Electricity economics and planning T.W. ßerrie
volume 18 Vacuum switcbgear A. Creenwood
volume 19 Electrical safety: a guide to causes and prevention of bazards
!. Maxwell Adams
volume 21 Electricity distribution network design, 2nd edition E. Lakervi and
E.!. Holmes
volume 22 Artiñcial intelligence tecbniques in power systems K. Warwick, A.C. Ekwue
and R. Aggarwal (Editors)
volume 24 Power system commissioning and maintenance practice K. Harker
volume 25 Engineers' bandbook of industrial microwave beating R.!. Meredith
volume 26 5mall electric motors H. Moczala ETAL
volume 27 AC-0C power system analysis !. Arrill and ß.C. Smith
volume 29 Higb voltage direct current transmission, 2nd edition !. Arrillaga
volume 30 Flexible AC Iransmission 5ystems (FACI5) Y-H. Song (Editor)
volume 31 Embedded generation N. !enkins ETAL
volume 32 Higb voltage engineering and testing, 2nd edition H.M. Ryan (Editor)
volume 33 0vervoltage protection of low-voltage systems, revised edition P. Hasse
volume 34 Ibe ligbtning ñasb v. Cooray
volume 35 Control tecbniques drives and controls bandbook W. Drury (Editor)
volume 36 Voltage quality in electrical power systems !. Schlabbach ETAL
volume 37 Electrical steels for rotating macbines P. ßeckley
volume 38 Ibe electric car: development and future of battery, bybrid and fuel-cell
cars M. Westbrook
volume 39 Power systems electromagnetic transients simulation !. Arrillaga and
N. Watson
volume 40 Advances in bigb voltage engineering M. Haddad and D. Warne
volume 41 Electrical operation of electrostatic precipitators K. Parker
volume 43 Ibermal power plant simulation and control D. Flynn
volume 44 Economic evaluation of projects in tbe electricity supply industry H. Khatib
volume 45 Propulsion systems for bybrid vebicles !. Miller
volume 46 0istribution switcbgear S. Stewart
volume 47 Protection of electricity distribution networks, 2nd edition !. Cers and
E. Holmes
volume 48 Wood pole overbead lines ß. Wareing
volume 49 Electric fuses, 3rd edition A. Wright and C. Newbery
volume 50 Wind power integration: connection and system operational aspects ß. Fox
ETAL.
volume 51 5bort circuit currents !. Schlabbach
volume 52 Nuclear power !. Wood
volume 53 Condition assessment of bigb voltage insulation in power system
equipment R.E. !ames and C. Su
volume 905 Power system protection, 4 volumes
Protection of Electricity
Distribution Networks
2nd Edition
!uan M. Cers and
Edward !. Holmes
The Institution of Engineering and Technology
Published by The Institution of Engineering and Technology, London, United Kingdom
© 1998, 2004 The Institution of Electrical Engineers
First published 1998 (0 85296 923 6)
Second edition hardback 2004 (0 86341 357 9)
Second edition paperback 2005 (0 86341 537 7)
This publication is copyright under the ßerne 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. Inquiries concerning reproduction outside those terms should be sent to
the publishers at the undermentioned address:
The Institution of Engineering and Technology
Michael Faraday House
Six Hills Way, Stevenage
Herts, SC1 2AY, United Kingdom
www.theiet.org
While the authors and the publishers 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 authors nor the publishers assume any liability to anyone for any loss
or damage caused by any error or omission in the work, whether such error or omission is
the result of negligence or any other cause. Any and all such liability is disclaimed.
The moral rights of the authors to be identihed as authors of this work have been asserted
by them in accordance with the Copyright, Designs and Patents Act 1988.
8ritisb Library Cataloguing in Publication 0ata
Cers, !uan M.
Protection of electricity distribution networks - 2nd edn
1. Electric power systems - Protection 2. Electric power distribution
3. Electric power transmission
I. Title II. Holmes, E.!. (Edward !), 1928 - III. Institution of Electrical Engineers
621.3'19
l58N (10 digit) 0 86341 537 7
l58N (13 digit) 978-0-86341-537-1
Typeset in India by Newgen Imaging Systems (P) Ltd, Chennai
First printed in the UK by MPC ßooks Ltd, ßodmin, Cornwall
Contents
Preface and acknowledgements xi
Preface to 2nd edition xiii
1 Introduction 1
1.1 General 1
1.2 Basic principles of electrical systems 3
1.3 Protection requirements 3
1.4 Protection zones 5
1.5 Primary and back-up protection 5
1.5.1 Primary protection 5
1.5.2 Back-up protection 5
1.6 Directional protection 7
1.7 Exercise 1.1 9
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 14
2.1.4 Calculation of fault duty values 16
2.2 Methods for calculating short-circuit currents 19
2.2.1 Importance and construction of sequence networks 22
2.2.2 Calculation of asymmetrical faults using symmetrical
components 23
2.2.3 Equivalent impedances for a power system 26
2.3 Supplying the current and voltage signals to protection
systems 26
2.4 Calculation of faults by computer 28
3 Classification and function of relays 31
3.1 Classification 31
3.1.1 Construction 31
vi Contents
3.1.2 Incoming signal 31
3.1.3 Function 32
3.1.4 International identification of electrical devices 32
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 39
3.4.1 General 39
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 43
4 Current and voltage transformers 45
4.1 Voltage transformers 45
4.1.1 Equivalent circuit 45
4.1.2 Errors 46
4.1.3 Burden 47
4.1.4 Selection of VTs 47
4.1.5 Capacitor voltage transformers 47
4.2 Current transformers 51
4.2.1 Equivalent circuit 51
4.2.2 Errors 52
4.2.3 AC saturation 52
4.2.4 Burden 53
4.2.5 Selection of CTs 53
4.2.6 DC saturation 59
4.2.7 Precautions when working with CTs 59
5 Overcurrent protection 63
5.1 General 63
5.2 Types of overcurrent relay 63
5.2.1 Definite-current relays 63
5.2.2 Definite-time/current or definite-time relays 66
5.2.3 Inverse-time relays 66
5.3 Setting overcurrent relays 66
5.3.1 Setting instantaneous units 67
5.3.2 Coverage of instantaneous units protecting lines
between substations 68
5.3.3 Setting the parameters of time delay overcurrent
relays 70
Contents vii
5.4 Constraints of relay co-ordination 73
5.4.1 Minimum short-circuit levels 73
5.4.2 Thermal limits 73
5.4.3 Pick-up values 75
5.5 Co-ordination across Dy transformers 86
5.6 Co-ordination with fuses 96
5.7 Co-ordination of negative-sequence units 96
5.8 Overcurrent relays with voltage control 97
5.9 Setting overcurrent relays using software techniques 98
5.10 Use of digital logic in numerical relaying 99
5.10.1 General 99
5.10.2 Principles of digital logic 99
5.10.3 Logic schemes 100
5.11 Adaptive protection with group settings change 102
5.12 Exercises 104
6 Fuses, reclosers and sectionalisers 109
6.1 Equipment 109
6.1.1 Reclosers 109
6.1.2 Sectionalisers 113
6.1.3 Fuses 114
6.2 Criteria for co-ordination of time/current devices in
distribution systems 117
6.2.1 Fuse-fuse co-ordination 117
6.2.2 Recloser-fuse co-ordination 117
6.2.3 Recloser-recloser co-ordination 120
6.2.4 Recloser-relay co-ordination 122
6.2.5 Recloser-sectionaliser co-ordination 123
6.2.6 Recloser-sectionaliser-fuse co-ordination 123
7 Directional overcurrent relays 127
7.1 Construction 127
7.2 Principle of operation 128
7.3 Relay connections 128
7.3.1 30

connection (0

AMT) 129
7.3.2 60

connection (0

AMT) 129
7.3.3 90

connection (30

AMT) 130
7.3.4 90

connection (45

AMT) 131
7.4 Directional earth-fault relays 131
7.5 Co-ordination of instantaneous units 137
7.6 Setting of time-delay directional overcurrent units 141
7.6.1 Pick-up setting 141
7.6.2 Time dial setting 142
7.7 Exercises 146
viii Contents
8 Differential protection 149
8.1 General 149
8.2 Classification of differential protection 152
8.3 Transformer differential protection 152
8.3.1 Basic considerations 153
8.3.2 Selection and connection of CTs 154
8.3.3 Percentage of winding protected by the differential
relay during an earth fault 159
8.3.4 Determination of the slope 161
8.3.5 Distribution of fault current in power transformers 162
8.4 Differential protection for generators and rotating machines 164
8.5 Line differential protection 164
8.6 Busbar differential protection 168
8.6.1 Differential system with multiple restraint 168
8.6.2 High impedance differential system 169
8.7 Exercises 170
9 Distance protection 173
9.1 General 173
9.2 Types of distance relays 174
9.2.1 Impedance relay 176
9.2.2 Directional relay 179
9.2.3 Reactance relay 180
9.2.4 Mho relay 181
9.2.5 Completely polarised mho relay 182
9.2.6 Relays with lens characteristics 183
9.2.7 Relays with polygonal characteristics 183
9.2.8 Relays with combined characteristics 185
9.3 Setting the reach and operating time of distance relays 185
9.4 The effect of infeeds on distance relays 188
9.5 The effect of arc resistance on distance protection 193
9.6 Residual compensation 194
9.7 Impedances seen by distance relays 195
9.7.1 Phase units 195
9.7.2 Earth-fault units 196
9.8 Power system oscillations 196
9.9 The effective cover of distance relays 199
9.10 Maximum load check 200
9.11 Drawing relay settings 203
9.12 Intertripping schemes 212
9.12.1 Under reach with direct tripping 213
9.12.2 Permissive under reach intertripping 213
9.12.3 Permissive over reach intertripping 214
9.13 Distance relays on series-compensated lines 214
Contents ix
9.14 Technical considerations of distance protection in tee circuits 215
9.14.1 Tee connection with infeeds at two terminals 215
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 222
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
10.2.2 Low voltage breakers 227
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 240
11.2.1 Simple machine model 240
11.2.2 Parameters for implementing a load shedding
system 241
11.3 Criteria for setting frequency relays 242
11.3.1 Operating times 242
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 243
11.4.4 Load shedding stages 243
11.4.5 Determination of the frequency relay settings 244
11.4.6 Verification of operation 247
12 Protection schemes and substation design diagrams 251
12.1 Protection schemes 251
12.1.1 Generator protection 251
12.1.2 Motor protection 252
12.1.3 Transformer protection 258
12.1.4 Line protection 261
12.2 Substation design diagrams 261
12.2.1 Single-line diagrams 262
12.2.2 Substation layout diagrams 263
x Contents
12.2.3 Diagrams of AC connections 264
12.2.4 Diagrams of DC connections 265
12.2.5 Wiring diagrams 266
12.2.6 Logic diagrams 268
12.2.7 Cabling lists 268
13 Processing alarms 269
13.1 General 269
13.2 Alarm processing methods 270
13.3 Expert systems 271
13.4 Equivalent alarms 272
13.5 Rules 273
13.6 Finger printing approach 273
13.7 Hypothesis approach 275
14 Installation, testing and maintenance of protection systems 283
14.1 Installation of protection equipment 283
14.2 Testing protection schemes 285
14.2.1 Factory tests 285
14.2.2 Precommissioning tests 285
14.2.3 Periodic maintenance 290
14.3 Commissioning numerical protection 292
14.3.1 Setting the parameters 292
14.3.2 Performance tests 293
Bibliography 297
Appendix: Solutions to exercises 301
Index 339
Preface and acknowledgements
The quality of electricity supplies is an important factor in the socio-economic
development of any area. Approximately 75 per cent 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
network design using proven equipment, it is also essential to provide suitable
protection schemes and relay settings to ensure that faults are quickly disconnected
to minimise outage times and improve the continuity of supplies to customers.
With this in mind 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 fromactual case studies in the field,
will provide worthwhile material for planning and design engineers and maintenance
staff, particularly those engaged in the co-ordination and setting of protection on
distribution systems.
The book is based on original material by Mr Gers and translated fromthe Spanish
by Mr Holmes. Subsequently the authors have expanded the text considerably 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 Ltd. 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 which 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
Preface to 2nd 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 nowmultifunctional
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 which allow them to work in
virtually any automated scheme. Therefore, modern relays nowoffer better protection
coverage and can be programmed to automatically adjust for changes in power system
topologies and different operating conditions owing 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 newtechnology available, while the testing procedures covered in
the last chapter nowinclude ample reference to numerical protection. The last chapter
was also thoroughly updated, particularly in respect of testing procedures applicable
to numerical relays.
We have also taken the opportunity to update sections of the original text and
have added a new chapter on the processing of alarms since the fast and efficient
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. 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
jmgers@gersusa.com ejholmes@compuserve.com
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 fur-
ther requirement is the safety of the electricity supply. Apriority 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 sup-
ply. The most important of these are the protection systems which are installed to clear
faults and limit any damage to distribution equipment. Amongst 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. In order 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 co-ordinated protection system is vital to ensure that an electricity
distribution network can operate within preset requirements for safety for individual
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 min-
imise 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
nondistributed 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
2 Protection of electricity distribution networks
Transformer
Generator
Disconnector
Circuit breaker
Fuse
Feeder Customer
rural
distribution
to adjacent
MV
network
LV board
LV network
urban distribution
MV/LV
MV cable
network
MV
overhead
network
HV/MV
MV
HV network
HV/MV
EHV/HV
EHV
Major HV
substation
Figure 1.1 EHV/HV/MV/LV network arrangements (reproduced from Electricity
Distribution Network Design)
Introduction 3
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 equip-
ment must be restricted to the minimumamount necessary to isolate the fault fromthe
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 maximumrated 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 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 primary transmission, is required. Such
systems operate in the 300 kV plus 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 either as 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 net-
works are supplied from EHV/HV substations which 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 sup-
plies direct to large customers, but the vast majority of customers are connected at
low voltage and supplied via MV/LV distribution substations and their associated
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
4 Protection of electricity distribution networks
Generation
Generation
Consumption
HV/MV
MV/LV
EHV
EHV/HV
HV
MV
LV
Consumption
Figure 1.2 Block schematic of transmission and distribution systems (reproduced
from Electricity Distribution Network Design)
a fault, and security, which is the ability to avoid incorrect operation during
faults.
2. Speed: minimum operating time to clear a fault in order to avoid damage to
equipment.
Introduction 5
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 simul-
taneously, inevitably a compromise is required to obtain the optimum protection
system.
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 systemto 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 protectionrelays tothe appropriate current transformers as illustratedinFigure 1.4.
1.5 Primary and back-up protection
All the elements of the power systemmust be correctly protected so that the relays only
operate on the occurrence of a fault. Some relays, designated as unit type protection,
operate onlyfor faults withintheir protectionzone. Other relays designatedas non-unit
protection, are able to detect faults both within a particular zone and also outside 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 protection 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 sensing
6 Protection of electricity distribution networks
M
~
M
~
Figure 1.3 Protection zones
Busbar protection
Line protection
Figure 1.4 Overlap of protection zones
Introduction 7
element which may or may not be similar to the primary protection, but which also
includes a time-delayfacilitytoslowdownthe operationof the relaysoas toallowtime
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.
1.6 Directional protection
An important characteristic of some types of protection is their capacity to be able to
determine the direction of the flowof 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 switchgear and thus improves the
security of the electricity supply. On protection schematic 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 Check on correct operation of protection
Using the power systemshown 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
A
2
G1
3 1
G2
F
1
F
2
F
3
F
4
4 6
B
G3
5
7 8 11
9
G4
10
Figure 1.5 Power system for Example 1.1
8 Protection of electricity distribution networks
of correct relay operation, with a final example of unnecessary relay operation. The
directional protection is indicated by the arrows below the corresponding breakers.
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.
Table 1.1 Relay/breaker operations for Example 1.1
Case Breakers
that
operated
Breakers
that
mal-operated
Tripped by
primary
protection
Tripped by
back-up
protection
F
1
1, 2, 4 3 4 1, 2
F
2
3, 5, 8 6 – 3, 5, 8
F
3
10 – 10 –
F
4
8, 11 8 11 –
G3
19
Tabor 115kV
17 18
F
3
16
Pailon
15
25
Diesel 34.5 kV
23 24
F
2
San Luis
21 22
Chipichape
20
26
27
Termoyumbo
Bajo Anchicayá
115 kV
11 12
10
13 14 7
F
1 5
G2 2
San Luis Chipichape
8
9
6
28
29
3 4
1
G1
115 kV
115 kV 115 kV
115kV 34,5 kV 34.5 kV
Figure 1.6 Schematic diagram for Exercise 1.1
Introduction 9
For fault F
1
, 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 G1
and G2. With fault F
2
, when breaker 6 failed to operate, the directional protection 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 F
3
was correctly cleared by the tripping of feeder breaker 10. Fault F
4
was
correctly cleared by the operation of breaker 11, so that the tripping of breaker 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
mal-operation from a dependability point of view, with the last one illustrating mal-
operation from a security standpoint.
1.7 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
that
operated
Breakers
that
mal-operated
Tripped by
primary
protection
Tripped by
back-up
protection
F
1
2, 3, 4, 5 2, 5
F
2
21, 22, 23
24, 27
F
3
10, 11, 17, 19
Chapter 2
Calculation of short-circuit currents
The current that flows through an element of a power systemis a parameter which 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 calculations illustrating
the methods used. While 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 applications, 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, which
produces a decaying DC component, and the performance of the rotating machinery,
which results in a decaying AC component.
2.1.1 Effect of the system impedance
Systemcurrents cannot change instantaneously when a fault occurs due to the equiva-
lent systemresistances 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. In order to illustrate the transient nature of the current, consider an RL
circuit as a simplified equivalent of the circuits in electricity distribution networks.
This simplification is important because all the system equipment must be modelled
12 Protection of electricity distribution networks
R V
max
sin (t +)
L
Figure 2.1 RL circuit for transient analysis study
in some way in order to quantify the transient values that can occur during the fault
condition.
For the circuit shown in Figure 2.1, the mathematical expression that defines the
behaviour of the current is
e(t ) =L
di
dt
+Ri(t ) (2.1)
This is a differential equation with constant coefficients, of which the solution is in
two parts:
i
a
(t ) : i
h
(t ) +i
p
(t )
where i
h
(t ) is the solution of the homogeneous equation corresponding to the transient
period and i
p
(t ) is 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:
i(t ) =
V
max
Z
¸
sin(ωt +α −φ) −sin(α −φ)e
−(R/L)t
¸
(2.2)
where
Z=

(R
2

2
L
2
)
α =the closing angle, which defines the point on the source sinusoidal voltage
when the fault occurs, and
φ =tan
−1
(ωL/R)
It can be seen that, in eqn. 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, while the second term can be recognised as the DC component
of the current having an initial maximum value when α −φ =±π/2, and zero value
when α =φ; 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
Calculation of short-circuit currents 13
(a)
(b)
Figure 2.2 Variation of fault current with time: (a) α −φ =0; (b) α −φ =−π/2
reach. If the tripping of the circuit, due to a fault, takes place when the sinusoidal
component is at its negative peak, the DCcomponent reaches its theoretical maximum
value half a cycle later.
An approximate formula for calculating the effective value of the total asymmetric
current, including the AC and DC components, with acceptable accuracy can be used
by assuming that these components are in quadrature with the following expression:
I
rms.asym
=

I
2
rms
+I
2
DC
(2.3)
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 induction current, can be
seen in Figure 2.3. This effect is known as armature reaction.
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 will be noted that the variation of current with time, I(t ), comes
close to the three discrete levels of current, I

, I

and I, the sub-transient, transient and
steady state currents respectively. The corresponding values of direct axis reactance
14 Protection of electricity distribution networks
ia
ib
ic
(a)
(b)
(c)
Figure 2.3 Transient short-circuit currents in a synchronous generator: (a) typical
A phase short circuit current; (b) typical B phase short circuit current;
(c) typical C phase short circuit current
are denoted by X

d
, X

d
and X
d
, and the typical variation with time for each of these
is illustrated in Figure 2.5.
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.
Calculation of short-circuit currents 15
I Љ

I
I (t)
t
1
t
2
Figure 2.4 Variation of current with time during a fault
X(t)
t
1
t
2
X
d
X
d
Ј
X
d
Љ
Figure 2.5 Variation of generator reactance with time during a fault
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
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 indicated
16 Protection of electricity distribution networks
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 currents. They are
calculated with DCoffset but no ACdecrement in the sources, and using the machine
sub-transient reactances. Peak currents correspond to the maximum currents during
the first cycle after the fault occurs and differ from the first cycle currents that are
totally asymmetrical r.m.s. currents.
Interrupting currents, also known as contact parting currents, are the values that
have to be cleared by interrupting equipment. In European Standards, these val-
ues are called breaking currents and typically are calculated in the range from 3 to
5 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 syn-
chronous 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 IEEEStandard 399-1990. For each case, asymmetrical or symmetrical r.m.s.
values can be defined depending on whether the DC component is included or not.
The peak values are obtained by multiplying the r.m.s. values by

2.
The asymmetrical values are calculated as the square root of the sumof the squares
of the DC component and the r.m.s. value of the AC current, i.e.
I
rms
=

I
2
DC
+I
2
AC
(2.4)
2.1.4 Calculation of fault duty values
The momentary current is used when specifying the closing current of switchgear.
Typically, the ACand DCcomponents decay to 90 per cent of their initial values after
the first half cycle. From this, the value of the r.m.s. current would then be
I
rms.asym.closing
=

I
2
DC
+I
2
AC.rms.sym.
=

(0.9

2V/X

d
)
2
+(0.9V/X

d
)
2
=1.56V/X

d
=1.56I
rms.sym.
(2.5)
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.
The peak value is obtained by arithmetically adding together the AC and DC
components. It should be noted that, in this case, the AC component is multiplied by
a factor of

2. Thus
I
peak
=I
DC
+I
AC
=(0.9

2V/X

d
) +(0.9

2V/X

d
)
=2.55I
rms.sym.
(2.6)
Calculation of short-circuit currents 17
Duty calculation
(see Note 1)
System component Reactance value
for medium- and
high-voltage
calculations per
IEEE Std
C37.010-1979
and IEEE Std
C37.5-1979
Reactance value
for low-voltage
calculations
(see Note 2)
First cycle Power company supply X
s
X
s
(momentary
calculations)
All turbine generators; all hydro-
generators with amortisseur wind-
ings; all condensers
1.0X

d
1.0X

d
Hydrogenerators without amortis-
seur windings
0.75X

d
0.75X

d
All synchronous motors 1.0X

d
1.0X

d
Induction motors
Above 1000 hp 1.0X

d
1.0X

d
Above 250 hp at 3600 rev/min 1.0X

d
1.0X

d
All others, 50 hp and above 1.2X

d
1.2X

d
All smaller than 50 hp 1.67X

d
1.67X

d
(see Note 6)
Interrupting Power company supply X
s
N/A
calculations All turbine generators; all hydro-
generators with amortisseur wind-
ings; all condensers
1.0X

d
N/A
Hydrogenerators without amortis-
seur windings
0.75X

d
N/A
All synchronous motors 1.5X

d
N/A
Induction motors
Above 1000 hp 1.5X

d
N/A
Above 250 hp at 3600 rev/min 1.5X

d
N/A
All others, 50 hp and above 3.0X

d
N/A
All smaller than 50 hp Neglect N/A
Notes:
1: 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.
2: Reactance (X) values to be used for low-voltage breaker duty calculations (see IEEEStd C37.13-
1990 and IEEE Std 242-1986).
3: X

d
of synchronous-rotating machines is the rated-voltage (saturated) direct-axis subtransient
reactance.
4: X

d
of synchronous-rotating machines is the rated-voltage (saturated) direct-axis transient
reactance.
5: X

d
of induction motors equals 1 divided by per-unit locked-rotor current at rated voltage.
6: 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)
18 Protection of electricity distribution networks
When considering the specification for the switchgear opening current, the so-
called r.m.s. value of interrupting current is used in which, again, the AC and DC
components are taken into account, and therefore
I
rms.asym.int.
=

I
2
DC
+I
2
AC.rms.sym.int.
Replacing the DC component by its exponential expression gives
I
rms.asym.int.
=


2I
rms.sym.int.
e
−(R/L)t

2
+I
2
rms.sym.int.
=I
rms.sym.int.

2e
−2(R/L)t
+1 (2.7)
The expression (I
rms.asym.int.
/I
rms.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.
10
20
30
40
50
60
70
80
90
100
110
120
130
140
150
0
1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7
Multiplying factors
R
a
t
i
o
X
/
R
1 2 3 4
C
o
n
t
a
c
t
-
p
a
r
t
i
n
g
t
i
m
e
,
c
y
c
l
e
s
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)
Calculation of short-circuit currents 19
As an illustration of the validity of the curves for any situation, consider a cir-
cuit 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 , is 60 Hz and the ratio X/R is given as 50, with t =2 cycles =0.033 s,
then (X/R) =(ωL/R) =50. Thus (L/R) =(50/ω) = (50/2πf ) =0.132. Therefore
I
asym
I
sym
=

2 e
−(0.033×2)/0.132
+1=1.49
as can be seen from Figure 2.7.
In protection co-ordination studies, r.m.s. 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 there-
fore can be treated by using a single-phase representation. This symmetry is lost
during asymmetric 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 which were named the symmetrical components of
the original phases. While 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
represents the values of the three phases. The three system balanced phasors are
designated as:
1. Positive-sequence components, which consist of three phasors of equal magni-
tude, 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 magni-
tude, 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.
20 Protection of electricity distribution networks
With this arrangement, voltage values of any three-phase system, V
a
, V
b
, and V
c
,
can be represented thus:
V
a
=V
a0
+V
a1
+V
a2
V
b
=V
b0
+V
b1
+V
b2
V
c
=V
c0
+V
c1
+V
c2
It can be demonstrated that
V
b
=V
a0
+a
2
V
a1
+aV
a2
V
c
=V
a0
+aV
a1
+a
2
V
a2
where a is a so-called operator which gives a phase shift of 120

clockwise and a
multiplication of unit magnitude i.e. a =1

120

, and a
2
similarly gives a phase shift
of 240

, i.e. a
2
=1

240

.
Therefore, the following matrix relationship can be established:


V
a
V
b
V
c


=


1 1 1
1 a
2
a
1 a a
2


×


V
a0
V
a1
V
a2


Inverting the matrix of coefficients


V
a0
V
a1
V
a2


=
1
3


1 1 1
1 a a
2
1 a
2
a


×


V
a
V
b
V
c


From the above matrix it can be deduced that
V
a0
=1/3(V
a
+V
b
+V
c
)
V
a1
=1/3(V
a
+aV
b
+a
2
V
c
)
V
a2
=1/3(V
a
+a
2
V
b
+aV
c
)
The foregoing procedure can also be applied directly to currents, and gives
I
a
=I
a0
+I
a1
+I
a2
I
b
=I
a0
+a
2
I
a1
+aI
a2
I
c
=I
a0
+aI
a1
+a
2
I
a2
Therefore
I
a0
=1/3(I
a
+I
b
+I
c
)
I
a1
=1/3(I
a
+aI
b
+a
2
I
c
)
I
a2
=1/3(I
a
+a
2
I
b
+aI
c
)
In three-phase systems the neutral current is equal to I
n
=(I
a
+I
b
+I
c
) and,
therefore, I
n
=3I
a0
. By way of illustration, a three-phase unbalanced systemis shown
in Figure 2.8 together with the associated symmetrical components.
Calculation of short-circuit currents 21
V
c
V
a
V
b
V
c1
V
a1
V
b1
V
a
=8.0
V
c2
V
a2
V
b2
V
a0
V
b0
V
c0
V
a0
V
a
V
a2
V
a1
V
b1
V
c
V
c1
V
c2
V
c0
V
b
V
b2
V
b0

V
a1
=9.8 18.4°
V
b1
=9.8 –101.6°
V
c1
=9.8 138.4°
V
a2
=4.3 –86.2°
V
b2
=4.3 33.8°
V
c2
=4.3 –206.2°
V
a0
=2.0 143.1°
V
b0
=2.0 143.1°
143.1°
V
c0
=2.0
V
b
=6.0 –90° V
c
=16.0 143.1°
Figure 2.8 Symmetrical components of an unbalanced three-phase system
22 Protection of electricity distribution networks
2.2.1 Importance and construction of sequence networks
The impedance of a circuit in which only positive-sequence currents are circulat-
ing is called the positive-sequence impedance and, similarly, those in which only
negative- and zero-sequence currents floware called the negative- and zero-sequence
impedances. These sequence impedances are designated Z
1
, Z
2
and Z
0
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 net-
works 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, Z
0
/Z
1
=2 when no earth wire is present and 3.5 with an earth wire. For a
double circuit line Z
0
/Z
1
=5.5. For underground cables Z
0
/Z
1
can be taken as 1 to
1.25 for single core, and 3 to 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, pro-
vided 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.
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

d
, X

d
and X
d
. In
fault studies the subtransient and transient reactances of generators and motors must
be included as appropriate, depending on the machine characteristics and fault clear-
ance 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 tran-
sient 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.
In connecting sequence networks together, the reference busbar for the positive-
andnegative-sequence networks is the generator neutral which, inthese networks, is at
earth potential so only zero-sequence currents flow through the impedances between
neutral and earth. The reference busbar for zero-sequence networks is the earth point
of the generator. The current that flows in the impedance Z
n
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, which has an impedance of Z
0
=3Z
n
+Z
e0
.
Calculation of short-circuit currents 23
Table 2.1 Typical per-unit reactances for three-phase synchronous machines
Type of machine X

d
X

d
X
d
X
2
X
0
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
Salient pole with dampers 0.20 0.30 1.25 0.20 0.18
Generator without dampers 0.28 0.30 1.20 0.35 0.12
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 a phase 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
=E
a
−I
a1
Z
1
V
a2
=−I
a2
Z
2
V
a0
=−I
a0
Z
0
where: E
a
=no load voltage to earth of the positive-sequence network; Z
1
=positive-
sequence impedance of the generator; Z
2
=negative-sequence impedance of the
generator; Z
0
=zero-sequence impedance of the generator (Z
g0
) plus three times
the impedance to earth.
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 to loaded generators, E
a
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 I
1
, I
2
and I
0
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.
Line-to-earth fault
The conditions for a solid fault from line a to earth are represented by the equations
I
b
=0, I
c
=0 and V
a
=0. As in the previous equations, it can easily be deduced
that I
a1
=I
a2
=I
a0
=E
a
/(Z
1
+Z
2
+Z
0
). Therefore, the sequence networks will be
connected 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.
24 Protection of electricity distribution networks
c b
a
Z
2
Z
2
Z
2
I
a2
I
b2
I
c2
Reference bus
Z
2
I
a2
a
V
a2
Reference bus
I
a0
a
V
a0
3Z
n
Z
e0
Z
0
c b
a
I
a0
I
b0
I
c0
Z
n
I
a0
I
a0
I
a0
I
b0
=
I
c0
= Z
e0
Z
e0
Z
e0
Z
1
+

Reference bus
a
I
a1
E
a
V
a1

c
+
Z
1
b

+
Z
1
Z
1
+
a
I
a1
I
b1
I
c1
E
b
E
a
E
c

(a)
(c)
(e)
(b)
(d)
(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
Calculation of short-circuit currents 25
Z
1
Z
1
Z
1
Z
2
Z
0
Z
2
Z
2
Z
0
E
a
E
a
E
a
I
a1
I
a1
I
a1
I
a2
I
a0
I
a2
V
a1
V
a1
V
a1
V
a2
V
a0
V
a2
V
a2
V
a0
(a)
(b)
(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
Line-to-line fault
The conditions for a solid fault between lines b and c are represented by the equations
I
a
=0, I
b
=−I
c
and V
b
=V
c
. Equally it can be shown that I
a0
=0 and I
a1
=E
a
/(Z
1
+
Z
2
) =−I
a2
. 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.
26 Protection of electricity distribution networks
Line-to-line-to-earth fault
The conditions for a fault between lines b and c and earth are represented by the
equations I
a
=0 and V
b
=V
c
=0. From these equations it can be proved that:
I
a1
=
E
a
Z
1
+(Z
0
Z
2
)/(Z
0
+Z
2
)
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.
The equivalent positive- and negative-sequence impedances can be calculated
directly from:
Z=V
2
/P
where: Z=equivalent positive- and negative-sequence impedances; V =nominal
phase-to-phase voltage; P =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., I
a1
=I
a2
=I
a0
=
V
LN
/(Z
1
+Z
2
+Z
0
), where V
LN
=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 distribu-
tion network fault studies, it may be assumed that overall Z
2
=Z
1
, which simplifies
the calculations. Thus, the above formula reduces to I
a
=3I
a0
=3V
LN
/(2Z
1
+Z
0
),
where V
LN
= line-to-neutral voltage and Z
0
=(3V
LN
/I
a
) −2Z
1
.
2.3 Supplying the current and voltage signals to protection systems
In the presence of a fault the current transformers (CTs) circulate current proportional
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 symmetrical 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
Calculation of short-circuit currents 27
c
,

a
F
a
u
l
t
P
o
s
i
t
i
v
e
-

s
e
q
u
e
n
c
e
n
e
t
w
o
r
k
c
u
r
r
e
n
t
N
e
g
a
t
i
v
e
-
s
e
q
u
e
n
c
e
n
e
t
w
o
r
k
c
u
r
r
e
n
t
a
,

b
,

c
a
,

b
b
,

c
Z
e
r
o
-
s
e
q
u
e
n
c
e
n
e
t
w
o
r
k
c
u
r
r
e
n
t
c
,

a
,

e
b
,

c
,

e
a
,

b
,

e
a
,

e
b
,

e
c
,

e
a
0
,

b
0
,

c
0
a
0
,

b
0
,

c
0
a
0
,

b
0
,

c
0
a
0
,

b
0
,

c
0
a
0
,

b
0
,

c
0
a
0
,

b
0
,

c
0
F
a
u
l
t
c
u
r
r
e
n
t
s
c
1
b
1
a
1
c
1
b
1
a
1
c
1
b
1
a
1
c
1
b
1
a
1
c
1
b
1
a
1
c
1
b
1
a
1
c
1
b
1
a
1
c
1
b
1
a
1
c
1
b
1
a
1
c
1
b
1
a
1
a
2
c
2
b
2
c
2
b
2
a
2
c
2
b
2
a
2
b
2
a
2
c
2
b
2
a
2
c
2
b
2
a
2
c
2
a
2
c
2
b
2
c
2
b
2
a
2
a
2
c
2
b
2
c
=
0
a
=
0
b
=
0
c
=
0
a
=
0
b
a
c
a
b
c
b
c
a
a
b
a a
b
=
0
c
c
b
a
=
c
=
0
b
=
c
=
0
b
a
=
b
=
0
c
c
,

a
F
a
u
l
t
P
o
s
i
t
i
v
e
-
s
e
q
u
e
n
c
e
n
e
t
w
o
r
k
v
o
l
t
a
g
e
N
e
g
a
t
i
v
e
-
s
e
q
u
e
n
c
e
n
e
t
w
o
r
k
v
o
l
t
a
g
e
a
,

b
,

c
a
,

b
b
,

c
Z
e
r
o
-
s
e
q
u
e
n
c
e
n
e
t
w
o
r
k
v
o
l
t
a
g
e
c
,

a
,

e
b
,

c
,

e
a
,

b
,

e
a
,

e
b
,

e
c
,

e
a
0
,

b
0
,

c
0
a
0
,

b
0
,

c
0
a
0
,

b
0
,

c
0
a
0
,

b
0
,

c
0
a
0
,

b
0
,

c
0
a
0
,

b
0
,

c
0
F
a
u
l
t
v
o
l
t
a
g
e
s
c
1
b
1
a
1
c
1
b
1
a
1
c
1
b
1
a
1
c
1
b
1
a
1
c
1
b
1
a
1
c
1
b
1
a
1
c
1
b
1
a
1
c
1
b
1
a
1
c
1
b
1
a
1
c
1
b
1
a
1
a
2
c
2
b
2
c
2
b
2
a
2
c
2
b
2
a
2
b
2
a
2
c
2
b
2
a
2
c
2
b
2
a
2
c
2
a
2
c
2
b
2
c
2
b
2
a
2
a
2
c
2
b
2
a
=
b
b
=
c
a
=
c
a
a
=
b
=
0
b
=
c
=
0
b
c
c
a
a
b
Z
e
r
o

a
1
f
a
u
l
t
a
a
c
b
a
b
b
c c a
=
c
=
0
a
=
0 b
=
0
c
=
0
(
a
)
(
b
)
F
i
g
u
r
e
2
.
1
1
C
u
r
r
e
n
t
s
a
n
d
v
o
l
t
a
g
e
s
f
o
r
v
a
r
i
o
u
s
t
y
p
e
s
o
f
f
a
u
l
t
s
:
(
a
)
s
e
q
u
e
n
c
e
c
u
r
r
e
n
t
s
f
o
r
d
i
f
f
e
r
e
n
t
t
y
p
e
s
o
f
f
a
u
l
t
;
(
b
)
s
e
q
u
e
n
c
e
v
o
l
t
a
g
e
s
f
o
r
d
i
f
f
e
r
e
n
t
t
y
p
e
s
o
f
f
a
u
l
t
28 Protection of electricity distribution networks
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 voltages and for currents.
Although these filters can be constructed for electromagnetic elements, the growth of
electronics has ledtotheir beingusedincreasinglyinlogic circuits. Amongst the relays
that require this type of filter in order to operate are those used in negative-sequence
and earth-fault protection.
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 com-
ponents; 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, both in 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 programs
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 calcula-
tions considerably bearing in mind that, as well as 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, etc.
Modern software packages enable the following features to be carried out:
• Calculations not only for the standard fault types, i.e. three-phase, line-to-line,
line-to-line-to-earth, line-to-earth, but also for faults such as those between
systems of different voltages;
Calculation of short-circuit currents 29
• 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, which can be slightly different especially in the pre-fault voltage level;
• Calculate simultaneous faults;
• Calculate faults along different line lengths.
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, generally 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-electric (thermal, pressure, etc.)
3.1.2 Incoming signal
• current
• voltage
• power
• frequency
• temperature
• pressure
• speed
• others
32 Protection of electricity distribution networks
3.1.3 Function
• overcurrent
• directional overcurrent
• distance
• overvoltage
• differential
• reverse power
• others
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 under-current or under-power 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
Classification and function of relays 33
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, T for transformer, etc.
Non-electric 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 components.
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 relay 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 which 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
Coil
Pivot
Armature
Figure 3.1 Armature-type relay
34 Protection of electricity distribution networks
Coil
Piston
Fixed contact
Moving contact
Restraining
spring
Figure 3.2 Solenoid-type relay
It can be shown that the force of attraction is equal to K
1
I
2
−K
2
, where K
1
depends upon the number of turns on the operating solenoid, the air gap, the effective
area and the reluctance of the magnetic circuit, amongst other factors. K
2
is the
restraining force, usually produced by a spring. When the relay is balanced, the
resultant force is zero and therefore K
1
I
2
=K
2
, so that I =

(K
1
/K
2
) =constant.
In order to control the value at which the relay starts to operate, the restraining tension
of the spring or the resistance of the solenoid circuit can be varied, thus modifying
the restricting force. Attraction relays effectively have no time delay and, for that
reason, are widely used when instantaneous operation is required.
3.2.2 Relays with moveable coils
This type of relay consists of a rotating movement with a small coil suspended or
pivoted with the freedom to rotate between the poles of a permanent magnet. The
coil is restrained by two springs which also serve as connections to carry the current
to the coil.
The torque produced in the coil is given by
T =BlaNi
where: T =torque; B =flux density; l =length of the coil; a =diameter of the coil;
N = number of turns on the coil; i =current flowing through the coil.
From the above equation it will be noted that the torque developed is proportional
to the current. The speed of movement is controlled by the damping action which is
Classification and function of relays 35
Multiples of tap value current
T
i
m
e

i
n

s
e
c
o
n
d
s
Figure 3.3 Inverse time characteristic
proportional to the torque. It thus follows that the relay has an inverse time charac-
teristic similar to that illustrated in Figure 3.3. The relay can be designed so that the
coil makes a large angular movement, for example 80

.
3.2.3 Induction relays
An induction relay works only with alternating current. It consists of an electromag-
netic system which operates on a moving conductor, generally in the form of a disc
or cup, and functions through the interaction of electromagnetic fluxes with the par-
asitic Foucalt currents that are induced in the rotor by these fluxes. These two fluxes,
which are mutually displaced both in angle and in position, produce a torque that
can be expressed by T =K
1

1

2
sin , where
1
and
2
are the interacting fluxes
and is the phase angle between
1
and
2
. It should be noted that the torque
is a maximum when the fluxes are out of phase by 90

and zero when they are in
phase.
It can be shown that
1
=
1
sin ωt , and
2
=
2
sin(ωt +), where is the
angle by which
2
leads
1
.
Then,
i

1

d
1
dt

1
cos ωt
and
i

2

d
2
dt

2
cos(ωt +)
36 Protection of electricity distribution networks
F
2
F
1
i
Φ
2
i
Φ
2
i
Φ
2
i
Φ
1
i
Φ
1
i
Φ
1
Φ
2
Φ
1
Figure 3.4 Electromagnetic forces in induction relays
Figure 3.4 shows the interrelationship between the currents and the opposing
forces.
Thus:
F =(F
2
−F
1
) ∝(
2
i

1

1
i

2
)
F ∝
2
sin(ωt +)
1
cos ωt −
1
sin ωt
2
cos(ωt +)
F ∝
1

2
[sin(ωt +) cos ωt −sin ωt cos(ωt +)]
F ∝
1

2
[sin{(ωt +) −ωt }]
F ∝
1

2
sin ∝T
Induction relays can be grouped into three classes as set out below.
(i) Shaded pole relay
In this case a portion of the electromagnetic section is short-circuited by means of
a copper ring or coil. This creates a flux in the area influenced by the short-circuited
section (the so-called shaded section) which lags the flux in the non-shaded section
(see Figure 3.5).
(ii) Wattmetric type relay
In its more common form, this type of relay uses an arrangement of coils above and
below the disc with the upper and lower coils fed by different values or, in some
cases, with just one supply for the top coil, which induces an out-of-phase flux in the
lower coil because of the air gap. Figure 3.6 illustrates a typical arrangement.
(iii) Cup type relay
This type of relay has a cylinder similar to a cup which can rotate in the annular
air gap between the poles of the coils, and has a fixed central core (see Figure 3.7).
Classification and function of relays 37
Core
Spindle
Moving contact
Fixed contact
Coil
Shading ring
Disc
Figure 3.5 Shaded-pole relay
Lower coils
Disc
Upper coil
Core
Figure 3.6 Wattmetric-type relay
The operation of this relay is very similar to that of an induction motor with salient
poles for the windings of the stator. Configurations with four or eight poles spaced
symmetrically around the circumference of the cup are often used. The movement
of the cylinder is limited to a small amount by the contact and the stops. A special
spring provides the restraining torque. The torque is a function of the product of
38 Protection of electricity distribution networks
Central core
8 pole stator
Air gap
Lateral coil
Corner
coil
Cup
Figure 3.7 Cup-type relay
the two currents through the coils and the cosine of the angle between them. The
torque equation is T =[KI
1
I
2
cos(
12
−) −K
s
], where K, K
s
, and are design
constants, I
1
andI
2
are the currents throughthe twocoils, and
12
is the angle between
I
1
and I
2
.
In the first two types of relay mentioned above, which are provided with a disc,
the inertia of the disc provides the time delay characteristic. The time delay can be
increased by the addition of a permanent magnet. The cup type relay has a small
inertia and is therefore principally used when high-speed operation is required, for
example in instantaneous units.
3.3 Evolution of protection relays
The evolution of protection relays started with the attraction type of relay referred to
earlier. However, the design of protection relays has changed significantly over the
past years with the advancement in microprocessor and signal processing technology.
As electronic technology progressed, electromechanical relays were superseded in the
1960s by electronic or static designs using transistors and similar types of electronic
elements. The integrated circuit (IC) enabled static designs to be further extended
and improved in the 1970s. Following the development of the microprocessor, basic
programmable microprocessor or micro-controlled multifunction protection relays
first started to appear in the early 1980s. Subsequently in the 1990s microprocessor
technology, along with the improvements in mathematical algorithms, spurred the
development of the so-called numerical relays which are extremely popular for their
multifunctional capabilities, low prices and reliability.
Classification and function of relays 39
3.4 Numerical protection
3.4.1 General
Numerical protection relays operate on the basis of sampling inputs and controlling
outputs to protect or control the monitored system. System currents and/or voltages,
for example, are not monitored on a continuous basis but, like all other quantities,
are sampled one at a time. After acquiring samples of the input waveforms, calcula-
tions are performed to convert the incremental sampled values into a final value that
represents the associated input quantity based on a defined algorithm. Once the final
value of an input quantity can be established, the appropriate comparison to a setting,
or reference value, or some other action, can be taken as necessary by the protection
relay. Depending upon the algorithm used, and other system design or protection
requirements, the final value may be calculated many times within a single sampling
cycle, or only once over many cycles.
Most numerical relays are multifunctional and can be regarded as intelligent elec-
tronic devices (IEDs). The proper handling of all the features requires a flexible
programmable logic platform for the user to apply the available functions with com-
plete flexibility and be able to customise the protection to meet the requirements of
the protected power system. Programmable I/O, extensive communication features
and an advanced human-machine interface (HMI), which is normally built into most
relays, provide easy access to the available features.
3.4.2 Characteristics of numerical relays
Numerical relays are technically superior to the conventional types of relays described
earlier in this chapter. Their general characteristics are:
• Reliability: incorrect operations are less likely with numerical relays.
• Self-diagnosis: numerical relays have the ability to conduct continuous self-
diagnosis in the form of watchdog circuitry, which includes memory checks and
analogue input module tests. In case of failure, normally the relays either lock
out or attempt a recovery, depending on the disturbance detected.
• Event and disturbance records: these relays can produce records of events when-
ever there is a protection function operation, the energising of a status input,
or any hardware failure. In addition, disturbance records can be generated in
a number of analogue channels, together with all the status input and relay output
information.
• Integration of digital systems: the present technology now includes many other
tasks at one substation, such as communications, measurement and control. These
functions can be integrated into one digital system so that a substation can be
operated in a more rapid and reliable manner. Fibre optics are now being used
to provide communication links between various system elements to avoid the
interference problems that can occur when using metallic conductors.
• Adaptive protection: with the programming and communication capacity of digi-
tal systems, the numerical relay can provide adaptive protection. This feature
40 Protection of electricity distribution networks
enables the relay setting to be changed depending on the operating conditions of
the network, thus guaranteeing suitable relay settings for the real-time situation
by not using a setting based on the most critical system arrangement, which
sometimes does not provide the most appropriate solution. The algorithms for
relay settings are usually in low-level languages because of the need for a short
time response which is not obtained with high-level languages such as Pascal or
Fortran.
3.4.3 Typical architectures of numerical relays
Numerical relays are made up from modules with well-defined functions. Figure 3.8
shows a block diagram for numerical relays using typical modules.
The main modules are as follows:
• Microprocessor: responsible for processing the protection algorithms. It includes
the memory module which is made up from two memory components:
– RAM (random access memory), which has various functions, including
retaining the incoming data that is input to the processor and is necessary
for storing information during the compilation of the protection algorithm.
– ROM (read only memory) or PROM (programmable ROM), which are used
for storing programs permanently.
• Input module: the analogue signals from the substation are captured and
sent to the microprocessor and the module typically contains the following
Display
Phase N
Phase C
Power
supply
Phase B
Phase A
Filter
and
analog
to
digital
converter
Com port
Micro-processor
Outputs
Isolation
Keypad
Inputs
Figure 3.8 General arrangement of numerical relays (reproduced by permission of
Basler Electric)
Classification and function of relays 41
elements:
– analogical filters, which are active low-bandpass filters that eliminate any
background noise that has been induced in the line;
– signal conditioner, which converts the signal from the CTs into a normalised
DC signal;
– analogue digital convertor, which converts the normalised DC signal into
a binary number that is then sent directly to the microprocessor or to a
communications buffer.
• Output module: conditions the microprocessor response signals and sends them
to the external elements that they control. It is made up of a digital output which
generates a pulse as a response signal, and a signal conditioner which amplifies
and isolates the pulse.
• Communication module: contains series and parallel ports to permit the intercon-
nection of the protection relays with the control and communications systems of
the substation.
3.4.4 Standard functions of numerical relays
Numerical multifunction relays are similar in nature to a panel of single-function
protection relays. Both must be wired together with ancillary devices to operate as
a complete protection and control system. In the single-function and electromagnetic
environment, diagrams provide information on wiring protection elements, switches,
meters, and indicator lights. In the digital multifunction environment the process
of wiring individual protection or control elements is replaced by entering logic
settings. The process of creating a logic scheme is the digital equivalent of wiring a
panel. It integrates the multifunction protection, control, and input/output elements
into a unique protection and control system.
(i) Protection
Protection functions of numerical relays may typically include one or more of
the following: directional/non-directional three-phase overcurrent; directional/non-
directional earth-fault overcurrent; negative-sequence overcurrent; directional power;
over-excitation; over- and under-voltage; over- and under-frequency; distance; field
loss; differential; breaker failure; automatic reclosing; breaker monitoring and
automatic reclosing.
(ii) Measurement
Numerical relays normally incorporate outstanding measurement functions. Three-
phase currents and voltages are digitally sampled and the fundamental is extracted
using a discrete Fourier transform (DFT) algorithm. Metering functions include volt-
age, current, frequency, power factor, apparent power, reactive power and true power.
Metered values are viewed through any communication port using serial commands
or at the front panel HMI if available.
42 Protection of electricity distribution networks
(iii) Control
Most numerical relays incorporate at least one virtual breaker control switch and
several virtual switches which can be accessed locally from the HMI or remotely
fromthe communications port. The virtual breaker control switch permits the tripping
and closing of a selected breaker. The virtual switches can be used to trip and close
additional switches and breakers, or enable and disable certain functions.
(iv) Communication
In numerical relaying, relay and power system information can be retrieved from
a remote location using the ASCII command interface which can also be used to enter
settings, retrieve reports and metering information, and perform control operations.
A communication port on the relay front panel provides a temporary local interface
for communication. Communication ports on the rear panel provide a permanent
communication interface.
Panel communication ports can be connected to computers, terminals, serial
printers, modems, and logic intermediate communication/control interfaces such as
RS-232 serial multiplexors. Most numerical relay communication protocols support
ASCII and binary data transmissions. ASCII data is used to send and receive human
readable data and commands. Binary data is used for computer communication and
transmission of raw oscillographic fault data if available. In most numerical relays,
at least one of the following protocols is available – Modbus, DNP, Courier, IEC
608750-5 and MMS/UCA2.
(v) Reporting and alarms
The fault reporting functions provide means of recording and reporting information
about faults that have been detected by the relay. The most basic fault reporting
functions provided by a protection relay are the signalling or visual flags which
indicate the type of fault, usually referred to as targets. In addition, the numerical relay
can provide many advanced fault reporting features. These include fault summary
reports, sequence of events recorder reports, and oscillographic records. Above all,
it is essential to be able to download information to COMTRADE files, which can
be opened by numerous programs.
In numerical relays, the settings are introduced as logic equations and trip expres-
sions are used by the fault reporting function to start logging targets for an event
and to record the fault current magnitudes at the time of trip. The HMI uses the
trip expression to display the respective target by means of the associated LED. The
breaker monitoring function also uses the trip expression to start counting the breaker
operating time.
Pick-up expressions are used by the fault reporting function to time-stamp the
fault summary record, time the length of the fault from pick-up to drop-out (fault
clearance time) and to control the recording of oscillograph data. The HMI also uses
the pick-up expression to control the flashing of a trip LED. A pick-up expression is
also used by the setting group selection function to prevent a setting group change
during a fault.
Classification and function of relays 43
3.5 Supplies to the relay circuits
Protection relays are usually designed for either alternating current or direct current
circuits, the two types generally being totally independent of each other. Voltage
and/or current signals are taken from measurement transformers to feed the AC cir-
cuits. In the case of the so-called primary relays the connection is made directly to
the supply system. The AC signals feed the control circuits of the relays which then
determine whether or not fault conditions exist. Various alarmand control signals (for
example, those used to open switches) are carried along DC circuits. These circuits
normally obtain their supply from banks of batteries so that faults on the AC system
do not affect the operation of the switchgear mechanisms.
Chapter 4
Current and voltage transformers
Current or voltage instrument transformers are necessary to isolate the protection,
control and measurement equipment from the high voltages of a power system, and
for supplying the equipment with the appropriate values of current and voltage –
generally these are 1 A or 5 A for the current coils, and 120 V for the voltage coils.
The behaviour of current and voltage transformers during and after the occurrence of
a fault is critical in electrical protection since errors in the signal from a transformer
can cause mal-operation of the relays. In addition, factors such as the transient period
and saturation must be taken into account when selecting the appropriate transformer.
When only voltage or current magnitudes are required to operate a relay then the
relative direction of the current flow in the transformer windings is not important.
However, the polarity must be kept in mind when the relays compare the sum or
difference of the currents.
4.1 Voltage transformers
With voltage transformers (VTs) it is essential that the voltage from the secondary
winding should be as near as possible proportional to the primary voltage. In order to
achieve this, VTs are designed in such a way that the voltage drops in the windings
are small and the flux density in the core is well below the saturation value so that
the magnetisation current is small; in this way a magnetisation impedance is obtained
that is practically constant over the required voltage range. The secondary voltage of
a VT is usually 115 or 120 V with corresponding line-to-neutral values. The majority
of protection relays have nominal voltages of 120 or 69 V, depending on whether their
connection is line-to-line or line-to-neutral.
4.1.1 Equivalent circuit
VTs can be considered as small power transformers so that their equivalent circuit is
the same as for power transformers, as shown in Figure 4.1a. The magnetisation
46 Protection of electricity distribution networks
V
p
a
b
n:1
n
X
m
Z
L
d
c
e
f
Z
B
V
p
b d
n
n:1
a c
f
Z
L
e
Z
B
I
L
V
p
R
m
V
s
V
s
Z
H
/n
2
Z
H
/n
2
V
p
(a)
(b)
Figure 4.1 Voltage transformer equivalent circuits: (a) equivalent circuit;
(b) simplified circuit
I
L
I
L
(R
H
/n
2
+R
L
)
jI
L
(X
H
/n
2
+X
L
)
V
s
=V
ef
n
V
p
V
cd
=
Figure 4.2 Vector diagram for a voltage transformer
branch can be ignored and the equivalent circuit then reduces to that shown in
Figure 4.1b.
The vector diagram for a VT is given in Figure 4.2, with the length of the voltage
drops increased for clarity. The secondary voltage V
s
lags the voltage V
p
/n and is
smaller in magnitude. In spite of this, the nominal maximum errors are relatively
small. VTs have an excellent transient behaviour and accurately reproduce abrupt
changes in the primary voltage.
4.1.2 Errors
When used for measurement instruments, for example for billing and control pur-
poses, the accuracy of a VT is important, especially for those values close to the
nominal system voltage. Notwithstanding this, although the precision requirements
of a VTfor protection applications are not so high at nominal voltages, due to the prob-
lems of having to cope with a variety of different relays, secondary wiring burdens
Current and voltage transformers 47
and the uncertainty of system parameters, errors should be contained within narrow
limits over a wide range of possible voltages under fault conditions. This range should
be between 5 and 173 per cent of the nominal primary voltage for VTs connected
between line and earth.
Referring to the circuit in Figure 4.1a, errors in a VT are due to differences in
magnitude and phase between V
p
/n and V
s
. These consist of the errors under open-
circuit conditions when the load impedance Z
B
is infinite, caused by the drop in
voltage from the circulation of the magnetisation current through the primary wind-
ing, and errors due to voltage drops as a result of the load current I
L
flowing through
both windings. Errors in magnitude can be calculated from Error
VT
={(nV
s
−V
p
)/
V
p
} ×100%. If the error is positive, then the secondary voltage exceeds the
nominal value.
4.1.3 Burden
The standard burden for voltage transformers is usually expressed in volt-amperes
(VA) at a specified power factor.
Table 4.1 gives standard burdens based on ANSI Standard C57.13. Voltage trans-
formers are specified in IEC publication 186A by the precision class, and the value
of volt-amperes (VA).
The allowable error limits corresponding to different class values are shown in
Table 4.2, where V
n
is the nominal voltage. The phase error is considered positive
when the secondary voltage leads the primary voltage. The voltage error is the per-
centage difference between the voltage at the secondary terminals, V
2
, multiplied by
the nominal transformation ratio, and the primary voltage V
1
.
4.1.4 Selection of VTs
Voltage transformers are connected between phases, or between phase and earth. The
connection between phase and earth is normally used with groups of three single-
phase units connected in star at substations operating with voltages at about 34.5 kV
or higher, or when it is necessary to measure the voltage and power factor of each
phase separately.
The nominal primary voltage of a VT is generally chosen with the higher nom-
inal insulation voltage (kV), and the nearest service voltage in mind. The nominal
secondary voltages are generally standardised at 115 and 120 V. In order to select the
nominal power of a VT, it is usual to add together all the nominal VA loadings of
the apparatus connected to the VT secondary winding. In addition, it is important to
take account of the voltage drops in the secondary wiring, especially if the distance
between the transformers and the relays is large.
4.1.5 Capacitor voltage transformers
In general, the size of an inductive VT is proportional to its nominal voltage and,
for this reason, the cost increases in a similar manner to that of a high voltage trans-
former. One alternative, and a more economic solution, is to use a capacitor voltage
48 Protection of electricity distribution networks
T
a
b
l
e
4
.
1
S
t
a
n
d
a
r
d
b
u
r
d
e
n
s
f
o
r
v
o
l
t
a
g
e
t
r
a
n
s
f
o
r
m
e
r
s
S
t
a
n
d
a
r
d
b
u
r
d
e
n
C
h
a
r
a
c
t
e
r
i
s
t
i
c
s
f
o
r
1
2
0
V
a
n
d
6
0
H
z
C
h
a
r
a
c
t
e
r
i
s
t
i
c
s
f
o
r
6
9
.
3
V
a
n
d
6
0
H
z
d
e
s
i
g
n
v
o
l
t
-
a
m
p
e
r
e
s
p
o
w
e
r
f
a
c
t
o
r
r
e
s
i
s
t
a
n
c
e
(

)
i
n
d
u
c
t
a
n
c
e
(
H
)
i
m
p
e
d
a
n
c
e
(

)
r
e
s
i
s
t
a
n
c
e
(

)
i
n
d
u
c
t
a
n
c
e
(
H
)
i
m
p
e
d
a
n
c
e
(

)
W
1
2
.
5
0
0
.
1
0
1
1
5
.
2
3
.
0
4
1
1
5
2
3
8
.
4
1
.
0
1
3
8
4
X
2
5
.
0
0
0
.
7
0
4
0
3
.
2
1
.
0
9
5
7
5
1
3
4
.
4
0
.
3
6
4
1
9
2
Y
7
5
.
0
0
0
.
8
5
1
6
3
.
2
0
.
2
6
8
1
9
2
5
4
.
4
0
.
0
8
9
6
4
Z
2
0
0
.
0
0
0
.
8
5
6
1
.
2
0
.
1
0
1
7
2
2
0
.
4
0
.
0
3
4
2
4
Z
Z
4
0
0
.
0
0
0
.
8
5
3
1
.
2
0
.
0
4
0
3
3
6
1
0
.
2
0
.
0
1
6
8
1
2
M
3
5
.
0
0
0
.
2
0
8
2
.
3
1
.
0
7
0
4
1
1
2
7
.
4
0
.
3
5
6
1
3
7
Current and voltage transformers 49
Table 4.2 Voltage transformer error limits
Class Primary voltage Voltage error
(±%)
Phase error
(±min)
0.1 0.1 0.5
0.2 0.8V
n
, 1.0V
n
and 0.2 10.0
0.5 1.2V
n
0.5 20.0
1.0 1.0 40.0
0.1 1.0 40.0
0.2 0.05V
n
1.0 40.0
0.5 1.0 40.0
1.0 2.0 80.0
0.1 0.2 80.0
0.2 V
n
2.0 80.0
0.5 2.0 80.0
1.0 3.0 120.0
transformer. This device is effectively a capacitance voltage divider and is similar to
a resistive divider in that the output voltage at the point of connection is affected by
the load – in fact the two parts of the divider taken together can be considered as the
source impedance which produces a drop in voltage when the load is connected.
The capacitor divider differs from the inductive divider in that the equivalent
impedance of the source is capacitive and the fact that this impedance can be com-
pensated for by connecting a reactance in series at the point of connection. With an
ideal reactance there are no regulation problems – however, in an actual situation on
a network some resistance is always present. The divider can reduce the voltage to a
value that enables errors to be kept within normally acceptable limits. For improved
accuracy a high voltage capacitor is used in order to obtain a bigger voltage at the
point of connection, which can be reduced to a standard voltage using a relatively
inexpensive transformer as shown in Figure 4.3.
Asimplified equivalent circuit of a capacitor VTis shown in Figure 4.4 in which V
i
is equal to the nominal primary voltage, C is the numerically equivalent impedance
equal to (C
1
+C
2
), L is the resonance inductance, R
i
represents the resistance of
the primary winding of transformer T plus the losses in C and L, and Z
e
is the
magnetisation impedance of transformer T . Referred to the intermediate voltage, the
resistance of the secondary circuit and the load impedance are represented by R

s
and
Z

B
, respectively, while V

s
and I

s
represent the secondary voltage and current.
It can be seen that, with the exception of C, the circuit in Figure 4.4 is the same
as the equivalent circuit of a power transformer. Therefore, at the system frequency
when C and L are resonating and cancelling out each other, under stable system
conditions the capacitor VT acts like a conventional transformer. R
i
and R

s
are not
large and, in addition, I
e
is small compared to I

s
so that the vector difference between
50 Protection of electricity distribution networks
C
1
C
2
Z
B
T
V
C2
L
V
p
V
S
V
S
Figure 4.3 Capacitor VT basic circuit
L

B
Z
e
I
e

S
E
L
E
C
C
I
i
R
i

S
V
C2

S
V
i
Figure 4.4 Capacitor VT equivalent circuit
I
i
R
i
E
L
E
C
V
C2


S

S

S

S
I
i
I
e
V
i

Figure 4.5 Capacitor VT vector diagram
V
i
and V

s
, which constitutes the error in the capacitor VT, is very small. This is
illustrated in the vector diagram shown in Figure 4.5, which is drawn for a power
factor close to unity. The voltage error is the difference in magnitude between V
i
and
V

s
, whereas the phase error is indicated by the angle . From the diagram it can be
Current and voltage transformers 51
seen that, for frequencies different from the resonant frequency, the values of E
L
and
E
C
predominate, causing serious errors in magnitude and phase.
Capacitor VTs have a better transient behaviour than electromagnetic VTs since
the inductive and capacitive reactances in series are large in relation to the load
impedance referred to the secondary voltage and thus, when the primary voltage
collapses, the secondary voltage is maintained for some milliseconds because of the
combination of the series and parallel resonant circuits represented by L, C and
the transformer T .
4.2 Current transformers
Even though the performance required from a current transformer (CT) varies with
the type of protection, high grade CTs must always be used. Good quality CTs are
more reliable and result in fewer application problems and, in general, provide better
protection. The quality of CTs is very important for differential protection schemes
where the operation of the relays is directly related to the accuracy of the CTs under
fault conditions as well as under normal load conditions.
CTs can become saturated at high current values caused by nearby faults; to avoid
this, care should be taken to ensure that under the most critical faults the CT operates
on the linear portion of the magnetisation curve. In all these cases the CT should be
able to supply sufficient current so that the relay operates satisfactorily.
4.2.1 Equivalent circuit
An approximate equivalent circuit for a CT is given in Figure 4.6a, where n
2
Z
H
represents the primary impedance Z
H
referred to the secondary side, and the second-
ary impedance is Z
L
. R
m
and X
m
represent the losses and the excitation of the core.
(a)
a
b
1:n
n
2
Z
H
Z
L
d
c
e
f
Z
B
(b)
b d
1:n
a c
f
R
L
e
Z
B
I
H
/n I
H
I
L
I
E
R
m
X
m
X
m
V
s
E
s
Figure 4.6 Current transformer equivalent circuits
52 Protection of electricity distribution networks
I
L
I
H
/n
I
e
I
q
I
e
I
r
I
L
R
L
V
s
E
s
Figure 4.7 Vector diagram for the CT equivalent circuit
The circuit in Figure 4.6a can be reduced to the arrangement shown in Figure 4.6b
where Z
H
can be ignored, since it does not influence either the current I
H
/n or the
voltage across X
m
. The current flowing through X
m
is the excitation current I
e
.
The vector diagram, with the voltage drops exaggerated for clarity, is shown in
Figure 4.7. In general Z
L
is resistive and I
e
lags V
s
by 90

, so that I
e
is the principal
source of error. Note that the net effect of I
e
is to make I
L
lag and be much smaller
than I
H
/n, the primary current referred to the secondary side.
4.2.2 Errors
The causes of errors in a CT are quite different to those associated with VTs. In effect,
the primary impedance of a CTdoes not have the same influence on the accuracy of the
equipment – it only adds an impedance in series with the line, which can be ignored.
The errors are principally due to the current that circulates through the magnetising
branch. The magnitude error is the difference in magnitude between I
H
/n and I
L
and
is equal to I
r
, the component of I
e
in line with I
L
(see Figure 4.7).
The phase error, represented by , is related to I
q
, the component of I
e
that
is in quadrature with I
L
. The values of the magnitude and phase errors depend on
the relative displacement between I
e
and I
L
, but neither of them can exceed the
vectorial error I
e
. It should be noted that a moderate inductive load, with I
e
and I
L
approximately in phase, has a small phase error and the excitation component results
almost entirely in an error in the magnitude.
4.2.3 AC saturation
CT errors result from excitation current, so much so that, in order to check if a CT is
functioning correctly, it is essential to measure or calculate the excitation curve. The
magnetisation current of a CTdepends on the cross section and length of the magnetic
circuit, the number of turns in the windings, and the magnetic characteristics of the
material. Thus, for a given CT, and referring to the equivalent circuit of Figure 4.6b,
it can be seen that the voltage across the magnetisation impedance, E
s
, is directly
proportional to the secondary current. From this it can be concluded that, when the
Current and voltage transformers 53
primary current and therefore the secondary current is increased, these currents reach
a point when the core commences to saturate and the magnetisation current becomes
sufficiently high enough to produce an excessive error.
When investigating the behaviour of a CT, the excitation current should be mea-
sured at various values of voltage – the so-called secondary injection test. Usually it
is more convenient to apply a variable voltage to the secondary winding, leaving the
primary winding open-circuited. Figure 4.8a shows the typical relationship between
the secondary voltage and the excitation current, determined in this way. In European
standards the point K
p
on the curve is called the saturation or knee point and is defined
as the point at which an increase in the excitation voltage of ten per cent produces
an increase of 50 per cent in the excitation current. This point is referred to in the
ANSI/IEEE standards as the intersection of the excitation curves with a 45

tangent
line as indicated in Figure 4.8b. The European knee point is at a higher voltage than
the ANSI/IEEE knee point.
4.2.4 Burden
The burden of a CT is the value in ohms of the impedance on the secondary side of
the CT due to the relays and the connections between the CT and the relays. By way
of example, the standard burdens for CTs with a nominal secondary current of 5 A
are shown in Table 4.3, based on ANSI Standard C57.13.
IEC Standard Publication 185(1987) specifies CTs by the class of accuracy
followed by the letter Mor P, which denotes if the transformer is suitable for measure-
ment or protection purposes respectively. The current and phase error limits for
measurement and protection CTs are given in Tables 4.4a and 4.4b. The phase error is
considered positive when the secondary current leads the primary current. The current
error is the percentage deviation of the secondary current multiplied by the nomi-
nal transformation ratio, from the primary current, i.e., {(CTR×I
2
) −I
1
} ÷I
1
(%),
where I
1
=primary current (A), I
2
=secondary current (A), and CTR= current trans-
former transformation ratio. Those CT classes marked with ‘ext’ denote wide range
(extended) current transformers with a rated continuous current of 1.2 or two times
the nameplate current rating.
4.2.5 Selection of CTs
When selecting a CT, it is important to ensure that the fault level and normal load
conditions do not result in saturation of the core and that the errors do not exceed
acceptable limits. These factors can be assessed from:
• formulae;
• CT magnetisation curves;
• CT classes of accuracy.
The first two methods provide precise facts for the selection of the CT. The third only
provides a qualitative estimation. The secondary voltage E
s
in Figure 4.6b has to be
determined for all three methods. If the impedance of the magnetic circuit, X
m
, is
54 Protection of electricity distribution networks
10% V
k
50% I
ek
V
k
mA
Volts
Secondary excitation current
S
e
c
o
n
d
a
r
y

e
x
c
i
t
a
t
i
o
n

v
o
l
t
a
g
e
(a)
(b)
S
e
c
o
n
d
a
r
y

e
x
c
i
t
a
t
i
o
n

v
o
l
t
a
g
e

(
V
e
f
)
,

V

r
.
m
.
s
.
Secondary excitation current (I
e
), A r.m.s.
1.0 0.0001 0.01
10
0.1
100
1000
10 100
100:5
200:5
400:5
300:5
500:5
600:5
1000:5
1200:5
800:5
900:5
45°
B
A
* OHMS at 75°C.
Below this line the
exciting current for a
given voltage for any
unit will not exceed
the curve value by
more than 25%
600:5
1200:5
800:5
900:5
1000:5
400:5
500:5
300:5
200:5
100:5
120:1
240:1
200:1
180:1
160:1
60:1
80:1
100:1
20:1
40:1
0.31
0.61
0.41
0.46
0.51
0.20
0.25
0.15
0.10
0.05
Current
ratio
Turns
ratio
Sec*
res
K
p
I
ek
Figure 4.8 CT magnetisation curves: (a) defining the knee point in a CT excitation
curve accordingto Europeanstandards; (b) typical excitationcurves for a
multi-ratio class C CT ( from IEEE Standard C576.13-1978; reproduced
by permission of the IEEE)
Current and voltage transformers 55
Table 4.3 Standard burdens for protection CTs with 5 A secondary current
Designation Resistance
()
Inductance
(mH)
Impedance
()
Volt-amps
(at 5 A)
Power
factor
B-1 0.5 2.3 1.0 25 0.5
B-2 1.0 4.6 2.0 50 0.5
B-4 2.0 9.2 4.0 100 0.5
B-8 4.0 18.4 8.0 200 0.5
Table 4.4a Error limits for measurement current transformers
Class % current error at the given proportion
of rated current shown below
% phase error at the given proportion
of the rated current shown below
2.0

1.2 1.0 0.5 0.2 0.1 0.05 2.0

1.2 1.0 0.5 0.2 0.1 0.05
0.1 – 0.1 0.1 – 0.20 0.25 – – 5 5 – 8 10 –
0.2 – 0.2 0.2 – 0.35 0.50 – – 10 10 – 15 20 –
0.5 – 0.5 0.5 – 0.75 1.00 – – 30 30 – 45 60 –
1.0 – 1.0 1.0 – 1.50 2.00 – – 60 60 – 90 120 –
3.0 – 3.0 – 3.0 – – – – 120 – 120 – – –
0.1 ext 0.1 – 0.1 – 0.20 0.25 0.40 5 – 5 – 8 10 15
0.2 ext 0.2 – 0.2 – 0.35 0.50 0.75 10 – 10 – 15 20 30
0.5 ext 0.5 – 0.5 – 0.75 1.00 1.50 30 – 30 – 45 60 90
1.0 ext 1.0 – 1.0 – 1.50 2.00 – 60 – 60 – 90 120 –
3.0 ext 3.0 – – 3.0 – – – 120 – – 120 – – –

ext =200%
Table 4.4b Error limits for protection current CTs
Class Current error (%) at
proportion of rated primary
current shown
Phase error (minutes) at
proportion of rated primary
current shown
1.0 0.5 0.2 0.1 1.0 0.5 0.2 0.1
5P and 5P ext 1.0 – 1.5 2.0 60 – 90 120
10P and 10P ext 3.0 3.0 – – 120 120 – –
Total error for nominal error limit current and nominal load is 5% for 5P and 5P ext CTs
and 10% for 10P and 10P ext CTs.
56 Protection of electricity distribution networks
high, this can be removed from the equivalent circuit with little error, giving E
s
=V
s
,
and thus:
V
s
=I
L
(Z
L
+Z
C
+Z
B
) (4.1)
where: V
s
=r.m.s voltage induced in the secondary winding; I
L
=maximum sec-
ondary current in amperes (this can be determined by dividing the maximum
fault current on the system by the transformer turns ratio selected); Z
B
=external
impedance connected; Z
L
=impedance of the secondary winding; Z
C
=impedance
of the connecting wiring.
Use of the formula
This method utilises the fundamental transformer equation:
V
s
=4.44f ANB
max
10
−8
volts (4.2)
where: f =frequency in Hz; A=cross-sectional area of core (in
2
); N =number of
turns; B
max
=flux density (lines/in
2
).
The cross-sectional area of metal and the saturation flux density are sometimes
difficult to obtain. The latter can be taken as equal to 100 000 lines/in
2
, which is a typ-
ical value for modern transformers. To use the formula, V
s
is determined fromeqn. 4.1,
and B
max
is then calculated using eqn. 4.2. If B
max
exceeds the saturation density,
there could be appreciable errors in the secondary current and the CT selected would
not be appropriate.
Example 4.1
Assume that a CT with a ratio of 2000/5 is available, having a steel core of high
permeability, a cross-sectional area of 3.25in
2
, and a secondary winding with a
resistance of 0.31. The impedance of the relays, including connections, is 2.
Determine whether the CT would be saturated by a fault of 35 000 A at 50 Hz.
Solution
If the CTis not saturated, then the secondary current, I
L
, is 35000×5/2000=87.5A.
N =2000/5=400 turns, and V
s
=87.5×(0.31+2) =202.1 V. Using eqn. 4.2, B
max
can now be calculated:
B
max
=
202.1×10
8
4.44×50×3.25×400
=70030 lines/in
2
Since the transformer in this example has a steel core of high permeability this
relatively low value of flux density should not result in saturation.
Using the magnetisation curve
Typical CT excitation curves, which are supplied by manufacturers, state the r.m.s.
current obtained on applying an r.m.s. voltage to the secondary winding, with the
primary winding open-circuited. The curves give the magnitude of the excitation
current required in order to obtain a specific secondary voltage. The method consists
Current and voltage transformers 57
b
c
I
e
Tap 1
Tap 3
Tap 2
a
Tap 4
I
L
d
I
H
e
V
s
Figure 4.9 Using the magnetisation curve: a: assume a value for I
L
; b: V
s
=
I
L
(Z
L
+Z
c
+Z
B
); c: find I
e
fromthe curve; d: I
H
=n(I
L
+I
e
); e: draw
the point on the curve
of producing a curve that shows the relationship between the primary and secondary
currents for one tap and specified load conditions, such as shown in Figure 4.9.
Starting with any value of secondary current, and with the help of the magnetisa-
tion curves, the value of the corresponding primary current can be determined. The
process is summarised in the following steps:
(a) Assume a value for I
L
.
(b) Calculate V
s
in accordance with eqn. 4.1.
(c) Locate the value of V
s
on the curve for the tap selected, and find the associated
value of the magnetisation current, I
e
.
(d) Calculate I
H
/n (=I
L
+I
e
) and multiply this value by n to refer it to the primary
side of the CT.
(e) This provides one point on the curve of I
L
versus I
H
. The process is then repeated
to obtain other values of I
L
and the resultant values of I
H
. By joining the points
together the curve of I
L
against I
H
is obtained.
This method incurs an error in calculating I
H
/n by adding I
e
and I
L
together
arithmetically and not vectorially, which implies not taking account of the load angle
and the magnetisation branch of the equivalent circuit. However, this error is not great
and the simplification makes it easier to carry out the calculations.
After constructing the curve it should be checked to confirm that the maximum
primary fault current is within the transformer saturation zone. If not, then it will
be necessary to repeat the process, changing the CT tap until the fault current is
within the linear part of the characteristic. In practice it is not necessary to draw the
complete curve because it is sufficient to take the known fault current and refer to
the secondary winding, assuming that there is no saturation for the tap selected. This
converted value can be taken as I
L
initially for the process described earlier. If the
tap is found to be suitable after finishing the calculations, then a value of I
H
can be
obtained that is closer to the fault current.
58 Protection of electricity distribution networks
Accuracy classes established by the ANSI standards
The ANSI accuracy class of a CT (Standard C57.13) is described by two symbols –
a letter and a nominal voltage; these define the capability of the CT. C indicates that
the transformation ratio can be calculated, while T indicates that the transformation
ratio can be determined by means of tests. The classification C includes those CTs
with uniformly distributed windings and other CTs with a dispersion flux that has
a negligible effect on the ratio, within defined limits. The classification T includes
those CTs whose dispersion flux considerably affects the transformation ratio.
For example, with a CT of class C-100, the ratio can be calculated and the error
should not exceed ten per cent if the secondary current does not go outside the range
of 1 to 20 times the nominal current and if the load does not exceed 1 (1×
5A×20=100V) at a minimum power factor of 0.5.
These accuracy classes are only applicable for complete windings. When con-
sidering a winding provided with taps, each tap will have a voltage capacity
proportionally smaller, and in consequence it can only feed a portion of the load
without exceeding the ten per cent error limit. The permissible load is defined as
Z
B
=(N
P
V
C
)/100, where Z
B
is the permissible load for a given tap of the CT, N
P
is
the fraction of the total number of turns being used and V
C
is the ANSI voltage
capacity for the complete CT.
Example 4.2
The maximum fault current in a given circuit is 12 000 A. The nominal CT ratio
is 1200/5 and the CT is to be used with a tap of 800/5. The CT class is C-200,
the resistance of the secondary is 0.2, the total secondary load is 2.4 and the
power factor is 0.6. Determine if, on the occurrence of a fault, the error will exceed
ten per cent.
Solution
The resistance of the secondary winding of the CTcan be ignored since, by definition,
the class C-200 indicates that the CT could withstand 200 V plus the drop produced
by the resistance of the secondary with a current range equal to 20 times the nominal
value and with a load power factor as low as 0.5. Notwithstanding this, the voltage
drops in the secondary can be ignored only if the current does not exceed 100 A. For
the example given, I
L
=12000×(5/800) =75A.
The permissible load is given by
Z
B
=(N
P
V
C
) ÷100
N
P
=800/1200=0.667
so that
Z
B
=(0.667×200V) ÷100A=1.334
Since the loading of the circuit, 2.4, is more than the maximum permissible
(1.33), then the error could exceed ten per cent during a fault of 12 000 A, which
results in a maximum secondary current of 75 A. Consequently, it is necessary to
Current and voltage transformers 59
reduce the load, increase the current transformer tap or use another CT of a higher
class.
4.2.6 DC saturation
Up to now, the behaviour of a CT has been discussed in terms of a steady state,
without considering the DCtransient component of the fault current. However, the DC
component has more influence in producing severe saturation than the ACcomponent.
Figure 4.10 shows an example of the distortion and reduction in the secondary
current that can be caused by DCsaturation. However, the DCcomponent of the fault
current does not produce saturationof the CTif V
K
≥6.28IRT, where: V
K
=voltage at
the knee point of the magnetisation curve, determined by the extension of the straight
part of the curve; I =secondary symmetrical current (amperes −r.m.s.); R=total
resistance of the secondary; T =DC time constant of the primary current in cycles,
i.e.
T =
L
p
R
p
f
where: L
p
=inductance of the primary circuit; R
p
=resistance of the primary circuit;
f =frequency.
DC saturation is particularly significant in complex protection schemes since, in
the case of external faults, high fault currents circulate through the CTs. If saturation
occurs in different CTs associated with a particular relay arrangement, this could result
in the circulation of unbalanced secondary currents that would cause the system to
malfunction.
4.2.7 Precautions when working with CTs
Working with CTs associated with energised network circuits can be extremely haz-
ardous. In particular, opening the secondary circuit of a CT could result in dangerous
Secondary current
without saturation
Secondary current
with saturation
Figure 4.10 The effect of DC saturation on the secondary current
60 Protection of electricity distribution networks
overvoltages, which might harm operational staff or lead to equipment being dam-
aged because the current transformers are designed to be used in power circuits that
have an impedance much greater than their own. As a consequence, when secondary
circuits are left open, the equivalent primary-circuit impedance is almost unaffected
but a high voltage will be developed by the primary current passing through the mag-
netising impedance. Thus, secondary circuits associated with CTs must always be
kept in a closed condition or short-circuited in order to prevent these adverse situa-
tions occurring. To illustrate this, an example is given next using typical data for a
CT and a 13.2 kV feeder.
Example 4.3
Consider a 13.2 kV feeder that is carrying a load of 10 MVA at 1.0 power factor.
Associated with this circuit is a 500/5 CT feeding a measurement system whose total
load is 10 VA. The equivalent circuit of the CT referred to the secondary side is shown
in Figure 4.11. Calculate the voltage that would occur in the secondary circuit of the
CT if the measurement system was accidentally opened.
Solution
The single line diagram is given in Figure 4.12 and the equivalent circuit in
Figure 4.13.
150 Ω j50 Ω
0.2 Ω
Figure 4.11 CT equivalent circuit, referred to the secondary side, for Example 4.3
Z
load
500/5 V=13200/ 3
Figure 4.12 Single-line diagram for Example 4.3
Current and voltage transformers 61
j50Ω 150 Ω
0.2 Ω
0.4 Ω
174240Ω
V=762102.3V
Figure 4.13 Equivalent circuit for Figure 4.12
Referring the values to the secondary side of the CT gives
V =
13200

3
×
500
5
=762102.36V
Z
load
=
13.2
2
10
×

500
5

2
=174240
Z
meter
=
10
5
2
=0.4
When the secondary circuit is closed, the voltage across the measurement system can
be calculated approximately, ignoring the shunt branch, as
V
meter
=
762102.36
174240+0.2+0.4
A×0.4=4.37A×0.4=1.75V
If the secondary circuit is opened, the current is only able to circulate across the shunt
branch. In these conditions the voltage that appears at the terminals of the CT is
V
CT
=
762102.36
174240+(150j50)
×(150j50) =207.47

71.55

V
Therefore the voltage increases by almost 120 times.
Chapter 5
Overcurrent protection
5.1 General
Very high current levels in electrical power systems are usually caused by faults
on the system. These currents can be used to determine the presence of faults and
operate protection devices, which can vary in design depending on the complexity
and accuracy required. Among the more common types of protection are thermo-
magnetic switches, moulded-case circuit breakers (MCCBs), fuses, and overcurrent
relays. The first two types have simple operating arrangements and are principally
used in the protection of low voltage equipment. Fuses are also often used at low
voltages, especially for protecting lines and distribution transformers.
Overcurrent relays, which form the basis of this chapter, are the most common
formof protection used to deal with excessive currents on power systems. They should
not be installed purely as a means of protecting systems against overloads – which
are associated with the thermal capacity of machines or lines – since overcurrent
protection is primarily intended to operate only under fault conditions. However, the
relay settings that are selected are often a compromise in order to cope with both
overload and overcurrent conditions.
5.2 Types of overcurrent relay
Based on the relay operating characteristics, overcurrent relays can be classified
into three groups: definite current or instantaneous, definite time, and inverse time.
The characteristic curves of these three types are shown in Figure 5.1, which also
illustrates the combination of an instantaneous relay with one having an inverse time
characteristic.
5.2.1 Definite-current relays
This type of relay operates instantaneously when the current reaches a predetermined
value. The setting is chosen so that, at the substation furthest away from the source,
64 Protection of electricity distribution networks
t
Inverse time
Definite time
A
A
t
1
t
Definite current
t
A
Inverse time with instantaneous unit
t
A
Figure 5.1 Time/current operating characteristics of overcurrent relays
Overcurrent protection 65
the relay will operate for a low current value and the relay operating currents are
progressively increased at each substation, moving towards the source. Thus, the
relay with the lower setting operates first and disconnects load at the point nearest to
the fault. This type of protection has the drawback of having little selectivity at high
values of short-circuit current. Another disadvantage is the difficulty of distinguishing
between the fault current at one point or another when the impedance between these
points is small in comparison to the impedance back to the source, leading to the
possibility of poor discrimination.
Figure 5.2a illustrates the effect of the source impedance on the short-circuit level
at a substation, and for a fault at point B down the line. From Figure 5.2b it can be
appreciated that the fault currents at F
1
and F
2
are almost the same, and it is this
that makes it difficult to obtain correct settings for the relays. When there is some
A B
Z
R
Z
S
(a)
(b)
F
1
F
2
F
3
V
S
/ 3
Figure 5.2 Illustration of different levels of fault current. (a) Z
R
= impedance of
protected element. (b) Z
S
= source impedance. I
sc(A)
=

V
S
/

3

×

1/Z
S

, I
sc(B)
=V
S
/

3(Z
S
+Z
R
)

66 Protection of electricity distribution networks
considerable impedance between F
1
and F
2
, for example when the fault F
1
is located
down a long line, then the fault current at F
1
will be less than at F
2
. Similarly, due to
the impedance of the transformer, there will be a considerable difference between the
currents for faults at F
2
and F
3
, even though these two points are physically close.
If the protection settings are based on maximum fault level conditions, then these
settings may not be appropriate for the situation when the fault level is lower. How-
ever, if a lower value of fault level is used when calculating the relay settings, this
could result in some breakers operating unnecessarily if the fault level increases. As
a consequence, definite current relays are not used as the only overcurrent protection,
but their use as an instantaneous unit is common where other types of protection are
in use.
5.2.2 Definite-time/current or definite-time relays
This type of relayenables the settingtobe variedtocope withdifferent levels of current
by using different operating times. The settings can be adjusted in such a way that
the breaker nearest to the fault is tripped in the shortest time, and then the remaining
breakers are tripped in succession using longer time delays, moving back towards the
source. The difference between the tripping times for the same current is called the
discrimination margin.
Since the operating time for definite-time relays can be adjusted in fixed steps, the
protection is more selective. The big disadvantage with this method of discrimination
is that faults near to the source, which result in bigger currents, may be cleared
in a relatively long time. This type of relay has a current or pick-up setting – also
known as the plug or tap setting – to select the value at which the relay will start,
plus a time dial setting to obtain the exact timing of the relay operation. It should
be noted that the time-delay setting is independent of the value of the overcurrent
required to operate the relay. These relays are used a great deal when the source
impedance is large compared to that of the power system element being protected
when fault levels at the relay position are similar to those at the end of the protected
element.
5.2.3 Inverse-time relays
The fundamental property of these relays is that they operate in a time that is inversely
proportional to the fault current, as illustrated by the characteristic curves shown later.
Their advantage over definite-time relays is that, for very high currents, much shorter
tripping times can be obtained without risk to the protection selectivity. Inverse-
time relays are generally classified in accordance with their characteristic curve that
indicates the speed of operation; based on this they are commonly defined as being
inverse, very inverse, or extremely inverse. Inverse-time relays are also referred to
as inverse definite minimum time or IDMT overcurrent relays.
5.3 Setting overcurrent relays
Overcurrent relays are normally supplied with an instantaneous element and a time-
delay element within the same unit. When electromechanical relays were more
Overcurrent protection 67
popular, the overcurrent protection was made up from separate single-phase units.
The more modern microprocessor protection has a three-phase overcurrent unit and
an earth-fault unit within the same case. Setting overcurrent relays involves selecting
the parameters that define the required time/current characteristic of both the time-
delay and instantaneous units. This process has to be carried out twice, once for the
phase relays and then repeated for the earth-fault relays. Although the two processes
are similar, the three-phase short-circuit current should be used for setting the phase
relays while the phase-to-earth fault current should be used for the earth-fault relays.
When calculating the fault currents, the power system is assumed to be in its normal
operating state. However, at a busbar that has two or more transformers connected
in parallel and protected with relays that do not have the facility of multiple setting
groups – the ability to be adjusted to accommodate the prevailing system conditions,
which is possible with numerical relays for example – then better discrimination is
obtained if the calculations are carried out on the basis of each one of the transformers
being out of service in turn. The same procedure can be applied to multiple circuit
arrangements.
5.3.1 Setting instantaneous units
Instantaneous units are more effective when the impedances of the power system
elements beingprotectedare large incomparisontothe source impedance, as indicated
earlier. They offer two fundamental advantages:
• they reduce the operating time of the relays for severe system faults;
• they avoid the loss of selectivity in a protection system consisting of relays with
different characteristics; this is obtained by setting the instantaneous units so that
they operate before the relay characteristics cross, as shown in Figure 5.3.
The criteria for setting instantaneous units vary depending on the location, and
the type of system element being protected. Three groups of elements can be defined:
lines between substations, distribution lines, and transformers.
(i) Lines between substations
The setting of instantaneous units is carried out by taking at least 125 per cent of the
symmetrical r.m.s. current for the maximum fault level at the next substation. The
t
A
Figure 5.3 Preservation of selectivity using instantaneous units
68 Protection of electricity distribution networks
procedure must be started fromthe furthest substation, then continued by moving back
towards the source. When the characteristics of two relays cross at a particular system
fault level, thus making it difficult to obtain correct co-ordination, it is necessary to
set the instantaneous unit of the relay at the substation that is furthest away from the
source to such a value that the relay operates for a slightly lower level of current,
thus avoiding loss of co-ordination. The 25 per cent margin avoids overlapping the
downstream instantaneous unit if a considerable DC component is present. In HV
systems operating at 220 kV or above, a higher value should be used since the X/R
ratio becomes larger, as does the DC component.
(ii) Distribution lines
The setting of the instantaneous element of relays on distribution lines that supply
only pole-mounted MV/LVtransformers is dealt with differently to the previous case,
since these lines are at the end of the MV system. They therefore do not have to fulfil
the co-ordination conditions that have to be met by lines between substations and so
one of the following two values can be used to set these units:
1. 50 per cent of the maximum short-circuit current at the point of connection of
the CT supplying the relay.
2. Between six and ten times the maximum circuit rating.
(iii) Transformer units
The instantaneous units of the overcurrent relays installed on the primary side of the
transformers should be set at a value between 125 and 150 per cent of the short-circuit
current existing at the busbar on the lowvoltage side, referred to the high voltage side.
This value is higher that those mentioned previously to avoid lack of co-ordination
with the higher currents encountered due to the magnetic inrush current when energis-
ing the transformer. If the instantaneous units of the transformer secondary winding
overcurrent protection and the feeder relays are subjected to the same short-circuit
level, then the transformer instantaneous units need to be overridden to avoid loss
of selectivity unless there are communication links between these units that can per-
mit the disabling of the transformer instantaneous overcurrent protection for faults
detected by the feeder instantaneous overcurrent protection.
5.3.2 Coverage of instantaneous units protecting lines between substations
The percentage of coverage of an instantaneous unit which protects a line, X, can be
illustrated by considering the system shown in Figure 5.4.
The following parameters are defined:
K
i
=
I
pickup
I
end
and
K
s
=
Z
source
Z
element
Overcurrent protection 69
A B
Z
S
Z
AB
50
V
Figure 5.4 Coverage of instantaneous units
From Figure 5.4:
I
pickup
=
V
Z
s
+XZ
ab
(5.1)
where: V =voltage at the relay CT point; Z
s
=source impedance; Z
ab
=impedance
of the element being protected=Z
element
; X=percentage of line protected; I
end
=
current at the end of the line; and I
pick up
=minimum current value for relay pick up.
I
end
=
V
Z
s
+Z
ab
(5.2)
K
i
=
Z
s
+Z
ab
Z
s
+XZ
ab
⇒X=
Z
s
+Z
ab
−Z
s
K
i
Z
ab
K
i
(5.3)
This gives:
K
s
=
Z
s
Z
ab
⇒X=
K
s
(1−K
i
) +1
K
i
(5.4)
For example, if K
i
=1.25, and K
s
=1, then X=0.6, i.e. the protection covers
60 per cent of the line.
Example 5.1
The effect of reducing the source impedance, Z
s
, on the coverage provided by the
instantaneous protection can be appreciated by considering the system in Figure 5.5,
and using a value of 1.25 for K
i
in eqn. 5.4. From this:
Z
S
() Z
AB
() I
A
(A) I
B
(A) % coverage
10 10 100 50 60
2 10 500 83 76
70 Protection of electricity distribution networks
A B
Z
S
Z
AB
1000 V
Figure 5.5 Equivalent circuit for Example 5.1
A B
B
A
t
current
Discrimination
margin
Figure 5.6 Overcurrent inverse-time relay curves associated with two breakers on
the same feeder
5.3.3 Setting the parameters of time delay overcurrent relays
The operating time of an overcurrent relay has to be delayed to ensure that, in the
presence of a fault, the relay does not trip before any other protection situated closer to
the fault. The curves of inverse-time overcurrent relays associated with two breakers
on the same feeder in a typical system are shown in Figure 5.6, illustrating the
difference in the operating time of these relays at the same fault levels in order to
satisfy the discrimination margin requirements. Definite-time relays and inverse-time
relays can be adjusted by selecting two parameters – the time dial or time multiplier
setting, and the pick-up or plug setting (tap setting).
The pick-up setting
The pick-up setting, or plug setting, is used to define the pick-up current of the relay,
and fault currents seen by the relay are expressed as multiples of this. This value is
usually referred to as the plug setting multiplier (PSM), which is defined as the ratio
Overcurrent protection 71
of the fault current in secondary amps to the relay pick-up or plug setting. For phase
relays the pick-up setting is determined by allowing a margin for overload above the
nominal current, as in the following expression:
Pick-up setting=(OLF×I
nom
) ÷CTR (5.5)
where: OLF =overload factor that depends on the element being protected;
I
nom
=nominal circuit current rating; CTR=CT ratio.
The overload factor recommended for motors is 1.05. For lines, transformers and
generators it is normally in the range of 1.25 to 1.5. In distribution systems where it is
possible to increase the loading on feeders under emergency conditions, the overload
factor can be of the order of 2. In any case I
nom
has to be smaller than those of the
CT and the thermal capacity of the conductor; otherwise the smallest value has to be
taken to calculate the pick-up setting.
For earth-fault relays, the pick-up setting is determined taking account of the max-
imum unbalance that would exist in the system under normal operating conditions. A
typical unbalance allowance is 20 per cent so that the expression in eqn. 5.5 becomes
Pick-up setting=(0.2×I
nom
) ÷CTR (5.6)
In HV transmission lines the unbalance allowance could go down to
10 per cent, while in rural distribution feeders the value could be as high as 30 per cent.
Time dial setting
The time dial setting adjusts the time delay before the relay operates whenever the
fault current reaches a value equal to, or greater than, the relay current setting. In
electromechanical relays the time delay is usually achieved by adjusting the physical
distance between the moving and fixed contacts; a smaller time dial value results in
shorter operating times. The time dial setting is also referred to as the time multiplier
setting.
The criteria and procedures for calculating the time dial setting, to obtain the
appropriate protection and co-ordination for the system, are considered next. These
criteria are mainly applicable to inverse-time relays, although the same methodology
is valid for definite-time relays.
1. Determine the required operating time t
1
of the relay furthest away from the
source by using the lowest time dial setting and considering the fault level for
which the instantaneous unit of this relay picks up. This time dial setting may
have to be higher if the load that flows when the circuit is re-energised after a
loss of supply is high (the cold load pick-up), or if it is necessary to co-ordinate
with devices installed downstream, e.g. fuses or reclosers.
2. Determine the operating time of the relay associated with the breaker in the next
substation towards the source, t
2a
=t
1
+t
margin
, where t
2a
is the operating time
of the back-up relay associated with breaker 2 and t
margin
is the discrimination
margin. The fault level used for this calculation is the same as that used to
determine the timing t
1
of the relay associated with the previous breaker.
72 Protection of electricity distribution networks
3. With the same fault current as in 1 and 2 above, and knowing t
2a
and the pick-
up value for relay 2, calculate the time dial setting for relay 2. Use the closest
available relay time dial setting whose characteristic is above the calculated value.
4. Determine the operating time (t
2b
) of relay 2, but now using the fault level just
before the operation of its instantaneous unit.
5. Continue with the sequence, starting from the second stage.
The procedure referred to above is appropriate if it can be assumed that the relays
have their characteristic curves scaled in seconds. For those relays where the time
adjustment is given as a percentage of the operating curve for one second, the time
dial setting can be determined starting from the fastest multiplier applied to the curve
for time dial 1. In most modern relays the time settings can start from values as low
as 0.1 s, in steps of 0.1 s.
Time discrimination margin
A time discrimination margin between two successive time/current characteristics of
the order of 0.25 to 0.4 s should be typically used. This value avoids losing selectivity
due to one or more of the following:
• breaker opening time;
• relay overrun time after the fault has been cleared;
• variations infault levels, deviations fromthe characteristic curves of the relays (for
example, due to manufacturing tolerances), and errors in the current transformers.
In numerical relays there is no overrun, and therefore the margin could be chosen as
low as 0.2 s.
Single-phase faults on the star side of a Dy transformer are not seen on the delta
side. Therefore, when setting earth-fault relays, the lowest available time dial setting
can be applied to the relays on the delta side, which makes it possible to considerably
reduce the settings and thus the operating times of earth-fault relays nearer the source
infeed.
Use of mathematical expressions for the relay characteristics
The procedure indicated above for phase and earth units can easily be used when the
operating characteristics of the relays are defined by mathematical formulae instead
of by curves on log-log paper. IEC and ANSI/IEEE Standards define the operating
time mathematically by the following expression:
t =

(I/I
s
)
α
−1
+L (5.7)
where: t =relay operating time in seconds; k =time dial, or time multiplier, setting;
I =fault current level in secondary amps; I
s
=pick-up current selected; L=constant.
The constants α and β determine the slope of the relay characteristics. The val-
ues of α, β and L for various standard overcurrent relay types manufactured under
ANSI/IEEE and IECStandards are given in Table 5.1. Typical characteristics for both
types are shown in Figures 5.7 and 5.8.
Overcurrent protection 73
Table 5.1 ANSI/IEEEand IECconstants for standard
overcurrent relays
Curve description Standard α β L
Moderately inverse IEEE 0.02 0.0515 0.114
Very inverse IEEE 2.0 19.61 0.491
Extremely inverse IEEE 2.0 28.2 0.1217
Inverse CO8 2.0 5.95 0.18
Short-time inverse CO2 0.02 0.0239 0.0169
Standard inverse IEC 0.02 0.14 0
Very inverse IEC 1.0 13.5 0
Extremely inverse IEC 2.0 80.0 0
Long-time inverse UK 1.0 120 0
Given the relay characteristic, it is a straightforward task to calculate the time
response for a given time dial setting k, pick-up setting, and the other values of the
expression in eqn. 5.7. Likewise, if a particular time response and pick-up setting have
been determined, the time dial setting is found by solving k from the same equation.
5.4 Constraints of relay co-ordination
5.4.1 Minimum short-circuit levels
When the time delay unit has been set, using maximum fault levels, it is necessary
to check that the relays will operate at the minimum fault levels, and in the correct
sequence. For this it is sufficient to verify that the plug setting multiplier – (I/I
s
) in
eqn. 5.7 – under these conditions is greater than 1.5.
5.4.2 Thermal limits
Once the curves for the overcurrent relays have been defined, a check should be
made to ensure that they lie below the curves for the designated thermal capacity
of machines and cables. In the case of conductors, manufacturers’ graphs, which
indicate the length of time that different sizes can withstand various short-circuit
values, should be used. A typical graph for copper conductors with thermoplastic
insulation is given in Figure 5.9. For motors, the manufacturers’ information should
also be consulted.
In the case of transformers, the magnitude of the fault current that they
can withstand during a given time is limited by their impedance. ANSI/IEEE
Standard 242-1986 defines curves of short-circuit capacity for four categories of
liquid-immersed transformers, based on the nominal kVA rating of the transformer
and the short-circuit impedance.
74 Protection of electricity distribution networks
O
p
e
r
a
t
i
n
g

t
i
m
e

(
s
)
0.1
1
1
Current (multiples of I
s
)
IEC EI
IEC VI
IEC SI
UK LTI
10 100
10
100
1000
Figure 5.7 IEC overcurrent relay curves
Figures 5.10 to 5.13 show the curves of thermal capacity of transformers with the
following characteristics:
(i) Category I – power rating between 5 and 500 kVA single phase; 15 to 500 kVA
three phase.
(ii) Category II – power rating between 501 and 1667 kVA single phase; 501 to
5000 kVA three phase.
(iii) Category III – power rating between 1668 and 10 000 kVA single phase;
5001 kVA to 30 000 kVA three phase.
(iv) Category IV – power rating above 10 000 kVA single phase; above 30 000 kVA
three phase.
Overcurrent protection 75
0.1
1
Current (multiples of I
s
)
O
p
e
r
a
t
i
n
g

t
i
m
e

(
s
)
IEEE MI
1
10
100
10 100
IEEE VI
IEEE EI
US C08
US C02
Figure 5.8 ANSI/IEEE overcurrent relay curves
The thermal limit curves for Dy transformers have to be shifted to the left by a
ratio of 1/

3 to make them more sensitive. This compensates for the lower value of
current seen by the relays installed on the primary side, relative to the currents seen
by the relays on the secondary side, during single-phase fault conditions, as discussed
in Section 5.5.
5.4.3 Pick-up values
It is also important to check that the relay settings are not going to present problems
when other system elements are energised. This is particularly critical for motors,
76 Protection of electricity distribution networks
1
6

c
y
c
l
e
s

0
.
2
6
7

s
e
c
o
n
d
s
8

c
y
c
l
e
s

0
.
1
3
3

s
e
c
o
n
d
s
4

c
y
c
l
e
s

0
.
0
6
7

s
e
c
o
n
d
s
S
h
o
r
t

c
i
r
c
u
i
t

c
u
r
r
e
n
t

(
k
A
)
300
AWG
kcmil 10
8
20
6
30
4
2
1
1.5
2.5
3
8
5
6
7
10
9
100 40
4
50 60 80 70
2
90
3/0 2/0 1/0 4/0
200
9
0

c
y
c
l
e
s

1
.
5

s
e
c
o
n
d
s
1
8
0

c
y
c
l
e
s

3
.
0

s
e
c
o
n
d
s
6
0

c
y
c
l
e
s

1
.
0

s
e
c
o
n
d
s
3
0

c
y
c
l
e
s

0
.
5

s
e
c
o
n
d
s
800 400 600 1000
Copper conductor – thermal plastic insulation 75°C
60
20
30
40
50
80
70
90
100
200
300
400
500
2

c
y
c
l
e
s

0
.
0
3
3

s
e
c
o
n
d
s
1

c
y
c
l
e

0
.
0
1
7

s
e
c
o
n
d
s
Figure 5.9 Thermal limits of copper conductors with thermoplastic insulation
and the appropriate code letter, which indicates the number of times nominal current
taken when the motor is starting, should always be borne in mind.
In the case of transformers, the initial magnetisation inrush current that a trans-
former takes can be expressed as I
Inrush
=K×I
nom
, where I
nom
is the nominal
transformer current, and the constant K depends on the transformer capacity; from
Overcurrent protection 77
T
i
m
e

i
n

s
e
c
o
n
d
s
Symmetrical fault current in times normal
base current (ANSI/IEEE C57.12.00-1980)
The maximum short-circuit withstand capability
of category I transformers is defined in
(ANSI/IEEE C57.12.00-1980).
0.3
t =
Note =
0.1
0.2
2
5
0.6
0.5
0.4
0.9
1
0.7
0.8
2
4
3
10
9
8
6
7
20
40
30
70
80
50
60
90
40
Times normal base current
5 4 3 7 8 6 9 20 10 30 50 3 2 6 5 4 10 9 8 7 20 50 30 40
400
200
100
300
800
600
700
500
1000
900
3000
2000
4000
7000
6000
8000
9000
10000
5000
Through-fault protection curve
for faults that will occur
frequently or infrequently
where I =
1250
I
2
Figure 5.10 Thermal capacity of transformers between 5 and 500 kVAsingle phase;
15 to 500 kVA three phase ( from ANSI/IEEE Standard 242-1986;
reproduced by permission of the IEEE)
500 to 2500 kVA, K=8, and above 2500 kVA, K=10. The inrush point then remains
defined by the appropriate inrush current during 0.1 s.
Example 5.2
For the system shown in Figure 5.14, and starting from the data that are given there,
carry out the following:
1. Calculate the nominal currents and three-phase short-circuit levels at each
breaker.
78 Protection of electricity distribution networks
Sample I
2
t =K curves have been plotted for
selected transformer impedances as noted.
Times normal base current
Symmetrical fault current in times normal
base current (ANSI/IEEE C57.12.00-1980)
Constant determined at maximum I with
t =2 seconds
0.8
4
possible : I
2
t =K
0.1
3 2
0.2
0.3
0.7
0.4
0.5
0.6
Note =
K=
where I =
10 9 8 7 6 5 20 50 30 40
T
i
m
e

i
n

s
e
c
o
n
d
s
For fault currents from 70% to 100% of maximum
K transformer impedance
0.9
1
3
2
8
6
7
4
5
9
10
4 5 7 8 6 12 10
80
30
20
60
50
70
40
100
90
300
200
4 2 3 8 7 910 6 5 30 20 40 50
This curve may also be
used for back-up
protection where the
transformer is exposed
to frequent faults
normally cleared by
high-speed relaying
Through-fault protection curve
for faults that will occur frequently
(typically not more than 10 in
transformer lifetime)
8000
Through-fault protection curve
for faults that will occur frequently
(typically not more than 10 in
transformer lifetime)
800
600
500
700
400
900
1000
2000
3000
7000
6000
5000
4000
10000
9000
Figure 5.11 Thermal capacity of transformers between 501 and 1667 single phase;
501 to 5000 kVA three phase ( from ANSI/IEEE Standard 242-1968;
reproduced by permission of the IEEE)
2. Select the transformation ratios of the CTs.
3. Determine the values of the pick-up setting, time dial and instantaneous settings
of all phase relays to ensure a co-ordinated protection arrangement.
4. Find the percentage of the line BC protected by the instantaneous unit of the
overcurrent relay associated with breaker 2.
5. Draw the time/current characteristics of the relays on the system.
Overcurrent protection 79
0.8
0.1
0.2
0.3
0.7
0.4
0.5
0.6
T
i
m
e

i
n

s
e
c
o
n
d
s
0.9
1
3
2
8
6
7
4
5
9
10
80
30
20
60
50
70
40
100
90
300
200
8000
800
600
500
700
400
900
1000
2000
3000
7000
6000
5000
4000
10000
9000
Times normal base current
4 3 2 10 9 8 7 6 5 20 50 30 40 4 2 3 8 7 910 6 5 30 20 40 50
Through-fault protection curve
for faults that will occur frequently
(typically not more than 5 in
transformer lifetime)
Through-fault protection curve
for faults that will occur frequently
(typically not more than 5 in
transformer lifetime)
Sample I
2
t =K curves have been plotted for
selected transformer impedances as noted.
Symmetrical fault current in times normal
base current (ANSI/IEEE C57.12.00-1980)
Constant determined at maximum I with
t =2 seconds
possible : I
2
t =K
Note =
K=
where I =
For fault currents from 50% to 100% of maximum
This curve may also be
used for back-up
protection where the
transformer is exposed
to frequent faults
normally cleared by
high-speed relaying
K transformer impedance
4 5 7 8 6 12 10
Figure 5.12 Thermal capacity of transformers between 1668 and 10000 kVA single
phase; 5001 to 30000 kVAthree phase ( fromANSI/IEEEStandard 242-
1968; reproduced by permission of the IEEE)
Take into account the following considerations:
1. The discrimination margin to be 0.4 s.
2. All relays have inverse time characteristics, as shown in Figure 5.15.
3. Relay data:
Pick-up setting: 1 to 12 A in steps of 1 A
Time dial setting: as in Figure 5.15
Instantaneous: 6 to 144 A in steps of 1 A.
80 Protection of electricity distribution networks
0.8
0.1
0.2
0.3
0.7
0.4
0.5
0.6
T
i
m
e

i
n

s
e
c
o
n
d
s
0.9
1
3
2
8
6
7
4
5
9
10
80
30
20
60
50
70
40
100
90
300
200
8000
800
600
500
700
400
900
1000
2000
3000
7000
6000
5000
4000
10000
9000
Times normal base current
4 3 2 10 9 8 7 6 5 20 50 30 40 4 2 3 8 7 910 6 5 30 20 40 50
Sample I
2
t =K curves have been plotted for
selected transformer impedances as noted.
Symmetrical fault current in times normal
base current (ANSI/IEEE C57.12.00-1980)
Constant determined at maximum I with
t =2 seconds
possible : I
2
t =K
Note =
K=
where I =
For fault currents from 50% to 100% of maximum
Through-fault protection curve
for faults that will occur
frequently or infrequently
K transformer impedance
4 5 7 8 6 12 10
Figure 5.13 Thermal capacity of transformers above 10000 kVA single phase;
above 30000 kVA three phase ( from ANSI/IEEE Standard 242-1968;
reproduced by permission of the IEEE)
Solution
Calculation of nominal currents and three-phase short-circuit levels
From Figure 5.14 the short-circuit level at busbar A, and the impedance of the line
BC, can be obtained:
Z
source
=
V
2
P
sc
=

115+10
3

2
950×10
6
=13.92 referred to 115 kV
Overcurrent protection 81
3 MVA 3 MVA 3 MVA
B
1
.
1
2
5
Ω
(
1
3
.
2
k
V
)
8
5
.
3
5
Ω
(
1
1
5
k
V
)
1
C
2
115/13.2 kV
3
Y
Y
25 MVA
Z%=4.8
950 MVASC
A
4
Figure 5.14 Schematic diagram for Example 5.2
Z
transf
=Z
pu
×Z
Base
=0.048×

115×10
3

2
25×10
6
=25.39 referred to 115 kV
Z
lineBC
=85.35 referred to 115 kV
The equivalent circuit of the system referred to 115 kV is shown in Figure 5.16.
Nominal currents
I
nom1
=
P

3×V
=
3×10
6

3

13.2×10
3
=131.2A
82 Protection of electricity distribution networks
0.01
0.02
0.03
0.10
0.09
0.08
0.07
0.06
0.05
0.04
1.0
0.2
0.5
0.3
0.4
0.7
0.6
0.9
0.8
10
2
5
3
4
7
6
9
8
100
20
50
30
40
70
60
90
80
1 2 3 4 5 6 7 8 910 40 30 20
T
i
m
e

(
s
e
c
o
n
d
s
)
½
1
2
3
4
5
6
7
8
9
10
11
Multiples of tap value current
Time dial
setting
1 9 3 2 5 4 7 8 6 10 30 40 20
Figure 5.15 Typical operating curves for an inverse-time relay
Overcurrent protection 83
13.92Ω
3
115 kV
25.39Ω
A B
85.35Ω
C
Figure 5.16 Equivalent circuit of the system shown in Figure 5.14
I
nom2
=3×I
nom1
=3×131.2=393.6A
I
nom3
=
25×10
6

3

13.2×10
3
=1093.5A
I
nom4
=
25×10
6

3

115×10
3
=I
nom3
×(13.2/115) =125.5A
Short-circuit levels
The equivalent circuit gives:
I
faultC
=
115×10
3

3(13.92+25.39+83.35)
=532.6A referred to 115 kV
=532.6(115/13.2) =4640.2A referred to 13.2 kV
I
faultB
=
115×10
3

3(13.92+25.39)
=1689.0A referred to 115 kV
=1689(115/13.2) =14714.8A referred to 13.2 kV
I
faultA
=
115×10
3

3×13.92
=4769.8A referred to 115 kV
Choice of CT transformer ratio
The transformation ratio of the CTs is determined by the larger of the two following
values:
(i) I
nom
(ii) maximum short-circuit current without saturation being present. To fulfil this
condition and assuming that a C100 core is used and that the total burden is 1 ,
then I
sc
(5/X) ≤100A where I
sc
is the short-circuit current.
Table 5.2 summarises the calculations.
84 Protection of electricity distribution networks
Table 5.2 Nominal currents, short-circuit currents and CT
ratios for Example 5.2
Breaker
number
P
nom
(MVA)
I
nom
(A)
I
sc
(A)
(5/100)I
sc
(A)
CT
ratio
1 3 131.2 4640.0 232.0 300/5
2 9 393.6 14714.8 1735.7 800/5
3 25 1093.5 14714.8 735.7 1100/5
4 25 125.5 4769.8 238.5 300/5
Determination of the pick-up setting, time dial and instantaneous setting values:
calculation of the pick-up settings
Relay 1: 1.5(131.2)5/300=3.28A; ⇒pick-up set at 4 A
Relay 2: 1.5(393.6)5/800=3.69A; ⇒pick-up set at 4 A
Relay 3: 1.5(1093.5)5/1100=7.46A; ⇒pick-up set at 8 A
Relay 4: 1.5(125.5)5/300=3.14A; ⇒pick-up set at 4 A
Determination of time dial setting and calibration of the instantaneous setting
Relay 1
I
pick up
=4×300/5=240A
Time dial setting selected is 1.0
Setting of instantaneous element =(0.5I
sc
)(1/CTR) =(0.5×4640) 5/300=38.67A;
set at 39 A.
I
inst. trip
=39×300/5=2340A primary at 13.2 kV.
Plug setting multiplier, PSM
b
=(2340A×5/300) ×1/4=9.75 times
From Figure 5.15, with a plug setting multiplier of 9.75 and a time dial setting at 1,
t
1b
=0.1s
Relay 2
2340 A should produce operation of t
2a
in at least 0.1+0.4=0.5s.
PSM
a
=2340A×5/800×1/4=3.66 times
With PSM at 3.66 times, and t
op
at least 0.5 s, time dial =2 is chosen
Instantaneous setting=(1.25 I
faultC
)(1/CTR) =1.25 (4640)(5/800) =36.25A; set
at 37 A
I
inst.prim
=(37) 800/5=5920A at 13.2 kV
PSM
b
=5920A×5/800×1/4=9.25 times
With 9.25 PSM and time dial setting=2⇒t
2b
=0.18s
Relay 3
To discriminate with relay 2, take I
inst.prim2
=5920A
Require operation of t
3a
in at least 0.18+0.4=0.58s
PSM
a
=5920A×5/1100×1/8=3.36 times
With 3.36 PSM and t
op
=0.58s, ⇒time dial setting=2
Overcurrent protection 85
However, the instantaneous element of the relay associated with breaker 3 is over-
ridden and the discrimination time is applied for a fault on busbar B to avoid lack of
co-ordination with the instantaneous units of the relays associated with the feeders
from the busbar, as referred to in Section 5.3.1.
Based on I
sc
=14714.8 at 13.2 kV, PSM
b
=14714.8A×5/1100×1/8=8.36 times
With 8.36 times PSM and time dial setting=2, ⇒t
3b
=0.21s
Relay 4
For 14714.8 A, PSM=14714.8A (13.2/115) 5/300×1/4=7.04 times.
Require t
4
=0.21+0.4=0.61s
With 7.04 PSM and t
op
=0.61s ⇒time dial setting=5.
Setting of instantaneous element =(1.25×I
faultB
)(1/CTR)
=1.25(1689) 5/300
=35.19A
Setting=36A.
I
inst.prim
=36(300/5) =2160A referred to 115 kV
I
inst.prim
=2160(115/13.2) =18818.2A referred to 13.2 kV
Table 5.3 summarises the four relay settings.
Percentage of line A-B protected by the instantaneous element of the relay
associated with breaker 2
X%=
K
s
(1−K
i
) +1
K
i
K
i
=
I
sc pick up
I
sc end
=
5920
4640
=1.28
K
s
=
Z
source
Z
element
=
13.92+25.39
85.35
=0.46
and
X%=
0.46(1−1.28) +1
1.28
=0.68
Therefore, the instantaneous element covers 68 per cent of the line BC.
Table 5.3 Summary of settings for Example 5.2
Relay associated
with breaker number
Pick-up
(A)
Time
dial
Instantaneous
I
sec
(A)
Instantaneous
I
prim
(A)
1 2.5 – 40 2400
2 4.0 2 37 5920
3 8.0 2 – –
4 4.0 5 36 18818
86 Protection of electricity distribution networks
1000 s
100 s
10 s
1.0 s
0.100 s
0.10 s
0.100 kA 1.0 kA – l(13 kV) 10 kA 100 kA 1000 kA
Relay settings
R1:
Tap =4.0 Amp
Time dial =1.0
Inst =39.0 Amp
CT=300/5 Amp
R2:
Tap =4.0 Amp
Time dial =2.0
Inst =37.0 Amp
CT=800/5 Amp
R3:
Tap =8.0 Amp
Time dial =2.0
Inst =Disable
CT=1100/5 Amp
R4:
Tap =4.0 Amp
Time dial =5.0
Inst =36.0 Amp
CT=300/5 Amp
Thermal limits
T1:
Power transformer
L1:
Feeder cable
T1
R4
L1
R3
R2
R1
Figure 5.17 Relay co-ordination curves for Example 5.2
The co-ordination curves of the relays associated with this system are shown
in Figure 5.17. It should be noted that these are all drawn for currents at the same
voltage – in this case 13.2 kV.
5.5 Co-ordination across Dy transformers
In the case of overcurrent relay co-ordination for Dy transformers the distribution
of currents in these transformers should be checked for three-phase, phase-to-phase,
and single-phase faults on the secondary winding, shown in Figure 5.18.
To simplify the operations, it can be assumed that the voltages between the phases
of the transformer are the same, for both the primary and the secondary windings.
Overcurrent protection 87
Phase-to-earth fault
Three-phase fault Phase-to-phase fault
Figure 5.18 Distribution of current for a fault on a Dy transformer
Thus, the number of turns on the primary is equal to

3 times the number of turns
of the secondary, i.e. N
1
=

3N
2
.
Three-phase fault
I
f
=
E
φ−n
X
=I (5.8)
I
delta
=I
N
2
N
1
=
I

3
(5.9)
I
primary
=

3I
delta
=I (5.10)
From the above it can be seen that the currents that flow through the relays associated
with the secondary winding are equal to the currents that flow through those relays
associated with the primary winding, as expected, since the primary and secondary
voltages are equal and the fault involves all three phases.
Phase-to-phase fault
I
f
=
E
φ−φ
2X
=

3×E
φ−n
2X
=

3
2
I (5.11)
I
delta
=

3
2
×I ×
N
2
N
1
=
I
2
(5.12)
I
primary
=2I
delta
=I (5.13)
88 Protection of electricity distribution networks
For this case, the current that goes through the relays installed in the secondary
winding circuit is equal to

3/2 times the current that flows through the relays
associated with the primary on the phase that has the largest current value. From
Figure 5.18 it is clear that, for this fault, the current distribution at the primary is
1-1-2, and 0-1-1 at the secondary.
Phase-to-earth fault
I
f
=
E
φ−n
X
=I (5.14)
I
delta
=I ×
N
2
N
1
=
I

3
(5.15)
I
primary
=
I

3
(5.16)
Thus, for a phase-to-earth fault, the current through the relays installed in the sec-
ondary winding circuit on the faulted phase is equal to

3 times the current that flows
through the relays associated with the primary winding on the same phase.
The results of the three cases are summarised in Table 5.4. Analysing the results,
it can be seen that the critical case for the co-ordination of overcurrent relays is the
phase-to-phase fault. In this case the relays installed in the secondary carry a current
less than the equivalent current flowing through the primary relays, which could lead
to a situation where the selectivity between the two relays is at risk. For this reason,
the discrimination margin between the relays is based on the operating time of the
secondary relays at a current equal to

3I
f
/2, and on the operating time for the
primary relays for the full fault current value I
f
, as shown in Figure 5.19.
Example 5.3
For the system shown in Figure 5.20, calculate the following:
1. The three-phase short-circuit levels on busbars 1 and 2.
2. The transformation ratios of the CTs associated with breakers 1 to 8, given that
the primary turns are multiples of 100 except for the CT for breaker number 9,
which has a ratio of 250/5. Assume that the total burden connected to each CT
is 1 and that C100 cores are used.
Table 5.4 Summary of fault conditions
Fault I
primary
I
secondary
Three-phase I I
Phase-to-phase I

3I/2
Phase-to-earth I

3I
Overcurrent protection 89
t
2
3
I
f
I
f
A
0.4 s
Figure 5.19 Co-ordination of overcurrent relays for a Dy transformer
1 MVA
1
13.2 kV
1 MVA 1 MVA
Busbar 1
2
7 6
5
4
3
1 MVA
Z%=7.3
3 MVA
34.5/13.2 kV
Dy1
Busbar 2
34.5 kV
15 MVA
YYo
Z%=10
8
9 250/5
115 kV
115/34.5 kV
Busbar 3
Busbar 4
Busbar 5
Busbar 6
Busbar 7
Z
L
1
=
1
.
0
8
6
p
u
TR1
TR2
3.060 kA
0
.
0
0
0
k
A
L
2
L3 L4
0
.
4
5
3
k
A
0
.
4
5
3
k
A
0
.
0
0
0
k
A
2
.
1
7
0
k
A
0
.
6
5
1
k
A
182.87MVA
34.5kV 34.5 kV
34.5 kV
34.5 kV
Figure 5.20 Single line diagram for Example 5.3 showing fault currents
90 Protection of electricity distribution networks
3. The settings of the instantaneous elements, and the pick-up and time dial settings
of the relays to guarantee a co-ordinated protection arrangement, allowing a
discrimination margin of 0.4 s.
4. The percentage of the 34.5 kV line protected by the instantaneous element of the
overcurrent relay associated with breaker 6.
Take into account the following additional information:
1. The p.u. impedances are calculated on the following bases:
V =34.5kV
P =100MVA
2. The settings of relay 7 are:
Pick-up=4A
Time dial =0.3
Instantaneous =1100A primary current
3. All the relays have an IEC very inverse time characteristic with the values given
in Table 5.1
4. Relay data:
Pick-up setting=1 to 12 A in steps of 1 A
Time dial setting=0.05 to 10 in steps of 0.05
Instantaneous: 6 to 14 A in steps of 1 A
5. The setting of the instantaneous elements of the relays associated with the feeders
is to be carried out on the basis of ten times the maximum nominal current.
6. The short-circuit MVA and fault currents, for a fault on the 34.5 kV busbar at
substation A, are given in Figure 5.20.
Solution
Calculation of equivalent impedance
The short-circuit level on the 34.5 kV busbar at substation A can be obtained from
the values in Figure 5.20 (183.11 MVA). Using this, the equivalent impedance of the
system behind the busbar is calculated as follows:
Z
base
=
V
2
P
sc
=
(34500)
2
183.11×10
6
=6.5, referred to 34.5 kV
Z
transf1
=0.1
(34500)
2
15×10
6
=7.93, referred to 34.5 kV
=88.17, referred to 115 kV
Z
transf 2
=0.073
(34500)
2
3×10
6
=28.96, referred to 34.5 kV
Z
line
=1.086
(34500)
2
100×10
6
=12.93, referred to 34.5 kV
Overcurrent protection 91
BUSBAR 2
34.5 kV
34.5 kV
3
6-SUB-A
34.5 kV
6.50 Ω 12.93 Ω 28.96 Ω
BUSBAR 1
34.5 kV
Figure 5.21 Positive sequence network for Example 5.3
The equivalent positive sequence network, referred to 34.5 kV, is shown in
Figure 5.21.
Nominal currents
I
nom1,2,3
=
1×10
6

3×13.2×10
3
=43.74A at 13.2kV
I
nom4
=
3×10
6

3×13.2×10
3
=131.22A at 13.2kV
I
nom5
=
3×10
6

3×34.5×10
3
=50.20A at 34.5 kV
I
nom6
=50.20A at 34.5 kV
I
nom7
=
1×10
6

3×34.5×10
3
=16.73A
I
nom8
=251.02A at 34.5 kV
I
nom9
=
15×10
6

3×115×10
3
=75.31A at 115 kV
Short-circuit levels
The short-circuit MVA (P
sc
) of the transformer at 34.5kV=

3×2170.34×34.5×
10
3
=129.69MVA.
Z
transf1
+Z
base
=
(34.5)
2
129.69
=9.18, referred to 34.5 kV
=101.97, referred to 115 kV
Z
system
=101.97−88.17=13.80, referred to 115 kV
92 Protection of electricity distribution networks
I
fault1,2,3,4
=
34.5×10
3

3(6.5+12.93+28.96)
=411.63A at 34.5kV,
=1075.84A at 13.2kV
I
fault5
=
34.5×10
3

3(6.5+12.93)
=1025.15A at 34.5kV
I
fault6,7
=
34.5×10
3

3(6.5)
=3064.40A at 34.5kV
I
fault8
=
129.69×10
6

3×34.5×10
3
=2170.34A at 34.5kV
I
fault9
=
115×10
3

3×13.80
=4811.25A at 115kV
Selection of current transformers
Table 5.5 gives the main values for determining the transformation ratio of the CTs,
which is taken as the larger of the two following values:
• nominal current;
• maximum short-circuit current for which no saturation is present.
Therefore, (I
sc
×5/X) ≤100⇒X≥(I
sc
×5/100)
Determining the pick-up (PU) values
I
load1,2,3
=43.74A; PU
1,2,3
=(1.5)(43.74)(5/100) =3.28A; PU
1,2,3
=4A
I
load4
=131.22A; PU
4
=(1.5)(131.22)(5/200) =4.92A; PU
4
=5A
I
load5
=50.20A; PU
5
=(1.5)(50.20)(5/100) =3.76A; PU
5
=4A
Table 5.5 Determination of CT ratios for Example 5.3
Breaker
number
P
nom
(MVA)
I
nom
(A)
I
sc
(A)
(5/100)I
sc
(A)
CT
ratio
9 15 75.31 4797.35 239.87 250/5
8 15 251.02 2170.40 108.51 300/5
7 1 16.73 3060.34 153.01 200/5
6 3 50.20 3060.34 153.01 200/5
5 3 50.20 1025.67 51.28 100/5
4 3 131.22 1076.06 53.80 200/5
1, 2, 3 1 43.74 1076.06 53.80 100/5
Overcurrent protection 93
I
load6
=50.20A; PU
6
=(1.5)(50.20)(5/200) =1.88A; PU
6
=2A
PU
7
=4A (as given in the example data)
I
load8
=251.02A; PU
8
=(1.5)(251.02)(5/300) =6.28A; PU
8
=7A
I
load9
=75.31A; PU
9
=(1.5)(75.31)(5/250) =2.26A; PU
9
=3A
Determining the instantaneous and time dial settings
Relays 1, 2 and 3
When calculating the settings for the relays situated at the end of the circuit, the
minimum time dial setting of 0.05 is selected. From the given information, the set-
ting of the instantaneous element is based on ten times the maximum load current
seen by the relay. Thus, I
inst. trip
=10×I
nom
×(1/CTR) =10×43.74×(5/100) =
21.87A⇒set at 22A. I
prim. trip
=22(100/5) =440A.
Given the constants for the IEC very inverse overcurrent relay are α =
1.0, β =13.5, and L=0 then, from eqn. 5.7, the relay operating time t is
[(time dial setting) ×13.5]/(PSM−1), where PSM is the ratio of the fault current in
secondary amps to the relay pick-up current. PSM=22/4=5.5 times and, with the
time dial setting of 0.05, the relay operating time is (0.05×13.5)/(5.5−1) =0.15s.
Relay 4
To discriminate with relay 3 at 440 A requires operation in t
4a
−0.15+0.4=0.55s.
PSM
4a
=(440×5/200)(1/5) =2.2 times. At 2.2 times, and t
4a
=0.55s, the time
dial setting=0.55×(2.2−1)/13.5=0.049⇒0.05.
This relay has no setting for the instantaneous element, as referred to in Section
5.3.1. The operating time for a line-to-line fault is determined by taking 86 per cent of
the three-phase fault current. PSM
4b
=(0.86)(1075.84×5/200)(1/5) =4.63 times.
By similar calculations to those for relays 1, 2 and 3, t
4b
=0.19s.
Relay 5
The back-up to relay 4 is obtained by considering the operating time for a line-to-line
fault of t
5a
=0.19+0.4=0.59s.
PSM
5a
=1075.84×(13.2/34.5) ×(5/100) ×(1/4) =5.15 times. At 5.15 times,
and t
5a
=0.59s, this gives a required time dial setting of 0.20.
The setting of the instantaneous element is 1.25(1075.84) ×(13.2/34.5) ×
(5/100) =25.73A⇒26A, so that I
prim. trip
=26(100/5) =520A.
The operating time of the time delay element is calculated fromPSM
5b
=(1/4) ×
26=6.5 times. At 6.5 times, and with a time dial setting of 0.20, from the relay
characteristics and eqn. 5.7, t
5b
=0.5s.
Relay 6
At 520 A, this relay has to operate in t
6a
=0.5+0.4=0.9s.
PSM
6a
=520×(5/200) ×(1/2) =6.5 times. At 6.5 times and t
6a
=0.9s, the time
dial setting=0.37⇒0.40.
Instantaneous setting=1.25(1025.15)(5/200) =32.04⇒32A. I
prim. trip
=32A×
200/5=1280A.
94 Protection of electricity distribution networks
Relay 7: PSM and DIAL as given in the example data.
Relay 8
Relay backs up relays 6 and 7 and should be co-ordinated with the slower of these two
relays. Relay 7 has an instantaneous primary current setting of 1100 A, equivalent to
27.5 Asecondary current which is less than the setting of relay 6, and so the operating
time of both relays is determined by this value.
For relay 7 PSM=1100×5/200×1/4=6.87 times. At 6.87 times and with a
time dial setting of 0.3, then t
op
=0.69s.
For relay 6 PSM=1100×5/200×1/2=13.75 times. At 13.75 times and with a
time dial setting of 0.4, t
op
=0.42s.
Therefore the operating time to give correct discrimination with relay 7 is t
8a
=
0.69+0.4=1.09s.
For back-up relay 8, the contributions to relay 6 from substations G and M are
not considered. Only the infeed from the transformer has to be taken into account,
so that PSM
8a
=1100×(2170.34/3060.40) ×(5/300) ×(1/7) =1.86 times. At 1.86
times and t
8a
=1.09s, the time dial setting=0.07⇒0.1.
Here also, no instantaneous setting is applied to relay 8 for the reasons given
in Section 5.2.1. The maximum short-circuit current to be used for this relay is that
which flows from the 115 kV busbar to the 34.5 kV busbar for a fault on the latter,
and PSM
8b
=2170.34×(5/300) ×(1/7) =5.17 times. At 5.17 times and with a time
dial setting of 0.1, t
8b
=0.32s.
Relay 9
This relay backs up relay 8 in a time of t
9a
=0.4+0.32=0.72s.
PSM
9a
=2170.34×(34.5/115) ×(5/250) ×(1/3) =4.34 times. At 4.34 times
and t
9a
=0.72s, the time dial setting=0.18⇒0.20.
The instantaneous setting=1.25×2170.39×(34.5/115) ×(5/250) =16.28A⇒
17A. I
prim. trip
=17×(250/5) =850A referred to 115 kV.
The co-ordination curves of the relays associated with this system are shown in
Figure 5.22, and summarised in Table 5.6.
Percentage of 34.5 kV line protected by the instantaneous element of the overcurrent
relay associated with breaker 6
Given
%=
K
s
(1−K
i
) −1
K
i
where
K
i
=
I
sc.pickup
I
sc.end
=
1280
1025.15
=1.25
and
K
s
=
Z
source
Z
element
Overcurrent protection 95
1000s
100s
10s
1.0 s
0.100s
0.010s
0.010kA 0.100 kA 1.0 kA– l(13 kV) 100 kA 10 kA 1000kA
100 kA 10 kA 1.0 kA–1(34 kV) 0.100 kA 0.010kA
Relay settings
R1, R2, R3:
Tap =4.0 Amp
Time dial =0.05
Inst =22.0 Amp
CT=100/5 Amp
R4:
Tap =5.00 Amp
Time dial =0.05
Inst =Disable
CT=200/5 Amp
R5:
Tap =4.0 Amp
Time dial =0.2
Inst =26.0 Amp
CT=100/5 Amp
R6:
Tap =2.0 Amp
Time dial =0.4
Inst =32.0 Amp
CT=200/5 Amp
Thermal limits
TR1, TR2:
Power transformer
L1
Feeder cable
R7:
Tap =4.0 Amp
Time dial =0.3
Inst =27.5 Amp
CT=200/5 Amp
R8:
Tap =7.0 Amp
Time dial =0.1
Inst =Disable
CT=300/5 Amp
R9:
Tap =3.0 Amp
Time dial =0.2
Inst =17.0 Amp
CT=250/5 Amp
TR2
TR1
R-8
R-7
R-5
R-6
R-9
R1
R2
R4
L1
R3
Figure 5.22 Relay co-ordination curves for Example 5.3
Table 5.6 Summary of settings for Example 5.3
Relay
number
CT ratio Pick-up
(A)
Time
dial
Instantaneous
I
sec
(A)
1, 2, 3 100/5 4 1/2 20.0
4 200/5 5 1/2 –
5 100/5 4 3 26.0
6 200/5 2 6 32.0
7 200/5 4 5 27.5
8 300/5 7 1 –
9 250/5 3 2 17.0
96 Protection of electricity distribution networks
From the computer listing
Z
f
=
V
2
P
=
34.5
2
183.11
=6.50
and
K
s
=6.50/12.93=0.50
Therefore
%=
0.50(1−1.25) +1
1.25
=0.70
so that the instantaneous element covers 70 per cent of the line.
5.6 Co-ordination with fuses
When co-ordinating overcurrent relays it may be necessary to consider the
time/current characteristics of fuses which are used to protect MV/LV substation
transformers. When a fuse operates, the circuit is left in an open-circuit condition
until the fuse is replaced. It is therefore necessary to consider the case of preventing
fuse operation because of the problems of replacing them after they operate, which is
called fuse saving. In these cases it may be preferable to forgo the selectivity of the
protection system by not taking account of the fuse characteristic curve, so that the
fuse will then act as a back-up.
5.7 Co-ordination of negative-sequence units
Sensitivity to phase-to-phase fault detection can be increased by the use of negative-
sequence relays (type 50/51 Q) because a balanced load has no negative-sequence
(I
2
) current component. This is also the situation for phase-to-earth faults if type
50/51 relays are used since a balanced load has no zero-sequence (I
0
) component.
Instantaneous overcurrent and time-delay overcurrent negative-sequence units are
common features of the new multifunction relays. It is important to ensure that the
settings of these units are checked for co-ordination with phase-only sensing devices
such as downstream fuses and reclosers, and/or earth-fault relays.
To determine the pick-up setting of the negative-sequence element, it is neces-
sary to keep in mind that the magnitude of the current for a phase-to-phase fault is

3/2 (87 per cent) of the current for a three-phase fault at the same location, as
indicated in eqn. 5.11. On the other hand, the magnitude of the negative-sequence
component for a phase-to-phase fault can be obtained from the following expression
taken fromSection 2.2: I
a2
=1/3(I
a
+a
2
I
b
+aI
c
). For a phase-to-phase fault, I
a
=0
and I
b
=−I
c
. Therefore the magnitude of the negative-sequence component is 1 /

3
(58 per cent) of the magnitude of the phase-fault current. When the two factors (

3/2
and 1/

3) are combined, the

3 factors cancel, leaving a 1/2 factor.
Overcurrent protection 97
From the above it is recommended that negative-sequence elements be set by
taking one half of the phase pick-up setting in order to achieve equal sensitivity to
phase-to-phase faults as to three-phase faults.
To plot the negative-sequence time/current characteristics on the same diagram
as the phase- and earth-fault devices it is necessary to adjust the negative-sequence
element pick-up value by a multiplier that is the ratio of the fault current to the
negative-sequence current. For a phase-to-phase fault this is 1.732. The negative-
sequence pick-up value should be multiplied by a value greater than 1.732 for a
phase-to-phase-to-earth fault, and by a factor of 3 for a phase-to-earth fault. Since
no negative-sequence current flows for a three-phase fault, negative-sequence relay
operation does not take place, and no multiplying factor is involved.
Consider a downstream time-delay phase overcurrent element with a pick-up
value of 100 A, and an upstream negative-sequence time-delay relay with a pick-up
value of 150 A. In order to check the co-ordination between these two units for a
phase-to-phase fault, the phase overcurrent element plot should be shifted to the right
by a factor of 1.732, with a pick-up value of 1.732×150=259.8A. Generally, for
co-ordination with downstream phase overcurrent devices, phase-to-phase faults are
the most critical with all other fault types resulting in an equal or greater shift of the
time/current characteristic curve to the right on the co-ordination graph.
5.8 Overcurrent relays with voltage control
Faults close to generator terminals may result in voltage drop and fault current reduc-
tion, especially if the generators are isolated and the faults are severe. Therefore, in
generation protection it is important to have voltage control on the overcurrent time-
delay units to ensure proper operation and co-ordination. These devices are used to
improve the reliability of the relay by ensuring that it operates before the generator
current becomes too low. There are two types of overcurrent relays with this feature –
voltage-controlled and voltage-restrained, which are generally referred to as type 51V
relays.
The voltage-controlled (51/27C) feature allows the relays to be set below rated
current, and operation is blocked until the voltage falls well below normal voltage.
The voltage-controlled approach typically inhibits operation until the voltage drops
below a pre-set value. It should be set to function below about 80 per cent of rated
voltage with a current pick-up of about 50 per cent of generator rated current.
The voltage-restrained (51/27R) feature causes the pick-up to decrease with
reducing voltage, as shown in Figure 5.23. For example, the relay can be set for
175 per cent of generator rated current with rated voltage applied. At 25 per cent volt-
age the relay picks up at 25 per cent of the relay setting (1.75×0.25=0.44 times
rated). The varying pick-up level makes it more difficult to co-ordinate the relay with
other fixed pick-up overcurrent relays.
Since the voltage-controlled type has a fixed pick-up, it can be more readily co-
ordinated with external relays than the voltage-restrained type. On the other hand,
compared to the voltage-controlled relay, the voltage-restrained type will be less
98 Protection of electricity distribution networks
25 50 75 100
25
50
75
100
% V
Restraint
%
I
P
i
c
k
u
p
Figure 5.23 Pick-up setting of 51/27R relay
susceptible to operation on swings or motor-starting conditions that depress the
voltage below the voltage-controlled undervoltage unit drop-out point.
5.9 Setting overcurrent relays using software techniques
The procedure for determining the settings of overcurrent relays, as illustrated in the
past sections, is relatively simple for radial or medium-sized interconnected systems.
However, for large systems, the procedure becomes cumbersome if performed manu-
ally and therefore software techniques are required, especially if different topologies
have to be analysed. This section introduces a very simple procedure to set overcurrent
relays using different algorithms. The entry data required are the short-circuit currents
for faults at all busbars, the margins and constraints of the system and the available
settings of the relays being co-ordinated. In addition, the settings of those relays
closest to the loads and at the boundaries of other networks have to be considered.
The overall process consists basically of three steps:
1. Locate the fault and obtain the current for setting the relays.
2. Identify the pairs of relays to be set, determining first which one is furthest away
from the source, and which is acting as the back-up. The program should define
the settings in accordance with the criteria given in Section 5.3.
3. Verify that the requirements given in Section 5.4 are fulfilled; otherwise the
process should be repeated with lower discrimination margins, or new relays
should be tried.
The single line diagram given in Figure 5.20 can be used to illustrate simply how
a typical computer program can tackle a co-ordination problem.
The algorithm files the discrimination margins required, the setting of the relays
closest to the loads, (1, 2 and 3 in this case), and those for relay 9 which corresponds to
the only boundary utility. The algorithm then establishes pairs of relays and identifies
Overcurrent protection 99
which relay acts as a back-up within each pair. For the system shown in Figure 5.20,
the algorithm will determine which is the slowest relay of 1, 2 and 3, and co-ordinate
this with relay 4 located upstream. Relay 4 will, in turn, be co-ordinated with relay 5,
and relay 5 with relay 6, by ensuring that the required time discrimination margin is
maintained in all cases. Similar procedures are carried out for those relays associated
with the rest of the lines connected to busbar 6. After this, the algorithm determines
the slowest of these relays which then has to be co-ordinated with relay 8, and finally
relay 8 is co-ordinated with relay 9. When the process is finished, the algorithm
carries out all the necessary checks in accordance with the restraints given in the data
entry. If any requirement is not fulfilled, the process is started again with a lower
discrimination margin or using relays with different characteristics until adequate
co-ordination is achieved.
During the execution of the program the critical route, which corresponds to the
one with the highest number of relay pairs, should be identified. The more inter-
connected the system, the larger and more complicated will be the critical route and
computer programs are being used more and more for large systems. However, for
small systems and fault-case analysis, manual methods are still used with the help
of software editors containing libraries with relay curves from many manufacturers.
This reduces the curve drawing process, which is very time consuming.
5.10 Use of digital logic in numerical relaying
5.10.1 General
When using numerical relays it is necessary to provide a suitable method for handling
the relay logic capabilities, which should include blocks with control inputs, virtual
outputs, and hardware outputs. A group of logic equations defining the function of
the multifunction relay is called a logic scheme.
Numerical relays can be configured to suit a particular specification by defining
the operating settings (pick-up thresholds and time delays) of the individual protection
and control functions. Operating settings and logic settings are interdependent, but
separately programmed functions. Changing logic settings is similar to rewiring a
panel, and is used for managing the input, output, protection, control, monitoring and
reportingcapabilities of multifunctionprotectionrelaysystems. Eachrelaysystemhas
multiple, normally-contained function blocks that have all of the inputs and outputs
of its discrete component counterpart. Each independent function block interacts with
control inputs, virtual outputs, and hardware outputs based on logic variables defined
in equation formwith relay logic. Relay logic equations entered and saved in the relay
system’s nonvolatile memory integrate the selected or enabled protection in order to
provide the operating settings that control the relay pick-up threshold and time delay
values.
5.10.2 Principles of digital logic
Digital systems are constructed by using three basic logic gates. These gates are
designated AND, OR and NOT. There also exist other logical gates, such as the
100 Protection of electricity distribution networks
A
B
AB
B
A
A+B
AND OR
NAND NOR
B
A
AB
B
A
A+B
EOR
A
B
A + B
NOT
A A
Figure 5.24 Logic gate symbols
NAND and EOR gates. The basic operations and logic gate symbols are summarised
in Figure 5.24.
The AND gate is an electronic circuit that gives a high output only if all its inputs
are high. A dot (.) is used to show the AND operation but this dot is usually omitted,
as in Figure 5.24. The OR gate gives a high output if one or more of its inputs are
high. A plus sign (+) is used to show the OR operation. The NOT gate produces
an inverted version of the input at its output. It is also known as an inverter. If the
variable is A, the inverted output is known as NOT A and is shown as A. The NAND
is a NOT-AND circuit that is equal to an AND circuit followed by a NOT circuit. The
outputs of all NAND circuits are high if any of the inputs are low. The NOR gate is a
NOT-ORcircuit that is equal to an ORcircuit followed by a NOT circuit. The outputs
of all NOR gates are low if any of the inputs are high. The EOR – the Exclusive OR
gate – is a circuit that will give a high output if either, but not both, of its two inputs
are high. An encircled plus sign, ⊕, is used to show the EOR operation.
Table 5.7 shows the input/output combinations for the gate functions mentioned
above. Note that a truth table with n inputs has 2n rows.
5.10.3 Logic schemes
Normally numerical relays have several pre-programmed logic schemes which are
stored in the relay memory. Each scheme is configured for a typical protection applica-
tion and virtually eliminates the need for start-from-scratch programming. Protection
scheme designers may select from a number of pre-programmed logic schemes
that perform the most common protection and control requirements. Alternatively
customised schemes can be created using the relay logic capabilities.
Figure 5.25 shows a typical pre-programmed logic scheme provided with a
numerical relay, where the features of each logic scheme are broken down into func-
tional groups. This logic scheme provides basic time and instantaneous overcurrent
protection. The protective elements include phase, neutral, and negative-sequence
overcurrent protection. Functions such as breaker failure, virtual breaker control,
Overcurrent protection 101
Table 5.7a Input/output combinations for
various gate functions
Inputs Outputs
A B AND OR NAND NOR EOR
0 0 0 0 1 1 0
0 1 0 1 1 0 1
1 0 0 1 1 0 1
1 1 1 1 0 0 0
Table 5.7b Input/output
combinations for
the NOT gate
Input Output
0 1
1 0
automatic reclosing and protective voltage features are not enabled in this scheme.
However, these features may be achieved by appropriate design of the relay logic.
This numerical relay has 4 programmable inputs IN1 to IN4; five programmable
outputs OUT1 to OUT5; one programmable alarm output OUTA; ten virtual outputs
VO6 to VO15; four virtual selection switches 43, 143, 243, 343 and four protection
setting groups with external or automatic selection modes.
The phase, neutral and negative-sequence elements are activated to provide timed
(51) and instantaneous (50) overcurrent protection in this scheme. A function block
is disabled by setting the pick-up set-point at zero in each of the four setting groups.
Virtual output VO11 is assigned for all protective trips. When VO11 becomes TRUE,
OUT1 will operate and trip the breaker. Contact outputs OUT2, OUT3, OUT4, and
OUT5 are designated to specific function blocks. OUT2 operates for instantaneous
phase overcurrent conditions, OUT3 trips for timed phase overcurrent situations,
OUT4 operates for instantaneous neutral and negative-sequence overcurrent con-
ditions, and OUT5 operates for timed neutral and negative-sequence overcurrent
conditions. Input 1 IN1 is typically assigned to monitor breaker status (52b). A set-
ting group can be selected automatically or by using the communication ports or the
front panel HMI. Automatic setting group changes are normally based on current
level and duration. Setting group changes initiated by contact sensing inputs are not
accommodated in this scheme, but the logic inputs can be programmed to provide
this function.
102 Protection of electricity distribution networks
Sg0
Sg1
Sg2
Sg3
D0
D1
D2
D3
Auto
Active
setting
group
control 0
Phase
IOC
(50TP)
BLK
50TPPU
50TPT
Neutral
IOC
(50TN)
BLK
50TNPU
50TNT
Neg seq
IOC
(50TQ)
BLK
50TQPU
50TQT
Phase
TOC
(51P)
BLK
51PPU
51PT
Neutral
TOC
(51N)
BLK
51NPU
51NT
Neg seq
TOC
(51Q)
BLK
51QPU
51QT
6
VO12 PROT PU
Co-outx
VO5
51NT+51QT
Out5
Co-outx
VO4
50TNT+50TQT
Out4
VO3– 51PT
Co-outx
Out3
VO2 –50TPT
Co-outx
Out2
VO1– 52TC
Co-outx
Out1
VOA– alarm
Co-outx
Output
logic
Output
logic
Output
logic
Output
logic
Output
logic
Output
logic
Outa
6
VO11 PROT TRIP
343
CO-343
CO-243
243
CO-143
143
CO-43
43
OPTO
IN4
IN3
IN2
IN1 +

+

OPTO
+

OPTO
+

OPTO
Note: For clarity, multiple variables going to the same OR gate are shown by a single line into the OR gate.
Figure 5.25 Typical pre-programmed logic scheme of a numerical relay
(reproduced by permission of Basler Electric)
5.11 Adaptive protection with group settings change
This section presents several examples of logic scheme customisation to provide
functions that normally are not incorporated in numerical relays as part of a man-
ufacturer’s package. Topology changes, for example, affect the short-circuit levels
and therefore an incorrect co-ordination might arise if the relays are not reset for the
prevailing power system conditions. To overcome this, adaptive protection, which
can be implemented by using the multiple setting groups feature included in most
numerical relays, is essential.
Figure 5.26 shows a portion of a power system that might have four scenarios as
follows:
• system normal;
• one of the transformers out of service for maintenance;
• the grid infeed not available;
• the in-house generator out of service.
For a fault on one of the feeders, resulting in a fault current of I
f
, from the co-
ordination curve R
1
in Figure 5.27 the feeder relay would operate in a time t
1
. With
both transformers in service the fault current passing through each transformer would
Overcurrent protection 103
G2
T1 T2
1
2
3
4
Grid
Figure 5.26 Electrical system to illustrate setting group change
0.5I
f
t
1
R
2
R
1
R
2a
t
2

2a

2

1

f
I
f
Figure 5.27 Overcurrent co-ordination curves considering adaptive relaying
104 Protection of electricity distribution networks
be 0.5I
f
. The relay R
2
on the low voltage side of the transformer would then operate
in time t
2
, giving a discrimination margin of (t
2
−t
1
).
However, when one transformer is out of service, the current through the remain-
ing transformer increases to I

f
. From curve R
2
, the transformer relay tripping time
would then reduce fromt
2
to t

2
, which is faster than the feeder relay operating time t

1
,
leading to incorrect relay operation. It is thus necessary to move the transformer low
voltage relay curve upwards to R
2a
, where the relay operating time at I

f
will be t

2a
,
in order to maintain discrimination with the feeder relays. The transformer relays can
be pre-programmed to ensure this shift of the relay curve takes place automatically
when one transformer is out of service.
5.12 Exercises
5.1 Consider a power system with the same single line diagram as used for
Example 5.2, but with a 115/34.5 kV 58.5 MVA transformer. The feeders from
substation C each have a capacity of 10 MVA. If the short-circuit level at substa-
tion A is 1400 MVA, determine the setting of relays 1, 2, 3 and 4 if the same type of
relay is used.
5.2 Calculate the pick-up setting, time dial setting and the instantaneous setting of
the phase relays installed in the high voltage and lowvoltage sides of the 115/13.2 kV
transformers T1 and T3, in the substation illustrated in Figure 5.28. The short-circuit
levels, CT ratios and other data are shown in the same diagram. The considerations
used for Examples 5.2 and 5.3 also apply here.
5.3 For the system shown in Figure 5.29, carry out the following calculations:
1. The maximum values of short-circuit current for three-phase faults at busbars A,
B and C, taking into account that busbar D has a fault level of 12906.89 A r.m.s.
symmetrical (2570.87 MVA).
2. a) The maximum peak values to which breakers 1, 5 and 8 can be subjected.
b) The r.m.s. asymmetrical values that breakers 1, 5 and 8 can withstand for 5
cycles for guarantee purposes.
For these calculations assume that the L/R ratio is 0.2.
3. The turns ratios of the CTs associated with breakers 1 to 8. The CT in breaker
6 is 100/5. Take into account that the secondaries are rated at 5 A and that the
ratios available in the primaries are multiples of 50 up to 400, and from then on
are in multiples of 100.
4. The instantaneous, pick-up and time dial settings for the phase relays in order
to guarantee a co-ordinated protection system, allowing a time discrimination
margin of 0.4 s.
5. The percentage of the 34.5 kV line that is protected by the instantaneous element
of the overcurrent relay associated with breaker 5.
Overcurrent protection 105
Typical
2
4
k
A

1
P
H

S
C
2
1
.
9
k
A

3
P
H

S
C
Fault
Feeder
115/13.2kV 115/13.2kV
1
6
.
0
9
k
A

1
P
H

S
C
1
4
.
7
k
A

3
P
H

S
C
2000/5A
Z%=10.1
41.7MVA
13.2 kV Busbar
(5MVA)
CO-11
50–600/5A
W W
CO-11
3 3
CO-11 CO-11
900/5A
CO-11
W
Z%=9.63
20 MVA
CO-11
W
1
.
6
k
A

3
P
H

S
C
MITSU
T3
300/5A
0

1
P
H

S
C
CO-11 CO-11
ASGEN
T1
150/5A
W
CO-11
115kV Busbar
3
50
51
50
51
3
1 1
3
50N
51N
50
51
50N
51N
50N
51N
51
50N
51N
51
1
0
.
8
2
k
A

3
P
H

S
C
0

1
P
H

S
C
7
.
2
1
k
A

3
P
H

S
C
7
.
9
1
k
A

1
P
H

S
C
Figure 5.28 Single line diagram for Exercise 5.2
106 Protection of electricity distribution networks
2.625MVA 2.625 MVA
1 2
34.5 kV
6 (CT 100/5)
34.5 kV
5.25 MVA
Z%=6.0
3
13.2 kV
A
34.5/13.2 kV
Dy
4
B
14.2 km
jO.625 Ω/km
5
C
115/34.5 kV
Yy
7
10.5 MVA
Z%=11.7
2570.87 MVASC
8
D
Figure 5.29 Single line diagram for Exercise 5.3
Overcurrent protection 107
Bear in mind the following additional information:
• The settings of relay 6 are as follows: pick-up 7 A, time dial 5, instantaneous
setting 1000 A primary current.
• All the relays are inverse time type, with the following characteristics:
Pick-up: 1 to 12 in steps of 2 A
Time dial: as in Figure 5.15
Instantaneous element: 6 to 144 in steps of 1 A
Calculate the setting of the instantaneous elements of the relays associated with the
feeders assuming 0.5 I
sc
on busbar A.
Chapter 6
Fuses, reclosers and sectionalisers
A wide variety of equipment is used to protect distribution networks. The particular
type of protection used depends on the system element being protected and the sys-
tem voltage level, and, even though there are no specific standards for the overall
protection of distribution networks, some general indication of how these systems
work can be made.
6.1 Equipment
The devices most used for distribution system protection are:
• overcurrent relays;
• reclosers;
• sectionalisers;
• fuses.
The co-ordinationof overcurrent relays was dealt withindetail inthe previous chapter,
and this chapter will cover the other three devices referred to above.
6.1.1 Reclosers
A recloser is a device with the ability to detect phase and phase-to-earth overcurrent
conditions, to interrupt the circuit if the overcurrent persists after a predetermined
time, and then to automatically reclose to re-energise the line. If the fault that origi-
nated the operation still exists, then the recloser will stay open after a preset number
of operations, thus isolating the faulted section fromthe rest of the system. In an over-
head distribution system between 80 to 95 per cent of the faults are of a temporary
nature and last, at the most, for a few cycles or seconds. Thus, the recloser, with its
opening/closing characteristic, prevents a distribution circuit being left out of service
for temporary faults. Typically, reclosers are designed to have up to three open-
close operations and, after these, a final open operation to lock out the sequence.
110 Protection of electricity distribution networks
One further closing operation by manual means is usually allowed. The counting
mechanisms register operations of the phase or earth-fault units which can also be
initiated by externally controlled devices when appropriate communication means
are available.
The operating time/current characteristic curves of reclosers normally incorpor-
ate three curves, one fast and two delayed, designated as A, B and C, respectively.
Figure 6.1 shows a typical set of time/current curves for reclosers. However,
new reclosers with microprocessor-based controls may have keyboard-selectable
Current, A
T
i
m
e
,

s
A
B
C
2
0
0
0
3
0
0
0
4
0
0
0
5
0
0
0
1
0
0
0
7
0
0
5
0
0
4
0
0
3
0
0
2
0
0
1
0
0
7
0
5
0
0.01
0.02
0.03
0.10
0.07
0.05
0.3
0.2
0.5
1.0
0.7
7
2
3
5
10
20
Figure 6.1 Time/current curves for reclosers
Fuses, reclosers and sectionalisers 111
time/current curves which enable an engineer to produce any curve to suit the co-
ordination requirements for both phase and earth-faults. This allows reprogramming
of the characteristics to tailor an arrangement to a customer’s specific needs without
the need to change components.
Co-ordination with other protection devices is important in order to ensure that,
when a fault occurs, the smallest section of the circuit is disconnected to minimise dis-
ruption of supplies to customers. Generally, the time characteristic and the sequence
of operation of the recloser are selected to co-ordinate with mechanisms upstream
towards the source. After selecting the size and sequence of operation of the recloser,
the devices downstream are adjusted in order to achieve correct co-ordination. A typ-
ical sequence of a recloser operation for a permanent fault is shown in Figure 6.2.
The first shot is carried out in instantaneous mode to clear temporary faults before
they cause damage to the lines. The three later ones operate in a timed manner with
predetermined time settings. If the fault is permanent, the time-delay operation allows
other protection devices nearer to the fault to open, limiting the amount of the network
being disconnected.
Earth faults are less severe than phase faults and, therefore, it is important that
the recloser has an appropriate sensitivity to detect them. One method is to use CTs
connected residually so that the resultant residual current under normal conditions is
approximately zero. The recloser should operate when the residual current exceeds
the setting value, as would occur during earth faults.
Reclosers can be classified as follows:
• single-phase and three-phase;
• mechanisms with hydraulic or electronic operation;
• oil, vacuum or SF
6
.
Single-phase reclosers are used when the load is predominantly single-phase.
In such a case, when a single-phase fault occurs the recloser should permanently
disconnect the faulted phase so that supplies are maintained on the other phases.
Three-phase reclosers are used when it is necessary to disconnect all three phases in
order to prevent unbalanced loading on the system.
Reclosers withhydraulic operatingmechanisms have a disconnectingcoil inseries
with the line. When the current exceeds the setting value, the coil attracts a piston that
opens the recloser main contacts and interrupts the circuit. The time characteristic
and operating sequence of the recloser are dependent on the flow of oil in different
chambers. The electronic type of control mechanism is normally located outside
the recloser and receives current signals from a CT-type bushing. When the current
exceeds the predetermined setting, a delayed shot is initiated that finally results in
a tripping signal being transmitted to the recloser control mechanism. The control
circuit determines the subsequent opening and closing of the mechanism, depending
on its setting. Reclosers with electronic operating mechanisms use a coil or motor
mechanism to close the contacts. Oil reclosers use the oil to extinguish the arc and
also to act as the basic insulation. The same oil can be used in the control mechanism.
Vacuum and SF
6
reclosers have the advantage of requiring less maintenance.
112 Protection of electricity distribution networks
F
a
u
l
t

c
u
r
r
e
n
t
S
t
a
r
t

o
f

f
a
u
l
t
R
e
c
l
o
s
i
n
g

i
n
t
e
r
v
a
l
s

(
c
o
n
t
a
c
t
s

o
p
e
n
)
T
i
m
e
d

o
p
e
r
a
t
i
o
n

(
c
o
n
t
a
c
t
s

c
l
o
s
e
d
)
L
o
a
d

c
u
r
r
e
n
t

(
c
o
n
t
a
c
t
s

c
l
o
s
e
d
)
R
a
p
i
d

o
p
e
r
a
t
i
o
n

(
c
o
n
t
a
c
t
s

c
l
o
s
e
d
)
F
i
g
u
r
e
6
.
2
T
y
p
i
c
a
l
s
e
q
u
e
n
c
e
f
o
r
r
e
c
l
o
s
e
r
o
p
e
r
a
t
i
o
n
Fuses, reclosers and sectionalisers 113
Reclosers are used at the following points on a distribution network:
• in substations, to provide primary protection for a circuit;
• in main feeder circuits, in order to permit the sectioning of long lines and thus
prevent the loss of a complete circuit due to a fault towards the end of the circuit;
• in branches or spurs, to prevent the tripping of the main circuit due to faults on
the spurs.
When installing reclosers it is necessary to take into account the following
factors:
1. System voltage.
2. Short-circuit level.
3. Maximum load current.
4. Minimum short-circuit current within the zone to be protected by the recloser.
5. Co-ordination with other mechanisms located upstream towards the source, and
downstream towards the load.
6. Sensitivity of operation for earth faults.
The voltage rating and the short-circuit capacity of the recloser should be equal
to, or greater than, the values that exist at the point of installation. The same criteria
should be applied to the current capability of the recloser in respect of the maximum
load current to be carried by the circuit. It is also necessary to ensure that the fault
current at the end of the line being protected is high enough to cause operation of the
recloser.
6.1.2 Sectionalisers
Asectionaliser is a device that automatically isolates faulted sections of a distribution
circuit once an upstream breaker or recloser has interrupted the fault current and is
usually installed downstream of a recloser. Since sectionalisers have no capacity to
break fault current, they must be used with a back-up device that has fault current
breaking capacity. Sectionalisers count the number of operations of the recloser during
fault conditions. After a preselected number of recloser openings, and while the
recloser is open, the sectionaliser opens and isolates the faulty section of line. This
permits the recloser to close and re-establish supplies to those areas free of faults. If
the fault is temporary, the operating mechanism of the sectionaliser is reset.
Sectionalisers are constructed in single- or three-phase arrangements with
hydraulic or electronic operating mechanisms. A sectionaliser does not have a
current/time operating characteristic, and can be used between two protective devices
whose operating curves are very close and where an additional step in co-ordination
is not practicable.
Sectionalisers with hydraulic operating mechanisms have an operating coil in
series with the line. Each time an overcurrent occurs the coil drives a piston that
activates a counting mechanism when the circuit is opened and the current is zero
by the displacement of oil across the chambers of the sectionaliser. After a pre-
arrangednumber of circuit openings, the sectionaliser contacts are openedbymeans of
114 Protection of electricity distribution networks
pretensioned springs. This type of sectionaliser can be closed manually. Sectionalisers
with electronic operating mechanisms are more flexible in operation and easier to set.
The load current is measured by means of CTs and the secondary current is fed
to a control circuit which counts the number of operations of the recloser or the
associated interrupter and then sends a tripping signal to the opening mechanism.
This type of sectionaliser is constructed with manual or motor closing.
The following factors should be considered when selecting a sectionaliser:
1. System voltage.
2. Maximum load current.
3. Maximum short-circuit level.
4. Co-ordination with protection devices installed upstream and downstream.
The nominal voltage and current of a sectionaliser should be equal to or greater
than the maximum values of voltage or load at the point of installation. The short-
circuit capacity (momentary rating) of a sectionaliser should be equal to or greater
than the fault level at the point of installation. The maximum clearance time of the
associated interrupter should not be permitted to exceed the short-circuit rating of
the sectionaliser. Co-ordination factors that need to be taken into account include
the starting current setting and the number of operations of the associated interrupter
before opening.
6.1.3 Fuses
A fuse is an overcurrent protection device; it possesses an element that is directly
heated by the passage of current and is destroyed when the current exceeds a predeter-
mined value. Asuitably selected fuse should open the circuit by the destruction of the
fuse element, eliminate the arc established during the destruction of the element and
then maintain circuit conditions open with nominal voltage applied to its terminals,
(i.e. no arcing across the fuse element).
The majority of fuses used in distribution systems operate on the expulsion
principle, i.e. they have a tube to confine the arc, with the interior covered with
de-ionising fibre, and a fusible element. In the presence of a fault, the interior fibre
is heated up when the fusible element melts and produces de-ionising gases which
accumulate in the tube. The arc is compressed and expelled out of the tube; in addi-
tion, the escape of gas from the ends of the tube causes the particles that sustain the
arc to be expelled. In this way, the arc is extinguished when current zero is reached.
The presence of de-ionising gases, and the turbulence within the tube, ensure that the
fault current is not re-established after the current passes through zero point. The zone
of operation is limited by two factors; the lower limit based on the minimum time
required for the fusing of the element (minimum melting time) with the upper limit
determined by the maximum total time that the fuse takes to clear the fault.
There are a number of standards to classify fuses according to the rated voltages,
rated currents, time/current characteristics, manufacturing features and other consid-
erations. For example, there are several sections of ANSI/UL198-1982 standards that
cover low voltage fuses of 600 V or less. For medium and high voltage fuses within
Fuses, reclosers and sectionalisers 115
the range 2.3–138 kV, standards such as ANSI/IEEE C37.40, 41, 42, 46, 47 and 48
apply. Other organisations and countries have their own standards; in addition, fuse
manufacturers have their own classifications and designations.
In distribution systems, the use of fuse links designated K and T for fast and slow
types, respectively, depending on the speed ratio, is very popular. The speed ratio is
the ratio of minimum melt current that causes fuse operation at 0.1 s to the minimum-
melt current for 300 s operation. For the K link, a speed ratio (SR) of 6-8 is defined
and, for a T link, 10-13. Figure 6.3 shows the comparative operating characteristics
of type 200 K and 200 T fuse links. For the 200 K fuse a 4400 A current is required
0.01
0.02
0.03
0.05
0.07
0.1
0.2
0.3
0.7
0.5
1
3
2
7
5
10
20
30
50
70
100
200
300
500
700
1000
0.01
0.02
0.05
0.03
0.1
0.07
0.3
0.2
0.7
0.5
1
2
3
7
5
10
30
20
70
50
100
300
200
700
500
1000
T
i
m
e
,

s
T
i
m
e
,

s
Current, A Current, A
(a) (b)
3
0
0
2
0
0
1
0
0
3
0
0
0
2
0
0
0
1
0
0
0
7
0
0
5
0
0
1
0
0
0
7
0
0
0
5
0
0
0
1
0
0
0
0
3
0
0
2
0
0
1
0
0
5
0
0
7
0
0
1
0
0
0
0
7
0
0
0
5
0
0
0
3
0
0
0
2
0
0
0
Figure 6.3 Time/current characteristics of typical fuse links: (a) 200 K fuse link;
(b) 200 T fuse link
116 Protection of electricity distribution networks
for 0.1 s clearance time and 560 A for 300 s, giving an SR of 7.86. For the 200 T
fuse, 6500 A is required for 0.1 s clearance, and 520 A for 300 s; for this case, the
SR is 12.5.
The following information is required in order to select a suitable fuse for use on
the distribution system:
1. Voltage and insulation level.
2. Type of system.
3. Maximum short-circuit level.
4. Load current.
The above four factors determine the fuse nominal current, voltage and short-
circuit capability characteristics.
Selection of nominal current
The nominal current of the fuse should be greater than the maximum continuous load
current at which the fuse will operate. An overload percentage should be allowed
according to the protected-equipment conditions. In the case of power transformers,
fuses should be selected such that the time/current characteristic is above the inrush
curve of the transformer and below its thermal limit. Some manufacturers have pro-
duced tables to assist in the proper fuse selection for different ratings and connection
arrangements.
Selection of nominal voltage
The nominal voltage of the fuse is determined by the following system
characteristics:
• maximum phase-to-phase or phase-to-earth voltage;
• type of earthing;
• number of phases (three or one).
The system characteristics determine the voltage seen by the fuse at the moment
when the fault current is interrupted. Such a voltage should be equal to, or less than,
the nominal voltage of the fuse. Therefore, the following criteria should be used:
1. In unearthed systems, the nominal voltage should be equal to, or greater than,
the maximum phase-to-phase voltage.
2. In three-phase earthed systems, for single-phase loads the nominal voltage should
be equal to, or greater than, the maximum line-to-earth voltage and for three-
phase loads the nominal voltage is selected on the basis of the line-to-line voltage.
Selection of short-circuit capacity
The symmetrical short-circuit capacity of the fuse should be equal to, or greater than,
the symmetrical fault current calculated for the point of installation of the fuse.
Fuses, reclosers and sectionalisers 117
Fuse notation
When two or more fuses are used on a system, the device nearest to the load is called
the main protection, and that upstream, towards the source, is called the back up. The
criteria for co-ordinating them will be discussed later.
6.2 Criteria for co-ordination of time/current devices in
distribution systems
The following basic criteria should be employed when co-ordinating time/current
devices in distribution systems:
1. The main protection should clear a permanent or temporary fault before the back-
up protection operates, or continue to operate until the circuit is disconnected.
However, if the main protection is a fuse and the back-up protection is a recloser,
it is normally acceptable to co-ordinate the fast operating curve or curves of the
recloser to operate first, followed by the fuse, if the fault is not cleared. (See
Section 6.2.2.)
2. Loss of supply caused by permanent faults should be restricted to the smallest
part of the system for the shortest time possible.
In the following sections criteria and recommendations are given for the
co-ordination of different devices used on distribution systems.
6.2.1 Fuse-fuse co-ordination
The essential criterion when using fuses is that the maximum clearance time for a
main fuse should not exceed 75 per cent of the minimum melting time of the back-
up fuse, for the same current level, as indicated in Figure 6.4. This ensures that
the main fuse interrupts and clears the fault before the back-up fuse is affected in
any way. The factor of 75 per cent compensates for effects such as load current and
ambient temperature, or fatigue in the fuse element caused by the heating effect of
fault currents that have passed through the fuse to a fault downstream but were not
sufficiently large enough to melt the fuse.
The co-ordination between two or more consecutive fuses can be achieved by
drawing their time/current characteristics, normally on log-log paper as for over-
current relays. In the past, co-ordination tables with data of the available fuses
were also used, which proved to be an easy and accurate method. However, the
graphic method is still popular not only because it gives more information but also
because computer-assisted design tools make it much easier to draw out the various
characteristics.
6.2.2 Recloser-fuse co-ordination
The criteria for determining recloser-fuse co-ordination depend on the relative loca-
tions of these devices, i.e. whether the fuse is at the source side and then backs up the
operation of the recloser that is at the load side, or vice versa. These possibilities are
treated in the following paragraphs.
118 Protection of electricity distribution networks
Current, A
T
i
m
e
,

s
t
1
t
2
Figure 6.4 Criteria for fuse-fuse co-ordination: t
1
<0.75 t
2
Fuse at the source side
When the fuse is at the source side, all the recloser operations should be faster than
the minimum melting time of the fuse. This can be achieved through the use of
multiplying factors on the recloser time/current curve to compensate for the fatigue
of the fuse link produced by the cumulative heating effect generated by successive
recloser operations. The recloser opening curve modified by the appropriate factor
then becomes slower but, even so, should be faster than the fuse curve. This is
illustrated in Figure 6.5.
The multiplying factors referred to above depend on the reclosing time in cycles
and on the number of the reclosing attempts. Some values proposed by Cooper Power
Systems are reproduced in Table 6.1.
It is convenient to mention that if the fuse is at the high voltage side of a power
transformer and the recloser at the low voltage side, either the fuse or the recloser
curve should be shifted horizontally on the current axis to allow for the transformer
turns ratio. Normally it is easier to shift the fuse curve, based on the transformer tap
that produces the highest current on the high voltage side. On the other hand, if the
transformer connection group is delta-star, the considerations given in Section 5.5
should be taken into account.
Fuses at the load side
The procedure to co-ordinate a recloser and a fuse, when the latter is at the load side,
is carried out with the following rules:
• the minimum melting time of the fuse must be greater than the fast curve of the
recloser times the multiplying factor, given in Table 6.2 and taken from the same
reference as above;
Fuses, reclosers and sectionalisers 119
Transformer
Recloser
Current, A
T
i
m
e
,

s
A
C

R
I
sc
max
at recloser installation
C times k
Fuse
R
Figure 6.5 Criteria for source-side fuse and recloser co-ordination
Table 6.1 k factor for source-side fuse link
Reclosing time in cycles Multipliers for:
two fast,
two delayed
sequence
one fast,
three delayed
sequence
four delayed
sequence
25 2.70 3.20 3.70
30 2.60 3.10 3.50
50 2.10 2.50 2.70
90 1.85 2.10 2.20
120 1.70 1.80 1.90
240 1.40 1.40 1.45
600 1.35 1.35 1.35
The k factor is used to multiply the time values of the delayed curve of the recloser.
120 Protection of electricity distribution networks
Table 6.2 k factor for the load-side fuse link
Reclosing time in cycles Multipliers for:
one fast two fast
operation operations
25–30 1.25 1.80
60 1.25 1.35
90 1.25 1.35
120 1.25 1.35
The k factor is used to multiply the time values of the recloser
fast curve.
• the maximum clearing time of the fuse must be smaller than the delayed curve of
the recloser without any multiplying factor; the recloser should have at least two
or more delayed operations to prevent loss of service in case the recloser trips
when the fuse operates.
The application of the two rules is illustrated in Figure 6.6.
Better co-ordination between a recloser and fuses is obtained by setting the
recloser to give two instantaneous operations followed by two timed operations.
In general, the first opening of a recloser will clear 80 per cent of the temporary
faults, while the second will clear a further 10 per cent . The load fuses are set
to operate before the third opening, clearing permanent faults. A less effective co-
ordination is obtained using one instantaneous operation followed by three timed
operations.
6.2.3 Recloser-recloser co-ordination
The co-ordination between reclosers is obtained by appropriately selecting the
amperes setting of the trip coil in the hydraulic reclosers, or of the pick-ups in
electronic reclosers.
Hydraulic reclosers
The co-ordination margins with hydraulic reclosers depend upon the type of
equipment used. In small reclosers, where the current coil and its piston produce
the opening of the contacts, the following criteria must be taken into account:
• separation of the curves by less than two cycles always results in simultaneous
operation;
• separation of the curves by between two and 12 cycles could result in simultaneous
operation;
• separation greater than 12 cycles ensures nonsimultaneous operation.
Fuses, reclosers and sectionalisers 121
T
i
m
e
,

s
Current, A
I
sc
max
at fuse installation
A times k
A

C
R
R Recloser Fuse
Figure 6.6 Criteria for recloser and load-side fuse co-ordination
With large capacity reclosers, the piston associated with the current coil only
actuates the opening mechanism. In such cases, the co-ordination margins are as
follows:
• separation of the curves by less than two cycles always results in simultaneous
operation;
• a separation of more than eight cycles guarantees non-simultaneous operation.
The principle of co-ordination between two large units in series is based on the
time of separation between the operating characteristics, in the same way as for
small units.
122 Protection of electricity distribution networks
Electronically-controlled reclosers
Adjacent reclosers of this type can be co-ordinated more closely since there are no
inherent errors such as those that exist with electromechanical mechanisms (due to
overspeed, inertia, etc.). The downstream recloser must be faster than the upstream
recloser, and the clearance time of the downstream recloser plus its tolerance should
be lower than the upstream recloser clearance time less its tolerance. Normally, the
setting of the recloser at the substation is used to achieve at least one fast reclosure, in
order to clear temporary faults on the line between the substation and the load recloser.
The latter should be set with the same, or a larger, number of rapid operations as the
recloser at the substation. It should be noted that the criteria of spacing between the
time/current characteristics of electronically controlled reclosers are different to those
used for hydraulically controlled reclosers.
6.2.4 Recloser-relay co-ordination
Two factors should be taken into account for the co-ordination of these devices;
the interrupter opens the circuit some cycles after the associated relay trips, and the
relay has to integrate the clearance time of the recloser. The reset time of the relay
is normally long and, if the fault current is re-applied before the relay has com-
pletely reset, the relay will move towards its operating point from this partially reset
position.
For example, consider a recloser with two fast and two delayed sequence
with reclosing intervals of two seconds, which is required to co-ordinate with an
inverse time-delay overcurrent relay that takes 0.6 s to close its contacts at the fault
level under question, and 16 s to completely reset. The impulse margin time of
the relay is neglected for the sake of this illustration. The rapid operating time of
the recloser is 0.030 s, and the delayed operating time is 0.30 s. The percentage
of the relay operation during which each of the two rapid recloser openings takes
place is (0.03 s/0.6 s) ×100 per cent =5 per cent . The percentage of relay reset that
takes place during the recloser interval is (2 s/16 s) ×100 per cent =12.5 per cent.
Therefore, the relay completely resets after both of the two rapid openings of the
recloser.
The percentage of the relay operation during the first time-delay opening
of the recloser is (0.3 s/0.6 s) ×100 per cent =50 per cent. The relay reset
for the third opening of the recloser =12.5 per cent, as previously, so that
the net percentage of relay operation after the third opening of the recloser =
50 per cent −12.5 per cent =37.5 per cent . The percentage of the relay operation
during the second time delay opening of the recloser takes place =(0.3 sec./0.6 sec) ×
100 per cent =50 per cent , and the total percentage of the relay operation after the
fourth opening of the recloser =37.5 per cent +50 per cent =87.5 per cent.
From the above analysis it can be concluded that the relay does not reach
100 per cent operation by the time the final opening shot starts, and therefore
co-ordination is guaranteed.
Fuses, reclosers and sectionalisers 123
6.2.5 Recloser-sectionaliser co-ordination
Since the sectionalisers have no time/current operating characteristic, their co-
ordination does not require an analysis of these curves.
The co-ordination criteria in this case are based upon the number of operations
of the back-up recloser. These operations can be any combination of rapid or timed
shots as mentioned previously, for example two fast and two delayed. The section-
aliser should be set for one shot less than those of the recloser, for example three
disconnections in this case. If a permanent fault occurs beyond the sectionaliser, the
sectionaliser will open and isolate the fault after the third opening of the recloser. The
recloser will then re-energise the section to restore the circuit. If additional section-
alisers are installed in series, the furthest recloser should be adjusted for a smaller
number of counts. A fault beyond the last sectionaliser results in the operation of the
recloser and the start of the counters in all the sectionalisers.
6.2.6 Recloser-sectionaliser-fuse co-ordination
Each one of the devices should be adjusted in order to co-ordinate with the recloser. In
turn, the sequence of operation of the recloser should be adjusted in order to obtain the
appropriate co-ordination for faults beyond the fuse by following the criteria already
mentioned.
Example 6.1
Figure 6.7 shows a portion of a 13.2 kVdistribution feeder that is protected by a set of
overcurrent relays at the substation location. A recloser and a sectionaliser have been
installed downstream to improve the reliability of supply to customers. The recloser
chosen has two fast and two delayed operations with 90 cycles intervals.
The time/current curves for the transformer and branch fuses, the recloser and the
relays, are shown in Figure 6.8. For a fault at the distribution transformer, its fuse
should operate first, being backed up by the recloser fast operating shots. If the fault
is still not cleared, then the branch fuse should operate next followed by the delayed
opening shots of the recloser and finally by the operation of the feeder relay. The
sectionaliser will isolate the faulted section of the network after the full number of
counts has elapsed, leaving that part of the feeder upstream still in service.
As the nominal current of the 112.5 kVA distribution transformer at 13.2 kV is
4.9 A, a 6 T fuse was selected on the basis of allowing a 20 per cent overload. The
fast curve of the recloser was chosen with the help of the following expression based
on the criteria already given, which guarantees that it lies in between the curves of
both fuses:
t
recloser
×k ≤t
MMT of branch fuse
×0.75 (6.1)
where t
MMT of branch fuse
is the minimum melting time. The 0.75 factor is used in
order to guarantee the co-ordination of the branch and transformer fuses, as indicated
in Section 6.2.1.
At the branch fuse location the short-circuit current is 2224 A, which results in
operation of the branch fuse in 0.02 s. From Table 6.2, the k factor for two fast
124 Protection of electricity distribution networks
Sectionaliser
Transformer
Recloser
Fuse
5
2097 A
SC
Transf.
fuse
4
Branch
fuse
2224 A
SC
3 Operations
3
2
Recloser
2 fast
2 delayed
1
112.5 kVA
31993 A
SC
7066 A
SC
31993 A
SC
51
50
51
50
S
R
R S
Figure 6.7 Portion of a distribution feeder for Example 6.1
Fuses, reclosers and sectionalisers 125
LVtransf. relay
Branch fuse
Recloser fast
Recloser (delay)
Feeder
relay
Transf. fuse
T
i
m
e

i
n

s
e
c
o
n
d
s
3B
2
1
3A
Transf. fuse
Kearney T 6E
Recloser fast
I(Start) =240
I(Start) =1.5 ×Inom. load
Branch fuse
Kearney T 40E
4
5
3B
1
RTC
Recloser (delay)
I(Start) =240
I(Start) =1.5 ×Inom. load
3A
Feeder relay
Siemens 7SK-88 VI
0.30 TD
Tap 4.3=1.5 ×Icond ×
600/5 CT
2
5 4
LV Transf. relay
West CO-11
2.0 TD 6.0 Tap
2000/5CT
Inst. overridden
1
0.5 1 10
Current in Amperes ×10
100 1000 10000
0.01
0.1
10
1
100
1000
Figure 6.8 Phase-current curves for Example 6.1
operations and a reclosing time of 90 cycles is 1.35. With these values, from eqn. 6.1
the maximum time for the recloser operation is (0.02×0.75/1.35) =0.011s. This
time, and the pick-up current of the recloser, determines the fast curve of the recloser.
The feeder relay curve is selected so that it is above that of the delayed curve
of the recloser, and so that the relay reset time is considered. Detailed calculations
have not been given for this particular example since the procedure was indicated in
Chapter 5, but the curves of Figure 6.8 shows that adequate co-ordination has been
achieved.
Chapter 7
Directional overcurrent relays
Directional overcurrent protection is used when it is necessary to protect the sys-
tem against fault currents that could circulate in both directions through a system
element, and when bi-directional overcurrent protection could produce unnecessary
disconnection of circuits. This can happen in ring or mesh-type systems and in sys-
tems with a number of infeed points. The use of directional overcurrent relays in the
two situations is shown in Figure 7.1.
7.1 Construction
Directional overcurrent relays are constructed using a normal overcurrent unit plus
a unit that can determine the direction of the power flow in the associated distribution
system element. In addition to the relay current this second unit usually requires
a reference signal to measure the angle of the fault and thus determine whether or not
(a) (b)
Figure 7.1 Application of directional overcurrent relays: (a) ring system; (b) multi-
source system
128 Protection of electricity distribution networks
D OC
(b) (a)
D
D
OC
OC
OC
D
Figure 7.2 Obtaining the direction of power flow: (a) by supervision; (b) by control
the relay should operate. Generally, the reference or polarisation signal is a voltage
but this can also be a current input.
Basically, there are two methods for obtaining the direction of the power flow –
supervision and control; both cases are illustrated in Figure 7.2 where D indicates the
directional unit and OC the overcurrent unit. It is better to use the control system to
determine the direction of the power flow, since the overcurrent unit only picks up
when the flowis in the correct direction. With the supervision method, the overcurrent
unit can pick up for the wrong power flow direction. In addition, when a breaker is
opened in a ring system the current flows will change and this could lead to the
consequential possibility of loss of co-ordination.
7.2 Principle of operation
The operating torque can be defined by T =K
1

2
sin where
1
and
2
are the
polarising values,
1
being proportional to the current and
2
proportional to the
voltage, with the angle between
1
and
2
. The torque is positive if 0<<180

and negative if 180

<<360

. It should be noted that is in phase with I but
lagging with respect to the voltage since V =−(d)/dt .
If I and V are in phase, then the fluxes are out of phase by 90

. Therefore, the
angle for maximum torque occurs when the current and voltage of the relay are in
phase. This can be obtained very simply by using the current and voltage from the
same phase. However, this is not practical since, for a fault on one phase, the voltage
of that phase might collapse. It is, therefore, common practice to use the current from
a different phase.
7.3 Relay connections
The connection of a directional relay is defined on the basis of the degrees by which
a current at unity power factor leads the polarisation voltage. The angle of maximum
torque, AMT, is the angle for which this displacement produces the maximum torque
and therefore is always aligned with the polarisation voltage.
Directional overcurrent relays 129
V
c
V
b
Zero torque
line
I
a
V
a
V
ac
AMT
60°
30°
Figure 7.3 Vector diagram for 30

connection (0

AMT)
7.3.1 30

connection (0

AMT)
Feeding the relays:

A
: I
a
,
B
: I
b
,
C
: I
c
V
ac
V
ba
V
cb
Maximum torque: when the phase current lags the phase-neutral voltage by 30

.
Angle of operation: current angles from 60

leading to 120

lagging.
Use: this type of connection should always be used on feeders, provided that it has
three elements, i.e. one per phase, since two phase elements and one earth element
can give rise to poor operation (see Figure 7.3). The three-phase unit arrangement
should not be used in transformer circuits where some faults can result in a reverse
current flow in one or more phases leading to relay mal-operation.
7.3.2 60

connection (0

AMT)
Feeding the relays:

A
: I
ab
,
B
: I
bc
,
C
: I
ca
V
ac
V
ba
V
cb
Maximum torque: when the current lags the phase-to-neutral voltage by 60

. I
ab
lags
V
ac
by 60

. I
a
lags V
a
by 60

at unity power factor.
Angle of operation: current I
ab
from 30

leading to 150

lagging, or I
a
leading 30

or lagging 150

at unity power factor.
Use: it is recommended that relays with this connection be used solely on feeders (see
Figure 7.4). However, they have the disadvantage that the CTs have to be connected in
delta. For this reason, and because they offer no advantages compared to the previous
case, they are little used.
130 Protection of electricity distribution networks
I
ab
Zero torque
line
AMT
V
ac
V
a
V
b
V
c
60°
30°
Figure 7.4 Vector diagram for 60

connection (0

AMT)
AMT
Zero torque
line
I
ab
V
c
V
b
V
a

bc
V
bc
30°
30°
30°
30°
Figure 7.5 Vector diagram for 90

connection (30

AMT)
7.3.3 90

connection (30

AMT)
Feeding the relays:

A
: I
a
,
B
: I
b
;
C
: I
c
V
bc
+30

V
ca
+30

V
ab
+30

Maximum torque: when the current lags the phase-to-neutral voltage by 60

.
Angle of operation: current angles from 30

leading to 150

lagging.
Use: in feeders where the source of zero-sequence components is behind the point of
connection of the relay (see Figure 7.5).
Directional overcurrent relays 131
Zero torque
line
I
a
AMT
V
a
V
b
V
c
bc

V
bc
45°
15°
30°
Figure 7.6 Vector diagram for 90

connection (45

AMT)
7.3.4 90

connection (45

AMT)
Feeding the relays:

A
: I
a
,
B
: I
b
,
C
: I
c
,
V
bc
+45

V
ca
+45

V
ab
+45

Maximum torque: when the current lags the phase-to-neutral voltage by 45

.
Angle of operation: current angles from 45

leading to 135

lagging.
Use: this arrangement is recommended for protecting transformers or feeders that
have a source of zero-sequence components in front of the relay (see Figure 7.6). This
connection is essential in the case of transformers in parallel, or transformer feeders,
especially for guaranteeing correct operation of the relays for faults beyond Yd trans-
formers. This configuration should always be used when single-phase directional
relays are applied to circuits that can have a 2-1-1 current distribution.
7.4 Directional earth-fault relays
Directional earth-fault relays are constructed on the basis that the residual voltage
is equal to three times the zero-sequence voltage drop in the source impedance, and
displaced with respect to the residual current for the angle characteristic of the source
impedance. When a set of suitable VTs is not available for obtaining the polarisation
voltage, current polarisation is employed using the earth current from a local trans-
former connected to earth. This is based on the principle that the neutral current always
flows towards the system from earth whereas, depending on the fault, the residual
current may flow in any direction. It should be stressed, however, that the possibility
of the failure of a voltage-polarised directional protection relay is minimal and it is
therefore recommended that this arrangement should be used wherever possible.
132 Protection of electricity distribution networks
67
115000/ √3
115/13.2 kV
500/5
T1
13.2/115 kV
A
T2
F
B
G
L
T2
X
2
=0.3
X
0
=0.3
X
1
=0.3
L
X
2
=0.4
X
0
=1.2
X
1
=0.4
T1
X
2
=0.11
X
0
=0.11
X
1
=0.11
G
X
2
=0.15
X
0
=0.09
X
1
=0.15
115√3
Figure 7.7 Single line diagram of system for Example 7.1
Example 7.1
A solid earth-fault on phases B and C is represented by the arrow at the point F in the
power system in Figure 7.7. Determine the current and voltage signals (in amps and
volts) that go into each one of the directional relays that have a 30

connection and
are fed as indicated below:

A
: I
A
,
B
: I
B
,
C
: I
C
V
AC
V
BA
V
CB
In addition, indicate which relays operate on the occurrence of the fault. In the
solution, ignore load currents and assume a pre-fault voltage equal to 1 p.u. The bases
at the generator location are 13.2 kV and 100 MVA.
N.B. Although the system is radial, the installation of a directional overcurrent
relay is justified by assuming that this circuit would be part of a ring in a future
system.
Solution
The conditions for a double phase-to-earth fault, B-C-N, are
I
A
=0, V
B
=0, V
C
=0
The three sequence networks are shown in Figure 7.8. The equivalent circuit is
obtained by connecting the three sequence networks in parallel as shown in Figure 7.9.
From Figure 7.8
Z
0
=0.111.5⇒Z
0
=0.102
Directional overcurrent relays 133
j0.4
j0.4
j1.2 j0.09 j0.11
F
j0.15
A
A
j0.11
F
1 p.u.
j0.15
A
j0.11
F
j0.3
B
j0.3
B
j0.3
B
Figure 7.8 Sequence networks for Example 7.1
1 p.u.
j0.26 j0.26
I
A
1
I
A
2
I
A
0
j0.102
Figure 7.9 Equivalent circuit for Figure 7.8
so that the three sequence currents in the network are
I
A1
=
1
( j0.26+( j0.26×j0.102)/( j0.362))
=−j3.0p.u.
I
A2
=−(−j3.0)
j0.102
j0.26+j0.102
=j0.845
I
A0
=−(−j3.0)
j0.26
j0.26+j0.102
=j2.155
134 Protection of electricity distribution networks
At the point of the fault:
I
A
=I
A1
+I
A2
+I
A0
=−j3.0+j0.845+j2.155
⇒I
A
=0 as was to be expected for a B-C-N fault
I
B
=a
2
I
A1
+aI
A2
+I
A0
=1

240

(−j3.0) +1

120

(j0.845) +j2.155
=−3.33+j3.2315=4.64

135.86

I
C
=aI
A1
+a
2
I
A2
+I
A0
=1

120

(−j3.0) +1

240

(j0.845) +j2.155
=3.33+j3.2315=4.64

44.14

At the point of fault on the network (not at the relay):
V
A1
=V
A2
=V
A0
=−I
A2
( j0.26) =0.220
V
A
=V
A1
+V
A2
+V
A0
=3V
A1
=3×0.22=0.66
V
B
=V
C
=0
V
AC
=V
A
−V
C
=V
A
−0=0.66
V
BA
=V
B
−V
A
=−V
A
=−0.66
V
CB
=V
C
−V
B
=0
The bases at the point of fault are
V =115kV; P =100MVA
I
Base
=
P

3V
=
100×10
6

3×115×10
3
=502.04A
Therefore, the values at the point of the fault are
I
A
=0
V
AC
=0.66×
115000

3
×
115

3
×

3
115000
=43.82

0

V
I
B
=4.64

135.86

×502.04×(5/500) =23.29

135.86

A
V
BA
=−0.66×
115000

3
×
1
1000
=43.82

180

V
I
C
=4.64

44.14

×502.04×(5/500) =23.29

44.14

A
V
CB
=0
Directional overcurrent relays 135
Corollary
I

=
1
j0.26
=−j3.48 I

=I
A1
+I
A2
+I
A0
=3I
A1
and
I
A1
=
1
j0.26+j0.26+j0.102
=−j1.6077
so that
I

=3×(−j1.6077) =−j4.823
At the point where the relay is located there will be equal positive- and negative-
sequence current values, as there are at the fault point; however, the zero-sequence
current in the relay itself is different because of the division of current.
In the relay, I
AB
=j2.154(1.5/1.61) =j2 p.u., so that at the relay
I
A
=I
A1
+I
A2
+I
A0
=−j3.0+j0.845+j2=−j0.155
and it will be seen that, in this case, I
A
=0:
I
B
=a
2
I
A1
+aI
A2
+I
A0
=1

240

(−j3.0) +1

120

(j0.845) +j2
=3.0

150

+0.845

210

+2.0

90

=−3.33+j3.077=4.534

137.26

I
C
=aI
A1
+a
2
I
A2
+I
A0
=1

120

(−j3.0) +1

240

(j0.845) +j2
=3.0

30

+0.845

−30

+2.0

90

=3.33+j3.077=4.534

42.73

V
A1
=V
A2
=I
A2
(j0.26) =0.845×0.26=0.22
V
A0
=−I
A0
(j0.11) =−j2.0(j0.11) =0.22
V
A
=V
A1
+V
A2
+V
A0
=3V
A1
=3×0.22=0.66
V
B
=V
C
=0
The CT is fed from the same fault point, so that
V
AC
=V
A
−V
C
=V
A
−0=0.66
V
BA
=V
B
−V
A
=−V
A
=−0.66
V
CB
=V
C
−V
B
=0
136 Protection of electricity distribution networks
The signals that feed the relay are

A
I
A
=0.155

−90

×502.04×(5/500) =0.728

−90

A
V
AC
=0.66×
115000

3
×
115

3
115000/

3
=43.82

0

V

B
I
B
=4.534

137.26

×502.04×(5/500) =22.76

137.26

A
V
BA
=−0.66×
115000

3
×
1
1000
=43.82

180

V

C
I
C
=4.534

42.73

×502.04×(5/500) =22.76

42.73

A
V
BC
=0
Analysis of operation of directional relays:
Polarisation:

A

B

C
I
A
I
B
I
C
V
AC
V
BA
V
CB
Phase A relay:
I
A
=0.728

−90

A
V
AC
=43.82

−0

V
For operation, −90

< angle of I
A
<90

. The relay in phase A is at the limit
of functioning (see Figure 7.10) thus creating some doubt about the operation of its
directional unit.
I
R
–90˚
+90˚
V
RT
Figure 7.10 Analysis of operation of relay for phase A
Directional overcurrent relays 137
V
SR
270°
180°
137.26°
I
S
90°
Figure 7.11 Analysis of operation of relay for phase B
Phase B relay:
I
B
=22.76

137.26

A
V
BA
=43.82

180

V
For operation, 90

<angle of I
B
<270

. The relay in phase B operates, since the
angle of I
B
is 137.26

; see Figure 7.11.
Phase C relay: this does not operate because V
CB
=0.
7.5 Co-ordination of instantaneous units
The calculations for the setting of an instantaneous unit in a ring system are carried
out using the short-circuit level at the next relay downstream, with the ring open,
multiplied by a safety overload factor in order to maintain co-ordination, taking into
account the DCtransient component of the current. The criterion used here is the same
as for setting bi-directional overcurrent relays protecting lines between substations.
When the ring has only one source, the relays installed at substations adjacent to the
source substation should never register any current from the substation towards the
source. Therefore, it is recommended that, for these relays, the instantaneous units
should be set at 1.5 times the maximumload current. Alower value should not be used
as this could result in a false trip if the directional element picks up inadvertently under
severe load transfer conditions. With these settings the instantaneous units then have
the same pick-up current value as that of the time-delay units so that co-ordination is
not compromised.
Example 7.2
For the system shown in Figure 7.12 determine the maximum load currents, the
transformation ratios of the CTs and the current setting of the instantaneous units
138 Protection of electricity distribution networks
3 MVA
7 MVA
150 MVA sc
3
4
K
2 Ω
2
5MVA
J
2 Ω
1 A
I
5
4 Ω
6
3 Ω
13.2 kV
8 B
7
L
Figure 7.12 Single-line diagram for a ring system with one infeed for Example 7.2
to guarantee a co-ordinated protection scheme. The instantaneous settings should be
given in primary amperes, since reference to specific relay models is not given.
Calculation of the maximum load currents
With the ring open at A:
total load current flowing through B=(7+3+5MVA)/

3(13200V) Relays
=(15MVA)/

3(13200V) =656.08A (7) (8)
total load current from L to K =(8MVA)/

3(13200V) =349.9A (5) (6)
total load current from K to J =(5MVA)/

3(13200V) =218.69A (3) (4)
total load current flowing from J =0A (1) (2)
With the ring open at B:
total load current flowing through A=(15MVA)/

3(13200V) =656.08A (1) (2)
total load current from J to K =(10MVA)/

3(13200V) =437.38A (3) (4)
total load current from K to L =(7MVA)/

3(13200V) =306.17A (5) (6)
total load flowing from L =0A (7) (8)
Selection of CTs
It is assumed that the available CTs have primary turns in multiples of 100 up to 600,
and thereafter in multiples of 200. The CT ratios are calculated for the maximumload
conditions first. Thus, the CT ratios selected are:
I
max 1
=656.08A CT ratio=800/5
I
max 2
=656.08A CT ratio=800/5
I
max 3
=437.38A CT ratio=500/5
I
max 4
=437.38A CT ratio=500/5
I
max 5
=349.90A CT ratio=400/5
Directional overcurrent relays 139
I
max 6
=349.90A CT ratio=400/5
I
max 7
=656.08A CT ratio=800/5
I
max 8
=656.08A CT ratio=800/5
In order to confirmthis selection, it is necessary to check whether or not saturation
is present at the maximumfault level at each breaker, using the above CT ratios. With
the most critical fault values, as calculated in the following section, it can be shown
that no CT is saturated since the value of 0.05×I
sc
is well below the number of
primary turns for each one.
Calculation of fault currents
Busbar I
I
sc
=(150MVA)/

3(13200V) =6560.8A
Z
source
=(13200V)
2
/(150MVA) =1.16
Busbar J
The equivalent circuit for a fault at J is shown in Figure 7.13:
I
scJ
=
13200

3(1.16+1.64)
=2721.79A, with the ring closed.
Apportioning the current flows in the inverse proportion of the circuit impedances:
I
scJ
(on the right hand side) =2721.79×(2÷11) =494.87A
I
scJ
(on the left hand side) =2721.79×(9÷11) =2226.92A
With breaker A open:
I
scJ
=
13200

3(1.16+9)
=750.1A
13.2
kV
√3
13.2
kV
√3
1.16 Ω 1.64 Ω 1.16Ω 9Ω

I
Figure 7.13 Equivalent circuit for a fault at J (Example 7.2)
140 Protection of electricity distribution networks
13.2
kV
I
√3
13.2
kV
√3
1.16Ω 2.54Ω 1.16 Ω 7 Ω
4 Ω
Figure 7.14 Equivalent circuit for a fault at K (Example 7.2)
With breaker B open:
I
scJ
=
13200

3(1.16+2)
=2411.71A
Busbar K
The equivalent circuit for a fault at K is shown in Figure 7.14:
I
scK
=
13200

3(1.16+2.54)
=2059.74A, with the ring closed.
I
scK
(on the right hand side) =2059.74×(4÷11) =748.99A
I
scK
(on the left hand side) =2059.74×(4÷11) =1310.74A
With breaker A open:
I
scK
=
13200

3(1.16+7)
=933.94A
With breaker B open:
I
scK
=
13200

3(1.16+4)
=1476.94A
Busbar L
The equivalent circuit for a fault at L is shown in Figure 7.15:
I
scL
=
13200

3(1.16+2.18)
=2281.74A, with the ring closed.
I
scL
(on the right hand side) =2281.74×(8÷11) =1659.45A
I
scL
(on the left hand side) =2281.74×(3÷11) =622.29A
With breaker A open:
I
scL
=
13200

3(1.16+3)
=1831.97A
Directional overcurrent relays 141
13.2
kV
√3
13.2
kV
√3
1.16Ω 2.18Ω 1.16Ω 3Ω

I
Figure 7.15 Equivalent circuit for a fault at L (Example 7.2)
With breaker B open:
I
scL
=
13200

3(1.16+8)
=831.99A
Setting of instantaneous unit
Clockwise direction:
Relay 7: 1.50 × 656.08 A = 984.12 A
Relay 5: 1.25 × 831.99 A = 1039.99 A
Relay 3: 1.25 × 1476.94 A = 1846.17 A
Relay 1: 1.25 × 2411.71 A = 3014.64 A
Anticlockwise direction:
Relay 2: 1.50 × 656.08 A = 984.12 A
Relay 4: 1.25 × 750.10 A = 937.62 A
Relay 6: 1.25 × 933.94 A = 1167.42 A
Relay 8: 1.25 × 1831.97 A = 2289.96 A
7.6 Setting of time-delay directional overcurrent units
As in the case of bi-directional overcurrent relays, the time-delay units of directional
overcurrent relays in a ring are set by selecting suitable values of the pick-up current
and time dial settings. The procedure for each one is indicated in the following
sections.
7.6.1 Pick-up setting
The pick-up setting of a directional overcurrent relay is calculated by considering the
maximum load transfer that can be seen by the relay in any direction, multiplied by
142 Protection of electricity distribution networks
the overload factor referred to in Section 5.3.3. The load transfer in both directions
is taken into account to avoid the possibility of relay mal-operation if the directional
unit is incorrectly activated by the wrong polarisation, especially under heavy transfer
conditions.
7.6.2 Time dial setting
The time dial setting can be defined by means of two procedures. The first one is based
on instantaneous setting values, whereas the second takes account of contact travel
and is more rigorous since it requires fault calculations for various ring topologies.
However, it has to be emphasised that both methods guarantee proper co-ordination
although the first one can produce slightly higher time dial values and is more used
in simple systems or when the locations of the co-ordination curves are not critical.
Time dial setting by direct method
The setting of time dial units by the direct method is based on the fault values used
to set the instantaneous units. As in the case of bi-directional relays, the time dial
value is adjusted so that, taking the instantaneous current setting given to the relay
downstream, its operating time is above that of the downstream relay by the required
discrimination time margin. This procedure should be carried out for all the relays
on the ring, both clockwise and anticlockwise, normally starting from the relays
associated with the main source busbar. The application of this method is illustrated
in the instantaneous settings calculated for the relays in Example 7.2.
Time dial setting considering contact travel
The time dial setting of directional relays, taking account of the contact travel of the
timer units, requires an iterative process as detailed below:
1. Determine the initial time dial values of the relays on the ring such that
co-ordination is guaranteed with the relays associated with the lines and machines
fed by the adjacent busbar in the direction of trip.
2. Calculate the time required for the first relay to operate for a fault at its associated
breaker terminals, with the ring closed. Any relay can be chosen as the first,
although it is usual to take one of the relays associated with equipment connected
to the main source busbar. For this condition a check should be made to ensure
that there is adequate discrimination between the chosen relay and the back-up
relays at adjacent substations. If not, then the time dial values of the relays at the
adjacent substations should be modified. In addition, the operating time of the
relay on the breaker at the opposite end of the line should be calculated as well
as the times for its back-up relays.
3. Next, consider a fault at the opposite end of the line with the ring open and, for
this condition, calculate the operating time of the relay closest to the fault and
check that there is adequate discrimination between it and the back-up relays at
the adjacent substations. As in the previous case, if co-ordination is not achieved
then the time dial values should be increased. For this case it is important to take
Directional overcurrent relays 143
into account the contact travel during the fault before the ring is opened by the
operation of the first relay. To do this, the following expressions should be used:
t
relay next to fault
=t
adjacent relay with ring closed
+t
relay next to fault with ring open

1−
t
adjacent relay with ring closed
t
relay next to fault with ring closed

and
t
back-up relay
=t
adjacent relay with ring closed
+t
back-up relay with ring open

1−
t
remote relay with ring closed
t
back-up relay with ring closed

t
back-up relay
≥t
relay next to fault
+t
discrimination margin
4. The same procedure is repeated for each relay, i.e. by considering a fault at
its associated breaker terminals with the ring closed, and then for a fault at the
opposite end of the line with the ring open. The procedure is completed when no
further time dial setting changes are required.
It should be noted that calculating the time dial setting based on contact travel
guarantees proper co-ordination of the relays on a ring, since the setting is carried
out for the most severe conditions, i.e. for a fault at the various busbars with the ring
closed, and with the ring open. To illustrate the above procedure, consider the system
shown in Figure 7.16, with a 34.5 kV ring connecting three busbars.
After setting the pick-up and instantaneous current values, the steps to set the
time dial values by considering the travel contact are as indicated below:
1. The time dial values of the relays are initially set in such a way that co-ordination
is guaranteed with those relays associated with lines or machines supplied from
the three busbars of the ring.
2. A fault at the breaker terminals associated with relay R
21
with the ring closed is
considered first. FromFigure 7.17a, the three-phase short-circuit level is 5157 A.
3. The operating times of relays R
13
and R
14
should be checked to ensure that there
is adequate co-ordination with relay R
21
. If not, their time dial settings should
be increased.
4. The operating times of relay R
11
and its back-up R
22
are calculated to ensure that
the specified fault will be cleared with the ring closed.
5. The short-circuit current value for a fault at the breaker terminals associated with
relay R
11
is calculated, with the ring open. From Figure 7.17b, this value is
2471 A.
6. The operating times of relay R
11
and its back-up relay R
22
are calculated for this
condition with the following expressions:
t

relay R
11
next to fault
=t
R
21
adjacent relay with ring closed
+t

R
11
relay next to fault with ring open

1−
t
R
21
adjacent relay with ring closed
t
R
11
relay next to fault with ring closed

144 Protection of electricity distribution networks
400/5
34.5 kV
800/5
C
R
11 800/5
R
12 600/5
D
4 km
R
22 R
13
800/5
34.5 kV
34.5 kV
7 km
B
R
21
800/5
R
14
600/5
R
23
5.7 km
110/34.5/11.67 kV
35 MVA
YNynd
Zps:9.62%
1400/5
A
R
24
115 kV
Figure 7.16 Ring systemto illustrate overcurrent directional relay setting procedure
and
t

R
22
back-up relay
=t
R
21
remote relay with ring closed
+t

R
22
back-up relay with ring open

1−
t
R
21
remote relay with ring closed
t
R
22
back-up relay with ring closed

where: t =operating time of the relay for the initial fault with the ring closed;
t

=operating time of the relay with the newtopology after the first relay operates;
t

=operating time of the relay considering the newtopology and taking account
of the contact travel.
7. Finally, the time dial setting of relay R
22
should be checked to confirm that it
satisfies the following expression:
t

R
22
≥t

R
11
+t
discrim.margin
This procedure is then repeated for the rest of the relays of the ring. Table 7.1
summarises the steps as a guideline to completing this co-ordination exercise.
Directional overcurrent relays 145
R
22
R
13
R
11
R
12
R
11
C
R
12
34.5 kV
2471 A
R
21
D
R
22
34.5 kV
R
13
R
23
34.5 kV
B
R
14
A
R
24
110/34.5/11.67 kV
35 MVA
YNynd
Zps:9.62%
115 kV
C
34.5 kV
D
34.5 kV
B
R
21
5157 A
34.5 kV
R
14
R
23
A
(a)
(b)
R
24
115 kV
110/34.5/11.67 kV
35 MVA
YNynd
Zps:9.62%
Indicates the location of the
opening of the line
Figure 7.17 Short-circuit currents for faults at breaker: (a) fault associated with R
21
with ring closed; (b) fault associated with R
11
with ring open at R
21
146 Protection of electricity distribution networks
Table 7.1 Summary of the procedures for the time dial
settings for ring system in Figure 7.16
Fault at Topology of
the ring
Calculation Co-ordination
check
B
21
closed t
R
21
, t
R
11
, t
R
22
t
R
13
, t
R
14
B
11
open t

R
11
, t

R
22
, t

R
11
t

R
22
B
12
closed t
R
12
, t
R
22
, t
R
23
t
R
21
B
22
open t

R
22
, t

R
23
, t

R
22
t

R
23
B
13
closed t
R
13
, t
R
23
, t
R
14
, t
R
11
t
R
12
B
23
open t

R
23
, t

R
14
, t

R
11
, t

R
23
t

R
14
, t

R
11
B
11
closed t
R
11
, t
R
21
, t
R
14
, t
R
13
t
R
22
B
21
open t

R
21
, t

R
14
, t

R
13
, t

R
21
t

R
13
, t

R
14
B
22
closed t
R
22
, t
R
12
, t
R
21
t
R
23
B
12
open t

R
12
, t

R
21
, t

R
12
t

R
21
B
23
closed t
R
23
, t
R
13
, t
R
12
t
R
14
, t
R
11
B
13
open t

R
13
, t

R
12
, t

R
13
t

R
12
5 4

1
B
3

2
C
6
A

150MVA
SC
Z
SOURCE
=1.16 Ω
Figure 7.18 Network for Exercise 7.1
7.7 Exercises
Exercise 7.1
The following short-circuit data refer to the ring system in Figure 7.18:
With the ring closed:
fault at A=6, 560A
fault at B=2, 731A 2, 235A (contribution from 1) plus 496 A from 2
fault at C=2, 280A 622A (from 1) +1, 658A (from 2)
Directional overcurrent relays 147
F
3
=
2
3
0
0
A
F
4
=
2
1
6
0
A
B
F
4
F
5
F
1
=
1
6
9
0
A
F
2
=
1
3
5
0
A
F
2
A
F
3
115 kV
F
6
=
4
4
4
0
A
F
5
=
2
8
8
0
A
F
6
C
34.5 kV
F
1
Indicates the location of the
opening of the line
Figure 7.19 Network for Exercise 7.2. F
1
, F
3
and F
5
correspond to faults with
the ring open at the ends indicated by the symbol
• •
. F
2
, F
4
and F
6
correspond to faults with the ring closed
With the ring open:
fault at B=750A (with 1 open)
=2411A (with 2 open)
fault at C=1832A (with 1 open)
=832A (with 2 open)
If the instantaneous element 6 at substation C is set to 424 A and protects up to the
middle of the line A-C with the ring closed, calculate the setting for the instantaneous
element of relay 4 at substation A.
Exercise 7.2
The short-circuit levels for different three-phase fault conditions on a 34.5 kV ring
system are shown in Figure 7.19. The fault currents F
1
, F
3
and F
5
take account of
those lines that are open at the points indicated for each fault, and assume that each
line is already open at the time the fault occurs.
Determine the relaytime dial andinstantaneous current settings for the overcurrent
relays associated with breakers A, B and C. The tap of each relay was calculated
beforehand and a value of 5 was selected for the three relays.
148 Protection of electricity distribution networks
Use the characteristic curves of the inverse-time relay, shown in Figure 5.15. The
relay has the following characteristics:
Pick-up: 1 to 12 A in steps of 1 A
Time dial: as in the diagram
Instantaneous settings: 6 to 80 A in steps of 1 A
Chapter 8
Differential protection
8.1 General
Differential protection functions when the vector difference of two or more similar
electrical magnitudes exceeds a predetermined value. Almost any type of relay can
function as differential protection – it is not so much the construction of the relay
that is important but rather its method of connection in the circuit. The majority
of the applications of differential relays are of the current-differential type, but they
can also be of the voltage-differential type, operating on the same principle as the
current relays; the difference lies in the fact that the operating signal is derived from
a voltage across a shunt resistance.
A simple example of a differential arrangement is shown in Figure 8.1. The
secondaries of the current transformers (CTs) are interconnected and the coil of an
overcurrent relay is connected across these. Although the currents I
1
and I
2
may
be different, provided that both sets of CTs have appropriate ratios and connections
then, under normal load conditions or when there is a fault outside the protection
I
1
I
2
CT
1
CT
2
Protected
element
K
1
I
1
K
2
I
2
Figure 8.1 Differential protection – current balance
150 Protection of electricity distribution networks
zone of the element, the secondary currents will circulate between the two CTs and
will not flow through the overcurrent relay. However, if a fault occurs in the section
between the two CTs the fault current would flow towards the short-circuit from
both sides and the sum of the secondary currents would flow through the differential
relay. In all cases the current in the differential relay would be proportional to the
vector difference between the currents that enter and leave the protected element; if
the current through the differential relay exceeds the setting value then the relay will
operate.
One arrangement that is extensively used is the differential relay with a variable-
percentage characteristic, alternatively described as being percentage biased, which
has an additional unit, the restraining coil, in addition to the operating coil as shown in
Figure 8.2. Since this type of relay is most common in differential relaying schemes,
all references to differential relays in this chapter will be on the basis of using this type
of relay. The current in the operating coil is proportional to (I
1
−I
2
). If N is equal
to the number of turns of the restraining coil with the operating coil connected to the
mid-point of the restraining coil, then the total ampere-turns are equal to I
1
(N/2)
plus I
2
(N/2), which is the same as if (I
1
+I
2
)/2 flowed through all of the restrain-
ing coil. The operating characteristic of the relay with this form of restraint is shown
in Figure 8.3.
CTs, although produced to the same specification, will not have identical sec-
ondary currents for the same primary currents because of slight differences in their
magnetising characteristics. The relay restraining force increases with the magnitude
of (I
1
+I
2
), thus preventing unnecessary tripping due to any CT unbalance errors. In
addition, the restraining torque is increased in the presence of through-fault currents
producing a more stable operating characteristic and preventing relay mal-operation.
In relays that have variable tappings in the restraint coil circuits, the tappings can be
set to balance out any currents due to differences in the CTs. If the relays do not have
these variable tappings, then the currents leaving the CTs should match closely in
order to avoid mal-operation of the relays.
I
1
I
1
I
2
N
R
N
R1
N
R2
N
op
I
2
CT
1
CT
2
Protected
element
N
R
Restraint coil
N
op
Operating coil
Figure 8.2 Differential relay with variable-percentage characteristic
Differential protection 151
(I
1
+I
2
)/2
No operation
Operation
=
N
op
I
res
N
R
I
op
m =
I
1
– I
2
Figure 8.3 Relay characteristic (variable-percentage type)
Restraint coil
Element with 3 terminals
Operating coil
Figure 8.4 Differential protection for element with three terminals
Differential relays can be used for power system elements that have more than
two terminals, as shown in Figure 8.4. Each of the three restraining coils has the same
number of turns and each coil produces a restraint that is independent of the others;
these add arithmetically. The slope of the operating characteristic of each relay varies
depending on the current distribution in the three restraining coils. There are also
other types of differential relays which use directional or overvoltage elements in
place of the overcurrent elements. Thus, all types are extensions of the fundamental
principles that have been described above.
152 Protection of electricity distribution networks
8.2 Classification of differential protection
Differential protection can be classified according to the type of element to be
protected, as follows:
• transformers;
• generators and rotating machines;
• lines and busbars.
8.3 Transformer differential protection
A differential system can protect a transformer effectively because of the inherent
reliability of the relays, which are highly efficient in operation, and the fact that
equivalent ampere-turns are developed in the primary and secondary windings of
the transformer. The CTs on the primary and secondary sides of the transformer are
connected in such a way that they form a circulating current system, as illustrated
in Figure 8.5. Faults on the terminals or in the windings are within the transformer
protection zone and should be cleared as quickly as possible in order to avoid internal
stress and the danger of fire. The majority of internal faults that occur in the windings
are to earth (across to the core) or between turns, with the severity depending on the
design of the transformer and the type of earthing.
Differential protection can also detect and clear insulation faults in the transformer
windings. The principal cause of these faults is arcing inside the bushings and faults in
the tapchanger. This type of protectionnot onlyresponds tophase-to-phase andphase-
to-earth faults but also in some degree to faults between turns. However, phase-to-
phase faults between the windings of a three-phase transformer are less common. An
internal fault that does not constitute an immediate danger is designated an incipient
fault and, if not detected in time, could degenerate into a major fault. The main faults
in this group are core faults, caused by the deterioration of the insulation between the
laminations that make up the core.
87 Differential relay
I
2
I
1
I
R1
I
R2
Figure 8.5 Transformer differential protection
Differential protection 153
8.3.1 Basic considerations
In order to apply the principles of differential protection to three-phase transformers,
the following factors should be taken into account:
Transformation ratio
The nominal currents in the primary and secondary sides of the transformer vary in
inverse ratio to the corresponding voltages. This should be compensated for by using
different transformation ratios for the CTs on the primary and secondary sides of the
transformer.
Transformer connections
When a transformer is connected in star/delta, the secondary current has a phase shift
of a multiple of 30

relative to the primary, depending on the vector group. This shift
can be offset by suitable secondary CT connections. Furthermore, the zero-sequence
current that flows in the star side of the transformer will not induce current in the
delta winding on the other side. The zero-sequence current can therefore be eliminated
from the star side by connecting the CTs in delta. For the same reason, the CTs on the
delta side of the transformer should be connected in star. When CTs are connected in
delta, their nominal secondary values should be multiplied by

3 so that the currents
flowing in the delta are balanced by the secondary currents of the CTs connected
in star.
Tap changer
If the transformer has the benefit of a tap changer it is possible to vary its transfor-
mation ratio, and any differential protection system should be able to cope with this
variation. Since it is not practical to vary the CT transformation ratios, the differential
protection should have a suitable tolerance range in order to be able to modify the
sensitivity of its response of operation. For this reason it is necessary to include some
form of biasing in the protection system together with some identifying markings of
the higher current input terminals.
Magnetisation inrush
This phenomenon occurs when a transformer is energised, or when the primary volt-
age returns to its normal value after the clearance of an external fault. The magnetising
inrush produces a current flowinto the primary winding that does not have any equiva-
lent in the secondary winding. The net effect is thus similar to the situation when there
is an internal fault on the transformer. Since the differential relay sees the magnetis-
ing current as an internal fault, it is necessary to have some method of distinguishing
between the magnetising current and the fault current. These methods include:
1. Using a differential relay with a suitable sensitivity to cope with the magnetising
current, usually obtained by a unit that introduces a time delay to cover the period
of the initial inrush peak.
154 Protection of electricity distribution networks
2. Using a harmonic-restraint unit, or a supervisory unit, in conjunction with a
differential unit.
3. Inhibiting the differential relay during the energising of the transformer.
8.3.2 Selection and connection of CTs
The following factors should be taken into account when considering the application
of differential protection systems:
1. In general, the CTs on the star side of a star/delta transformer should be con-
nected in delta, and those on the delta side should be connected in star. This
arrangement compensates for the phase shift across the transformer and blocks
the zero-sequence current in the event of external faults to earth.
2. The relays should be connected to accept the load current entering one side of the
transformer and leaving by the other side. If there are more than two windings it
is necessary to consider all combinations, taking two windings at a time.
3. The CT ratios should be selected in order to produce the maximum possible
balance between the secondary currents of both sides of the transformer under
maximumload conditions. If there are more than two windings, all combinations
should be considered, taking two windings at a time and the nominal power of the
primary winding. If the available CT ratios do not enable adequate compensation
to be made for any variation in secondary current from CTs, then compensation
transformers can be used to offset the phase shift across the transformer.
The following examples show the connections of the CTs, the calculation of
their transformation ratios, and the connection of the differential relays as applied to
transformer protection schemes.
Example 8.1
Consider a 30 MVA, 11.5/69 kV, Yd1 transformer as shown in the single-line diagram
in Figure 8.6. Determine the transformation ratio and connections of the CTs required
in order to set the differential relays. CTs with ratios in steps of 50/5 up to 250/5,
and in steps of 100/5 thereafter, should be used. Use relays with a variable-percentage
87
3
152
30 MVA
11.5/69 kV
Yd1
252
Figure 8.6 Single-line diagram for Example 8.1
Differential protection 155
characteristic. The available current taps are: 5.0-5.0, 5.0-5.5, 5.0-6.0, 5.0-6.6,
5.0-7.3, 5.0-8.0, 5.0-9.0, and 5.0-10.0 A.
Solution
Figure 8.7 shows the complete schematic of the three-phase connections. The currents
in the windings and in the lines are drawn and show that the restraint currents on the
star and delta sides of the relay are in phase.
For a throughput of 30 MVA the load currents are
I
load
(69kV) =
30MVA

3×69kV
=251.0A
I
load
(11.5kV) =
30MVA

3×11.5kV
=1506.13A
In order to increase the sensitivity, the CT ratio at 11.5 kV is selected as close
as possible to the maximum load current and, therefore, a CT ratio (11.5 kV) of
R R OP
R
R
30 MVA
Yd1
R
R
OP
OP
I
A
I
A
I
B
I
C
I
A
I
B
I
C
I
a
I
b
I
c
I
B
I
C
A
B
C
1500/5 115 kV 69 kV
I
A
– I
C
I
a
– I
c
I
b
– I
a
I
c
– I
b
I
a
–I
c
I
b
– I
a
I
c
–I
b
I
a
– I
c
I
b
– I
a
I
c
– I
b
I
B
– I
A
I
C
– I
B
250/5
Figure 8.7 Three-phase connection diagram for Example 8.1
156 Protection of electricity distribution networks
Compensation
transformer
87
100/5
Dy1
12.5 MVA
115/13.2 kV
600/5
Figure 8.8 Single-line diagram for Example 8.2
1500/5 is chosen. When calculating the other CT ratio, a balance of currents has to
be achieved, i.e. 1506.13×(5/1500) ×

3=251×(5/X)A⇒X=144. This would
suggest using a CT ratio of 150/5. However, taking into account the fact that the
differential relay has several taps, it is not necessary to have exactly the same current
values at its terminals and therefore another CT ratio can be used. In this case a value
approximating to the nominal current is preferred and the ratio 250/5 is selected.
Finally, this ratio is checked to see if it is compatible with the taps that are available
on the relay.
With the two ratios selected in this way, the currents in the windings of the relay
for nominal conditions are
I
relay
at 69kV=251×(5/250) =5.02A
I
relay
at 11.5kV=1506.13×(5/1500) ×

3=8.69A
Therefore the tap range 5.0-9.0 A should be selected.
Example 8.2
For the transformer shown in Figure 8.8, determine the transformation ratios and the
connections required for the compensation transformers. Use differential relays of
the same type as for Example 8.1. Draw the complete schematic for the three-phase
connections and identify the currents in each of the elements.
Solution
Knowing the vector group of the transformer it is possible to determine the connec-
tions of the windings. Once this is obtained, the complete schematic diagram of the
connections can be drawn, indicating how the currents circulate in order to check
the correct functioning of the differential relay. The completed schematic is given in
Figure 8.9.
The load currents are
I
load
(13.2kV) =(12.5MVA)/(

3×13.2kV) =546.7A
I
load
(115kV) =(12.5MVA)/(

3×115kV) =62.75A
Differential protection 157
Dy1
OP
OP
OP
R
a:1
115 kV 100/5 13.2 kV 600/5
R
R R
R
R
I
A
– I
B
I
a
–I
b
I
b
–I
c
I
c
– I
a
I
a
– I
b
I
b
– I
c
I
c
– I
a
I
a
–I
b
I
b
–I
c
I
c
–I
a
I
C
I
B
I
A
I
C
I
B
I
A
I
a
I
b
I
c
I
a
I
b
I
c
I
a
I
b
I
c
I
B
– I
C
I
C
– I
A
Figure 8.9 Three-phase connection diagram for Example 8.2
In order to select the ratio of the compensation transformers it is necessary to
obtain a correct balance of currents, taking into account the delta connection on one
of the sides, i.e. 62.75A×(5/100) =546.7A×(5/600) ×(1/a) ×

3, from which
a =2.51. However, let us assume that this value of a is not included on commercially
available relays but that a value of a =

3 is, this being a very typical value provided
on this type of relay. On this basis the currents feeding into the relay are:
I
relay
(115kV) =62.75×(5/100) =3.13A
I
relay
(13.2kV) =546.7×(5/600) ×(1/

3) ×

3=4.55A
Selection of the tap setting:
4.55
3.13
=
X
5
,
fromwhich X=7.26A, so that for these conditions the tap to be selected is 5.0-7.3 A.
Example 8.3
In the system in Figure 8.10, given the power rating and voltage ratio of the power
transformer and the CTratios, determine the transformation ratios and the connections
of the compensation transformers required in order to set the differential relays, which
are not provided with taps. Produce the complete three-phase connection schematic
and identify the currents in each of the elements.
158 Protection of electricity distribution networks
500/5
Comp. trfr. 1
Comp. trfr. 2
13.2 kV
5 MVA
150/5
300/5
87
Primary–secondary Yd5
Primary–tertiary Yy0
25MVA
34.5 kV
115 kV
25MVA
Figure 8.10 Single-line diagram for Example 8.3
Solution
The three-phase diagram of connections is given in Figure 8.11; the connections
of the windings of the power transformer can be obtained from the vector group. If
the differential relays are connected on the Y side, then compensation transformer
number 2 should be connected in Yd5 in order to compensate for the phase difference
between the primary and secondary currents. There is no need for phase compensation
between the primary and tertiary winding, and a Yy0 connection for compensation
transformer 1 is therefore the most appropriate arrangement. It should be noted that the
common point of the operating coils and the neutral points of the compensation trans-
formers and CTs must be connected to only one earthing point in order to avoid mal-
operation during external faults. If several earthing points are used circulating currents
could be induced during external faults causing the relays to pick up inadvertently.
When determining the ratios of the compensation transformers, the calculations
involving the primary and secondary windings should be carried out on the basis of the
main transformer having only two windings with no current circulating in the tertiary.
The calculations involving the primary and tertiary windings should be treated in a
similar way. This method ensures that a correct selection is made which will cover
any fault or load current distribution.
Considering the currents on the primary and secondary sides, I
relay
(115kV) =
[25MVA/(

3×115kV)] ×(5/150) =125.51×(5/150) =4.18A. The current in
the relay associated with the 34.5 kV side should be equal to 4.18 A, so that
I
relay
(34.5kV) =4.18A=
25MVA

3×34.5kV
×
5
500
×
1

3
×
1
a
2
from which a
2
=0.578. The turns ratio of compensator 2 is therefore 1/0.578=1.73,
provided that the delta side has the higher number of turns.
The current in the relay for the 13.2 kV side is calculated assuming that the
power of the tertiary winding is equal to that of the primary winding, thus obtain-
ing a correct balance of magnitudes. This connection is equivalent to taking the
primary and the tertiary and treating them as a transformer with two windings so
Differential protection 159
OP
R
Comp.trfr.1
a
1
:1
Yy0
I
A
I
B
I
C
I
A
I
A
I
B
I
C
I
A
I
A
I
B
I
C
I
a
I
b
I
c
I
B
I
C
I
B
I
C
Yd5
Comp.trfr. 2
Yd5
R
OP
R
a
2
:1
I
b
–I
a
I
c
– I
b
I
a
– I
c
Yy0
I
a
I
a
I
a
I
a
I
b
I
c
I
a
I
b
I
c
I
a
I
b
I
c
I
b
I
c
I
b
I
c
I
b
I
c
115kV 34.5 kV
150/5
13.2 kV
R
OP
R
R
R
R
R
I
B
–I
A
I
C
– I
B
I
A
– I
C
Figure 8.11 Three-phase connection diagram for Example 8.3
that I
relay
(13.2kV) =4.18A=(25MVA/(

3×13.2kV)) ×(5/300) ×(1/a
1
), from
which a
1
=4.36. With this setting, it can be shown that, for any load distribution, the
primary restraint current is equal to the sum of the secondary and tertiary restraint
currents.
8.3.3 Percentage of winding protected by the differential relay during
an earth fault
Although differential protection is very reliable for protecting power transformers,
the windings are not always fully protected, especially in the case of single-phase
faults. Consider the case of a delta/star transformer as shown in Figure 8.12a, in
which the star winding has been earthed via a resistor. Assume an internal earth fault
occurs at point F at a distance X from the neutral point, involving X per cent turns,
160 Protection of electricity distribution networks
Distance X of fault from neutral (%)
F
a
u
l
t

c
u
r
r
e
n
t

I
(
p
e
r
-
u
n
i
t
)
Relay pick up
Winding not protected
20
0.3
0.2
0.1
0
60 40
0.9
0.8
0.7
0.6
0.5
0.4
I
S
I
L
1.0
100 80
Source B
C
I
L
I
L
=I
P
I
p
I
s
R
f
X
N
p
N
s
A
F
(a)
(b)
Figure 8.12 Delta/star transformer; star winding earthed via a resistor, with a fault
on the star side: (a) connection diagram; (b) fault current values for
primary and secondary
and that the resistor has been set so that nominal current I
nom
will flow for a fault
on the terminals, (with full line-to-neutral voltage applied between phase and earth).
The numbers of primary and secondary turns are N
p
and N
s
respectively.
The secondary current for a fault at F is produced by X per cent of the line-to-
neutral voltage. Therefore, by direct ratio, the current will be XI
nom
. In addition, the
Differential protection 161
number of turns involved in the fault is XN
s
. The distribution of current in the delta
side for an earth fault on the star side results in a line current I

L
equal to the phase
current. Therefore
I

L
=XI
nom
×(XN
s
/N
p
) =X
2
I
nom
(N
s
/N
p
) (8.1)
Under normal conditions, the line current in the delta side, I
L
, is
I
L
=

3I
nom
×(N
s
/N
p
) (8.2)
If the differential relay is set to operate for 20 per cent of the nominal line current
then, for operation of the relay, the following should apply:
I

L
≥0.2×I
L
i.e.
X
2
I
nom
(N
s
/N
p
) ≥0.2×

3×I
nom
×(N
s
/N
p
)
X
2
≥0.2

3, i.e. X≥59 per cent
Therefore, 59 per cent of the secondary winding will remain unprotected. It should
be noted that to protect 80 per cent of the winding (X≥0.2) would require an effective
relay setting of 2.3 per cent of the nominal primary current. This level of setting can
be very difficult to achieve with certain types of differential relays.
Figure 8.12b illustrates typical primary and secondary currents for delta/star trans-
formers, where the secondary star winding is earthed via a resistor, and also the effect
of the location of a fault along the star winding on the pick up of the differential
relays.
8.3.4 Determination of the slope
The setting of the slope of differential relays is carried out with the aim of ensuring
that there will be no mal-operations because of differences in the currents in the
restraint windings due to the transformation ratios of the CTs and the operation of the
tap changer under load conditions. In order to determine the slope, the restraint and
operating torques are calculated on the basis of the currents and the number of turns
in the respective coils as set out below:
T
res
=I
1
N
R1
+I
2
N
R2
T
op
=|I
1
−I
2
|N
op
where: I
1
, I
2
=currents in the secondaries of the CTs; N
R1
, N
R2
=number of turns
on the restraint coils (see Figure 8.2).
In order for the relay to operate, T
op
>T
res
, i.e. |I
1
−I
2
|N
op
>I
1
N
R1
+I
2
N
R2
. If
N
R1
=N
R2
=N
R
/2, this then gives T
res
=(I
1
+I
2
)N
R
/2.
For relay operation, the slope
|I
1
−I
2
|
0.5|I
1
+I
2
|

N
res
N
op
=m
A typical operating curve is shown in Figure 8.3.
162 Protection of electricity distribution networks
8.3.5 Distribution of fault current in power transformers
When considering the operation of differential protection it is very important to take
into account the distribution of fault current in all the windings in order to ensure that
the settings that have been selected have a suitable sensitivity. This is particularly
critical for single-phase faults in transformers that are earthed via an impedance. The
following example illustrates the procedure.
Example 8.4
A 115/13.2 kV Dy1 transformer rated at 25 MVA has differential protection as indi-
cated in Figure 8.13. The transformer is connected to a radial system, with the source
on the 115 kV side. The minimum operating current of the relays is 1 A. The trans-
former 13.2 kV winding is earthed via a resistor which is set so that the current for
a single-phase fault on its secondary terminals is equal to the nominal load current.
Draw the complete three-phase diagram and indicate on it the current values in
all the elements for:
(i) Full load conditions.
(ii) When a fault occurs at the middle of the winding on phase C, on the 13.2 kV
side, assuming that the transformer is not loaded.
For both cases indicate if there is any relay operation.
Solution
Full load conditions
The full load conditions for the maximum load of the transformer are as follows:
I
nom(13.2kV)
=
25×10
6
VA

3×13.2×10
3
V
=1093.47A
and
I
nom(115kV)
=
25×10
6
VA

3×115×10
3
V
=125.51A
It should be noted that, based on the primary currents given, the phase rotation
A-B-C is negative, i.e. clockwise. Therefore, the respective currents at the secondary
lead the primary currents by 30

in order to provide the required phase shifting for
the transformer vector group (Dy1). Figure 8.13 shows the current values through
the HV and LV connections, and it is clear that balanced currents are presented to the
differential relays, which therefore do not pick up, as expected.
Fault at the middle of 13.2 kV winding C
Since the transformer is earthed through a resistor that limits the current for faults at
the transformer 13.2 kV bushings to the rating of the winding, and since the fault is
at the middle of the winding, the fault current is then equal to half the rated value as
follows:
I
fault
=(I
nom(13.2kV)
)/2=1093.47/2=546.74A
Differential protection 163
2
5
M
V
A

D
y
1
S
u
p
p
l
y
A B C
1
2
5
.
5
1
0
°
7
2
.
4
6
3
0
°
7
2
.
4
6
1
5
0
°
7
2
.
4
6
2
7
0
°
1
5
0
/
5
R
0
.
0
3

1
A

=
>
D
o
e
s

n
o
t

o
p
e
r
a
t
e
0
.
0
3
0
.
0
3
0
.
0
3
2
2
5
0
/
5
2
.
4
3
L
o
a
d
R
R
O
P
R
R
O
P
R
R
O
P
Ω
4
.
1
8
0
°
Ω
1
2
5
.
5
1
1
2
0
°
Ω
4
.
1
8
1
2
0
°
Ω
1
2
5
.
5
1
2
4
0
°
Ω
4
.
1
8
2
4
0
°
Ω
4
.
2
1
0
°
Ω
4
.
2
1
1
2
0
°
Ω
4
.
2
1
2
4
0
°
Ω
1
0
9
3
.
4
7
3
0
°
Ω
1
0
9
3
.
4
7
1
5
0
°
Ω
1
0
9
3
.
4
7
2
7
0
°
Ω
3
0
°
2
.
4
3
1
5
0
°
2
.
4
3
2
7
0
°
F
i
g
u
r
e
8
.
1
3
T
h
r
e
e
-
p
h
a
s
e
c
o
n
n
e
c
t
i
o
n
d
i
a
g
r
a
m
f
o
r
E
x
a
m
p
l
e
8
.
4
164 Protection of electricity distribution networks
The primary current within the delta winding is
I
prim
=I
fault
×
(N
2
/2)
N
1
, since
N
2
N
1
=
V
2

3×V
1
Then
I
prim
=I
fault
×
V
2


3×V
1
=546.47×
13.2


3×115
=18.12A
Figure 8.14 shows the current values through the HV and LV connections, from
which it can be seen that, for this case also, the differential relays do not operate
since the current through their operating coils is only 0.6 A, which is less than the 1 A
required for relay operation.
8.4 Differential protection for generators and rotating machines
Differential protection for generators and other rotating machines is similar to that
for transformers in several ways. Internal generator winding faults include phase-to-
phase short-circuits, short-circuited turns, open circuits and faults to earth, and should
be disconnected by opening the circuit as quickly as possible. In order to obtain the
most effective form of differential protection the neutral of the generator should be
well earthed, either solidly or via a resistor or a reactor. The differential protection
should satisfy the following requirements:
1. It should be sensitive enough to detect damage in the winding of the generator
stator, and yet not operate for faults outside the machine.
2. It should operate quickly in such a way that the generator is disconnected before
any serious damage can result.
3. It should be designed so that the main breaker is opened, as well as the neutral
breaker and the field circuit breaker.
The arrangements of the CTs and the differential relays for a machine connected
in star can be seen in Figure 8.15, and in Figure 8.16 for a delta connection. If the
neutral connection is made inside the generator and the neutral taken outside as shown
in Figure 8.17, the differential protection provided will only cover earth faults.
8.5 Line differential protection
The form of differential protection using only one set of relays as illustrated in
Figure 8.2 is not suitable for long overhead lines since the ends of a line are too far
apart to be able to interconnect the CT secondaries satisfactorily. It is therefore neces-
sary to install a set of relays at each end of the circuit and interconnect them by some
suitable communication link. Pilot protection is an adaptation of the principles of dif-
ferential protection that can be used on such lines, the term pilot indicating that there
is an interconnecting channel between the ends of the lines through which information
Differential protection 165
2
5
M
V
A

D
y
1
0
.
6
0
.
6
A B C
1
5
0
/
5
0
.
6
A
0
.
6
A
1
8
.
1
2
A
1
8
.
1
2
A
1
8
.
1
2
A
S
u
p
p
l
y
0
.
6
0
.
6
R
5
4
8
.
7
2
2
5
0
/
5
L
o
a
d
T S R
R
R
O
P
R
R
O
P
R
R
O
P
F
i
g
u
r
e
8
.
1
4
C
o
n
d
i
t
i
o
n
s
f
o
r
a
f
a
u
l
t
a
t
t
h
e
m
i
d
d
l
e
o
f
t
h
e
w
i
n
d
i
n
g
o
n
p
h
a
s
e
C
o
n
t
h
e
1
3
.
2
k
V
s
i
d
e
(
E
x
a
m
p
l
e
8
.
4
)
166 Protection of electricity distribution networks
OP
Generator
windings
R
R
OP
b a c
Differential
relay
R
R
R
OP
R
52
Figure 8.15 Differential protection for a generator connected in star, with four
terminals
can be transmitted. There are three different types of interconnecting channels – pilot
wires, current-carrying wires and centimetric-wave systems. A pilot wire arrange-
ment generally consists of two telephone-line type wires, either overhead or cable. In
a current-carrying pilot system, high frequency low tension currents are transmitted
along the length of a power conductor to a receiver at the other end; the earth or
guard wire generally acts as the return conductor. Centimetric waves are transmitted
by means of a very high frequency radio system, usually operating above 900 MHz.
The principle of operation of pilot differential protection is similar to the differen-
tial systems for protecting generators and transformers, but the relays have different
settings because the breakers at the ends of the line are more widely separated and
a single relay should not be used to operate two tripping circuits. This method of pro-
tection is ideal from a theoretical point of view as both ends of the line should open
instantaneously for faults wherever they occur on the line. In addition, the system
should not operate for faults outside the section and is therefore inherently selective.
Many of the operational difficulties with conventional schemes due to induced cur-
rents have been overcome by the use of fibre optics, which has greatly improved the
reliability of this type of protection.
Differential protection 167
Generator
windings
52
OP
OP
OP
b a c
Differential
relay
R
R R
R R
R
Figure 8.16 Differential protection for a generator connected in delta
Generator
windings
52
OP
R
a b c
R R
Figure 8.17 Differential protection for a generator connected in star
168 Protection of electricity distribution networks
8.6 Busbar differential protection
Busbar differential protection is based on the same principles as transformer and
generator differential protection. Under normal system conditions the power that
enters a busbar is identical to the power that leaves; a fault inside the differential
circuit unbalances the system and current thus flows in the operating coil of the relay,
which then results in the tripping of all the breakers associated with that busbar.
There can be many circuits connected to the busbar, which necessarily implies the
connection of a number of CT secondaries in parallel. In busbar differential schemes
that involve bushing type CTs, six to eight secondaries can usually be connected in
parallel without difficulty. Although some busbar differential protection schemes still
use multiple restraint features, high impedance relays predominate because of their
better performance. The majority of faults on a busbar involve one phase and earth
and are due to many causes such as lightning and imperfections in the insulation of
switchgear equipment. However, a large proportion of busbar faults are the result of
human error rather than faults on switchgear equipment.
8.6.1 Differential system with multiple restraint
Differential relays with a variable-percentage characteristic should be used in a
scheme involving multiple restraint, as shown in Figure 8.18. The secondaries of
R
R
R
R
OP
OP
52
52
OP
OP
Busbar
52
52
R
R
R
R
Figure 8.18 Multiple restraint busbar differential protection
Differential protection 169
the CTs on the feeders on the outgoing side of the busbar are connected in parallel
and across the differential relay, together with the secondaries of CTs of the circuits
on the incoming side of the busbar which also are connected in parallel.
8.6.2 High impedance differential system
The high impedance arrangement tends to force any incorrect differential current to
flow through the CTs instead of through the operating coil of the relay, and thus
avoids mal-operation for external fault or overload conditions when the secondary
currents of all the CTs are not the same because of differences in the magnetisation
characteristics.
Connecting CTs in parallel
This arrangement requires only one high impedance relay, connected across the ter-
minals of the CT secondaries which are connected in parallel to a set of CTs per
circuit, as shown in Figure 8.19. However, with the connections made in this way,
the busbar is only protected from earth faults. In order for the scheme to be effective,
the resistance of the CT secondary wiring should be as low as possible.
The relay basically consists of an instantaneous overvoltage unit which is set by
calculating the maximum voltage at the relay terminals for an external fault, taking
into account the maximum primary fault current, the resistance of the secondary
windings and the wiring, and the transformation ratios of the CTs, plus a safety
margin. Consequently, during an external fault, the voltage across the terminals of
the relay is relatively low and does not initiate any relay operation. During internal
faults the voltage across the relay terminals is higher and results in the operation of
the instantaneous overvoltage unit which sends a tripping signal to the appropriate
breakers.
a b c
High impedance
differential relay
87
Busbar
Figure 8.19 High impedance differential protection scheme with CTs in parallel
170 Protection of electricity distribution networks
87A
A
B
C
N
87C 87B
Busbar
Figure 8.20 High impedance differential protection arrangement using a common
bus for each phase
Using common buses for each phase
In this scheme balanced groups of CTs are formed in each one of the phases, using a
common bus to feed three elements of a high impedance differential relay as illustrated
in Figure 8.20, so that the busbar is protected against both phase and earth faults.
8.7 Exercises
Exercise 8.1
The transformer shown in Figure 8.21 is protected by a differential relay provided
with the following taps in each restraint circuit:
2.9-3.2-3.5-3.8-4.2-4.6-5.0-8.7
If the current transformers have the ratios indicated in the diagram, and have been
selected in such a way that there is no saturation, determine the taps for each
restraint coil.
Exercise 8.2
Carry out the same calculations as for Exercise 8.1 for the transformer in Figure 8.22,
using a differential relay with the same taps for each restraint circuit.
Differential protection 171
1000/5
12.4/69kV
16/20MVA
200/5
Figure 8.21 Diagram for Exercise 8.1
161 kV
30/40 MVA
69 kV
30/40MVA
500/5
200/5
12.4 kV
7.5/10 MVA
400/5
Figure 8.22 Diagram for Exercise 8.2
200/5
4500
4500
Dy1
10 MVA
34.5/13.2kV I
B
I
C
4.5 kA
900/5
R R
OP
R R
OP
R R
OP
Figure 8.23 Diagram for Exercise 8.3
Exercise 8.3
Consider the 10 MVA, 34.5/13.2 kV, Dy1 transformer whose connections are given
in Figure 8.23. The CTs on the 34.5 kV side of the transformer have a ratio of 200/5,
and those on the 13.2 kV side a ratio of 900/5.
172 Protection of electricity distribution networks
a:1
I
A
I
B
I
C
600/5
1200/5
Yd7
69/12.4 kV
R R
OP
R R
OP
R R
OP
Figure 8.24 Diagram for Exercise 8.4
Calculate the magnitude and direction of the currents in the primary circuits
and the CT secondaries. In addition, determine if these result in the differential
relays operating for a fault between phases b and c in the secondary windings of the
transformer which are carrying a fault current of 4.5 kA.
Check if the operation of the differential relays is appropriate. If the answer is
negative, indicate what correction should be applied.
Note: The differential relays operate for a current equal to 20 per cent of nominal
current, which is 5 A. Therefore, they operate for currents above 1 A.
Exercise 8.4
For the 16/20 MVA, 69/12.4 kV, Yd7 transformer shown in Figure 8.24, determine the
transformation ratios and connections of the compensation transformers in order to
connect the differential relays, which are not provided with taps to vary the settings.
The primary and secondary windings of the compensators can only be connected in
star or delta.
Draw the three-phase schematic diagram of the connections and identify the cur-
rents in each of the elements of the system, taking into account that I
A
, I
B
and I
C
feed into the transformer primary.
Chapter 9
Distance protection
9.1 General
It is essential that any faults on a power system circuit are cleared quickly; otherwise
they could result in the disconnection of customers, loss of stability in the system and
damage to equipment. Distance protection meets the requirements of reliability and
speed needed to protect these circuits, and for these reasons is extensively used on
power system networks.
Distance protection is a nonunit type of protection and has the ability to dis-
criminate between faults occurring in different parts of the system, depending on the
impedance measured. Essentially, this involves comparing the fault current, as seen
by the relay, against the voltage at the relay location to determine the impedance down
the line to the fault. For the system shown in Figure 9.1, a relay located at A uses the
A
F
1
B
F
2
A
F
1
B
F
2
Figure 9.1 Faults occurring on different parts of a power system
174 Protection of electricity distribution networks
line current and the line voltage to evaluate Z=V/I. The value of the impedance Z
for a fault at F
1
would be Z
AF1
, and (Z
AB
+Z
BF2
) for a fault at F
2
.
The main advantage of using a distance relay is that its zone of protection depends
on the impedance of the protected line that is a constant virtually independent of the
magnitudes of the voltage and current. Thus, the distance relay has a fixed reach, in
contrast to overcurrent units where the reach varies depending on system conditions.
9.2 Types of distance relays
Distance relays are classified depending on their characteristics in the R-X plane, the
number of incoming signals and the methods used to compare the incoming signals.
The most common type compares the magnitude or phase of the two incoming signals
in order to obtain the operating characteristics, which are straight or circular lines
when drawn in the R-X plane. Any type of characteristic obtainable with one type
of comparator can also be obtained with the other, although the quantities compared
would be different in each case.
If Z
R
is the impedance setting of the distance relay, it should operate when Z
R

V/I, or when IZ
R
≥V. As shown in Figure 9.2, this condition can be obtained in the
amplitude comparator that operates when the ampere-turns of the current circuit are
greater than the ampere-turns of the voltage circuit. However, it is difficult to provide
an amplitude comparator that functions correctly under fault conditions when the
phase displacement between V and I tends to be 90

and transients are present,
which leads to incorrect r.m.s. values of V and I which are required to evaluate
IZ
R
≥V. For these reasons, the use of amplitude comparators is limited; it is more
convenient to compare two signals by their phase difference than by their amplitudes.
Z
s
Fault
Restraint coil
Relay
To trip

V
f
R
+
Operating coil
V
f
=I
f
Z
f
I
f
I
f
Z
L
Figure 9.2 Relay based on an amplitude comparator
Distance protection 175
The following analysis shows that for two signals, S
0
and S
r
, which are to be compared
in magnitude, there exist two other signals S
1
and S
2
that can be compared by phase.
The relationship between the signals is as follows:
S
0
=S
1
+S
2
S
r
=S
1
−S
2
(9.1)
From eqns. 9.1
S
1
=
(S
0
+S
r
)
2
S
2
=
(S
0
−S
r
)
2
(9.2)
The comparison of the amplitudes is given by:
|S
0
| ≥|S
r
|
|S
1
+S
2
| ≥|S
1
−S
2
|
(9.3)
Defining S
1
/S
2
=C, the relationship 9.3 can be expressed as
|C+1| ≥|C−1| (9.4)
Drawing C in the R-X plane, as shown in Figure 9.3, it can be seen that condition
9.4 is satisfied in the semiplane on the right. This semiplane is defined for all the
points C

in such a way that −90

≤≤+90

.
Given that C

=(S
1

α)/(S
2

β), then the relationship 9.4 is satisfied when
−90

≤α −β ≤+90

(9.5)
The above relationships demonstrate that two signals, obtained for use with an
amplitude comparator, can be converted in order to be used by a phase angle com-
parator. The signals to be compared are analysed in the following paragraphs to obtain
the operating characteristics of the main types of distance relays.
Operating zone
–1
C– 1 C C+1
+1
Real
Imaginary
Figure 9.3 Comparison of phases in a complex plane: C=S
1
/S
2
176 Protection of electricity distribution networks
9.2.1 Impedance relay
The impedance relay does not take into account the phase angle between the voltage
and the current applied to the relay and, for this reason, its operating characteristic in
the R-X plane is a circle with its centre at the origin of the co-ordinates and a radius
equal to the setting in ohms. The relay operates for all values of impedance less than
the setting, i.e. for all the points inside the circle. Thus, if Z
R
is the impedance setting,
it is required that the relay will operate when Z
R
≥V/I, or when IZ
R
≥V. In order
for an impedance relay to work as a phase comparator, the following signals should
be assigned to S
0
and S
r
:
S
0
=IZ
R
S
r
=KV
(9.6)
The constant K takes into account the transformation ratios of the CTs and VTs.
The corresponding signals for a phase comparator are
S
1
=KV +IZ
R
S
2
=−KV +IZ
R
(9.7)
Dividing eqns. 9.7 by KI, gives
S
1
=Z+Z
R
/K
S
2
=−Z+Z
R
/K
(9.8)
where
Z=V/I.
Note that the magnitudes of signals S
1
and S
2
have been changed when dividing
by KI. However, this is not important since the main purpose is to retain the phase
difference between them. It should be noted that drawing S
1
and S
2
in one or the
other scale does not affect the phase relationship between the two signals.
Drawing Z
R
/K and the eqns. 9.8 in the R-X plane, the operating characteristic
of the relay is determined by the locus of the points Z such that , the phase angle
between S
1
and S
2
, is given by −90

≤≤+90

. The construction is shown in
Figure 9.4. Eqns. 9.8 give the origin of the rhomboid OABC which has diagonals of
S
1
and S
2
. From the properties of the rhomboid, the angle between S
1
and S
2
is 90

if |Z| =|Z
R
/K|. Therefore point C is the limit of the operating zone, and the locus
of point C for the different values of Z is a circle of radius Z
R
/K.
If Z<Z
R
/K, the situation shown in Figure 9.5 is obtained. In this case is less
than 90

and consequently the vector for Z is inside the relay operating zone. If, on
the other hand, Z>Z
R
/K, as in Figure 9.6, then is greater than 90

and Z is outside
the operating zone of the relay, which then will not operate. Being nondirectional, the
impedance relay will operate for all faults along the vector AB (see Figure 9.7) and
for all faults behind the busbar, i.e. along the vector AC. The vector AB represents
the impedance in front of the relay between its location at A and the end of the line
AB, while the vector AC represents the impedance of the line behind the site of the
relay.
Distance protection 177
Limit of
operating zone
R
O
K
Z
R
A
S
2
S
1
Z
Θ=90°
C
X
B
Figure 9.4 Operating characteristic of an impedance relay obtained using a phase
comparator
R
Operating zone
O
Z
X
K
Z
R
S
2
S
1
Θ<90°
Figure 9.5 Impedance Z inside the operating zone of an impedance relay
178 Protection of electricity distribution networks
R
O
Z
K
Z
R
S
1
S
2
Θ>90°
X
Figure 9.6 Impedance Z outside the operating zone of an impedance relay
A
C
X
R
B
C
Line AC
A
Line AB
21
B
Figure 9.7 Impedance relay characteristic in the complex plane
Distance protection 179
The impedance relay has three main disadvantages:
1. It is not directional; it will see faults in front and behind its location and therefore
requires a directional element in order to obtain correct discrimination. This can
be obtained by adding an independent directional relay to restrict or prevent the
tripping of the distance relay when power flows out of the protected zone during
a fault.
2. It is affected by the arc resistance.
3. It is highly sensitive to oscillations on the power system, due to the large area
covered by its circular characteristic.
9.2.2 Directional relay
Directional relays are elements that produce tripping when the impedance measured
is situated in one half of the R-X plane. They are commonly used together with
impedance relays in order to limit the operating zone to a semi-circle.
The operating characteristic is obtained froma phase comparison of the following
signals:
S
1
=KV
S
2
=Z
R
I
(9.9)
Dividing by KI, and defining Z=V/I, gives
S
1
=Z
S
2
=Z
R
/K
(9.10)
The operating zone of the directional relay is defined by the values of Z and Z
R
,
which result in a phase difference between S
1
and S
2
of less than 90

. The construction
of the characteristic is shown in Figure 9.8, in which S
1
and S
2
are drawn.
K
Θ<90°
Operating zone
Z
R
X
Z=S
1
=S
2
R
Figure 9.8 Operating zone of a directional relay
180 Protection of electricity distribution networks
9.2.3 Reactance relay
The reactance relay is designed to measure only the reactive component of the line
impedance; consequently, its setting is achieved by using a value determined by the
reactance X
R
. In this case the pair of equations for S
1
and S
2
is as follows:
S
1
=−KV +X
R
I
S
2
=X
R
I
(9.11)
and, dividing by KI, gives
S
1
=−Z+X
R
/K
S
2
=X
R
/K
(9.12)
The operating characteristics are obtained by drawing eqns. 9.12 in the complex
plane and determining those values of Zfor which is less than 90

. The construction
is shown in Figure 9.9; here, the limit of the operating zone is a straight line parallel
to the resistance axis, drawn for a reactance setting of X
R
/K.
As the impedance of the fault is almost always resistive, it might be assumed that
the fault resistance has no effect on the reactance relays. In a radial system this is
generally true, but not necessarily if the fault is fed from two or more points since
the voltage drop in the fault resistance is added to the drop in the line and affects the
relay voltage. Unless the current in the relay is exactly in phase with the fault current,
the voltage drop in the fault resistance will result in a component 90

out of phase
to the relay current, producing an effect similar to the line reactance. This apparent
reactance can be positive or negative and add to, or subtract from, the impedance
measured by the relay, thus affecting its operation. If the fault resistance is large in
comparison to the line reactance the effect could be serious and this type of relay
should not be used.
Figure 9.10 shows the voltage seen by the relay in the presence of faults with
arc resistance and two infeeds. From the diagram it will be seen that the relay will
measure a value that is smaller than the actual reactance between the relay point and
the fault.
K
Θ<90°
X
R
X
S
2
=
R
Operating zone
0
Z
S
1
Figure 9.9 Operating zone of a reactance relay
Distance protection 181
Relay
I
1
+I
2
I
1
I
2
R
1
+jX
1
R
2
+jX
2
R
f
Fault
resistance
(I
1
+I
2
) R
f
I
1
+I
2
I
1
X
1
I
1
R
1
: Error due to the fault
resistance
Voltage seen
by the relay
I
1
I
2

(a)
(b)
Figure 9.10 Voltage seen by a relay in the presence of faults with arc resistance and
two infeeds: (a) schematic of circuit; (b) vector diagram
9.2.4 Mho relay
The mho relay combines the properties of impedance and directional relays. Its char-
acteristic is inherently directional and the relay only operates for faults in front of
the relay location; in addition it has the advantage that the reach of the relay varies
with the fault angle. The characteristic, drawn in the R-X plane, is a circle with a
circumference that passes through the origin of the co-ordinates and is obtained by
assigning the signals the following values:
S
1
=−KV +Z
R
I
S
2
=KV
(9.13)
from which
S
1
=−Z+Z
R
/K
S
2
=Z
(9.14)
Drawing Z
R
/K and the eqns. 9.14 in the R-X plane, the relay characteristic is
determined by the locus for the values of Z that are fulfilled when is less than 90

.
In this case the limit of the operating zone (=90

), as shown in Figure 9.11, is
traced by a circle with a diameter of Z
R
/K and a circumference that passes through
the origin of the co-ordinates. For values of Z located inside the circumference,
will be less than 90

, as is shown in Figure 9.12, and this will result in operation of
the relay.
182 Protection of electricity distribution networks
K
Z
R
Θ=90°
O
R
Z=S
2
S
1
X
Limit of
operating zone
Figure 9.11 Operating characteristic of a mho relay
K
Z
R
R
Z=S
2
S
1
O
X
Θ<90°
Figure 9.12 Impedance Z inside the operating zone of a mho relay
9.2.5 Completely polarised mho relay
One of the disadvantages of the autopolarised mho relay is that, when it is used on long
lines and the reach does not cover the section sufficiently along the resistance axis,
then it is incapable of detecting faults with high arc or fault resistances. The problem
is aggravated in the case of short lines since the setting is low and the amount of the
R axis covered by the mho circle is small in relation to the values of arc resistance
expected.
Distance protection 183
K
Z
R
O
Operating characteristic
for unbalanced fault
R
X
Mho circle
Figure 9.13 Operating characteristic of a completely polarised mho relay
One practical solution to this problem is to use a completely polarised mho relay
where the circular characteristic is extended along the Raxis for all unbalanced faults,
as illustrated in Figure 9.13. This characteristic can be obtained by means of a phase
comparator, which is fed by the following signals:
S
1
=V
pol
S
2
=V −IZ
R
(9.15)
where: V =voltage at the location of the relay, on the faulted phase or phases; V
pol
=
polarisation voltage taken from the phase, or phases, not involved with the fault;
I =fault current; Z
R
=setting of the distance relay.
9.2.6 Relays with lens characteristics
Distance relays with lens characteristics are very useful for protecting high impedance
lines that have highpower transfers. Under these conditions the line impedance values,
which are equal to V
2
/S, become small and get close to the impedance characteristics
of the relay, especially that of zone 3. This offset lens characteristic, which can be
adjusted to offset the mho circular characteristic as shown in Figure 9.14, is common
in some relays.
9.2.7 Relays with polygonal characteristics
Relays with polygonal characteristics provide an extended reach in order to cover
the fault resistance, in particular for short lines, since the position of the resistance
line can be set in the tripping characteristic (see line 2 in Figure 9.15 which shows a
typical polygonal operating characteristic).
The polygonal tripping characteristic is obtained from three independent mea-
suring elements – reactance, resistance and directional. In order to achieve this
184 Protection of electricity distribution networks
Zone 2
Zone 1
X
Zone 3
R
Z
1
, Z
2
Figure 9.14 Zone 3 offset lens characteristic
X
R
Zone 4
(T
4
)
Zone 3
(T
3
)
Zone 2
(T
2
)
Zone 1
(T
1
)
Zone 5
(T
5
)
Line
characteristic
R
1
R
2
R
3
R
4
X
4
Z
4
Z
3
Z
2
Z
1
X
3
X
2
X
1
Z
5
Figure 9.15 Polygonal relay operating characteristic
Distance protection 185
X
Z
K
R
X
R
Z
K
(a) (b)
Figure 9.16 Typical combined operating characteristics: (a) R/X ratio =0.5;
(b) R/X ratio =2
characteristic, the measuring elements are suitably combined. The relay is tripped
only when all three elements have operated; in this way the required polygonal
characteristic is obtained.
9.2.8 Relays with combined characteristics
Atypical combined operating characteristic is defined in the impedance plane by lines
running parallel to the resistive and reactive axes which cross each other at the setting
point for Z
K
, as shown in Figure 9.16. In order to achieve the required directionality,
a mho circle that passes through Z
K
is employed. In relays with this characteristic the
reaches in the resistive and reactive directions have the same range of settings and
can be adjusted independently of each other.
9.3 Setting the reach and operating time of distance relays
Distance relays are set on the basis of the positive-sequence impedance fromthe relay
location up to the point on the line to be protected. Line impedances are proportional
to the line lengths and it is this property that is used to determine the position of the
fault, starting from the location of the relay. However, this value is obtained by using
systemcurrents and voltages fromthe measurement transformers that feed the relays.
Therefore, in order to convert the primary impedance into a secondary value, which
is used for the setting of the distance relay, the following expression is used:
V
prim
I
prim
=Z
prim
=
V
sec
×VTR
I
sec
×CTR
(9.16)
Thus
Z
sec
=Z
prim
×(CTR/VTR) (9.17)
where CTR and VTR are the transformation ratios of the current and voltage
transformers, respectively.
186 Protection of electricity distribution networks
A
x
Relay
t
AB+BC+25%CD AB+50%BC 80%AB
B
t
2
C
t
3
D
Figure 9.17 Distance relay protection zones for a radial system
Normally, three protection zones in the direction of the fault are used in order
to cover a section of line and to provide back-up protection to remote sections (see
Figure 9.17). Some relays have one or two additional zones in the direction of the
fault plus another in the opposite sense, the latter acting as a back-up to protect the
busbars. In the majority of cases the setting of the reach of the three main protection
zones is made in accordance with the following criteria:
• zone 1: this is set to cover between 80 and 85 per cent of the length of the
protected line;
• zone 2: this is set to cover all the protected line plus 50 per cent of the shortest
next line;
• zone 3: this is set to cover all the protected line plus 100 per cent of the second
longest line, plus 25 per cent of the shortest next line.
In addition to the unit for setting the reach, each zone unit has a timer unit. The
operating time for zone 1, t
1
, is normally set by the manufacturer to trip instanta-
neously since any fault on the protected line detected by the zone 1 unit should be
cleared immediately without the need to wait for any other device to operate. The
operating time for zone 2 is usually of the order of 0.25 to 0.4 s, and that of zone 3 is
in the range of 0.6 to 1.0 s. When there are power transformers at adjacent substations
the zone 2 timer should have a margin of 0.2 s over the tripping time of any associated
transformer overcurrent protection. In the case of zone 3, when the settings of relays
at different locations overlap, then the timer for the zone 3 of the furthest relay should
be increased by at least 0.2 s to avoid incorrect co-ordination. However, the operating
time for the zone 3 units should also be set at a value that will ensure that system
stability is maintained and therefore, if necessary, consideration may have to be given
to reducing the zone 3 operating time in such circumstances.
Since the tripping produced by zone 1 is instantaneous, it should not reach as
far as the busbar at the end of the first line (see Figure 9.17) so it is set to cover
only 80–85 per cent of the protected line. The remaining 20–15 per cent provides
a factor of safety in order to mitigate against errors introduced by the measurement
transformers and line impedance calculations. The 20–15 per cent to the end of the
line is protected by zone 2, which operates in t
2
s. Zone 3 provides the back-up and
operates with a delay of t
3
s. Since the reach and therefore the operating time of the
Distance protection 187
Zone 3 unit
Zone 1 unit
Zone 2 unit
A
R
X
A B
C
B
C
Figure 9.18 Operating characteristic of distance protection located at A
distance relays are fixed, their co-ordination is much easier than that for overcurrent
relays.
In order to illustrate the philosophy referred to earlier, consider the case of the
system in Figure 9.18 in which it is required to protect the lines AB and BC. For this,
it is necessary to have three relays at A to set the three zones. All three units should
operate for a fault within the operating characteristic of zone 1. For a fault on line BC,
but within the cover of the zone 2 unit at A, both the zone 2 and zone 3 units should
operate. Since there is also distance protection at substation B, the relay at A should
provide an opportunity for the breaker at B to clear the fault; it is for this reason that
the zone 2 and zone 3 units operate with an appropriate time delay in order to obtain
discrimination between faults on lines AB and BC. The diagram of operating times
is shown in Figure 9.19.
Some methods for setting distance relays use different criteria to those already
mentioned, mainly with regard to the reach of zones 2 and 3. In particular, there is the
method where it is recommended that the reach of zone 2 should be 120 per cent of
the impedance of the line to be protected, and that for zone 3 should be 120 per cent of
the sum of impedances of the protected line and of its longest adjacent line. In this
case the times for zones 2 and 3 should not have a fixed value, but should be based
on the opening time of the breakers and the reach of the relays to guarantee that there
will be no overlap in the same zones covered by adjacent relays. Since the same
philosophy is used as the basis for either method, no specific recommendation is
188 Protection of electricity distribution networks
A
Operating
time
t
3
t
2
Zone 3 (A)
Zone 2 (A)
Zone 1 (A)
B C
Distance
Figure 9.19 Operating times for distance protection at A
made to use one or the other, given that the actual selection is generally dependent
on the characteristics of the system in question.
Modern distance relays, especially the numerical types, offer zones 4 and 5 to
reinforce the back-up functions, as shown in Figure 9.15. In these cases, zones 3 and
4 provide cover only in the forward direction and zone 5 in the backward direction.
The setting for the first three zones is the same as discussed before, but the settings for
zones 4 and 5 can vary from utility to utility. Some accepted criteria suggest setting
zone 4 at 120 per cent of zone 3, and zone 5 at 20 per cent of zone 1. The time delay
for zones 4 and 5 is normally the same as that for zone 3 but increased by a margin
of, typically, 400 ms. Care should be taken to ensure that the zones with the higher
settings, i.e. zones 3 and 4, do not overlap different voltage levels through step-up or
step-down transformers, or load impedance values.
9.4 The effect of infeeds on distance relays
The effect of infeeds needs to be taken into account when there are one or more
generation sources within the protection zone of a distance relay which can contribute
to the fault current without being seen by the distance relay.
Analysing the case illustrated in Figure 9.20, it can be appreciated that the
impedance seen by the distance relay at A for a fault beyond busbar B is greater
than actually occurs. In fact, if a solid earth-fault is present at F, the voltage at the
relay at A would be
V
A
=I
A
Z
A
+(I
A
+I
B
)Z
B
(9.18)
from which
V
A
I
A
=Z
A
+
¸
1+
I
B
I
A
¸
Z
B
(9.19)
The relay therefore sees an impedance of KZ
B
, the infeed constant K being equal
to I
B
/I
A
, in addition to the line impedance Z
A
, which implies that its reach is reduced.
Distance protection 189
I
A
I
A
+I
B
I
B
A Z
A
Z
B
F
B
Figure 9.20 Effect of an infeed on distance protection
The setting of zones 2 and 3 for the relay at A should then take the following form:
Z
relay
=Z
A
+(1+K)Z
B
(9.20)
where K is given as
K=
I
total infeed
I
relay
(9.21)
It is necessary to take into account the fact that the distance relay can over reach
if the sources that represent the infeed are disconnected, so that a check should be
made for these conditions to ensure that there is no overlap with the adjacent zone 2.
For systems in which zones 2 and 3 cover lines that are not part of a ring, the value
of K is constant and independent of the location of the fault, given the linearity of
electrical systems. By way of illustration, consider the system shown in Figure 9.21a.
Figure 9.21b shows the impedance that is seen by a distance relay located in substation
C. For a fault between B and D, the value of K will be the same for faults at either
substation B or D, or at some other point on the line linking the two substations. If
the fault location moves from B to D, the current values diminish but the ratio of the
total infeed to the current seen by the relay will be the same.
Since the value of the infeed constant depends on the zone under consideration,
several infeed constants, referred to as K
1
, K
2
and K
3
, need to be calculated. K
1
is used to calculate the infeed for the second zone. K
2
and K
3
are used for zone 3,
K
2
taking account of the infeed on the adjacent line and K
3
that in the remote line.
Based on the criteria in Section 9.3, and considering the infeed as discussed before,
the expressions for calculating the reach of the three zones for the relay located at
busbar C would be
Z
1
=0.8 to 0.85 times Z
AB
Z
2
=Z
CB
+0.5(1+K
1
)Z
BD
Z
3
=Z
CB
+(1+K
2
)Z
BF
+0.25(1+K
3
)Z
FH
where
K
1
=
I
A
+I
E
+I
F
I
C
K
2
=
I
A
+I
D
+I
E
I
C
K
3
=
I
A
+I
D
+I
E
+I
G
I
C
190 Protection of electricity distribution networks
Z
L1 Z=m.d(1 +K)
Z=m.d
m=slope
d
21
C
A
D
F
B E
I
A
I
E
I
C
I
D
G
H
I
F
I
H
I
G
(a)
(b)
Figure 9.21 Distance protection with a fixed value of K: (a) typical system for
analysis of infeed with fixed K; (b) impedance seen by distance
relay at C
Since the coverage on the remote line is not very critical, K
3
can be assumed to
be equal to K
2
, which reduces the reach of the zone 3. However, this difference is
normally negligible, and the assumption is therefore acceptable.
In drawing the impedance as a function of the distance, as shown in Figure 9.21b,
it can be seen that the slope of line CB is constant, as is to be expected. However, in
section BD the slope continues to be constant, although of a different value to that
in the section CB, due to the inclusion of the term K. The important point is that the
impedance seen by the relay at C for faults in section BD is directly proportional to
the location of the fault, the same as for faults in section CB. When zones 2 and 3
cover part of the lines of a double circuit, or of rings, the value of K will depend on
Distance protection 191
I
A
A
I
A
A
I
C
E
C
B
I
D
I
C
C
D
B
I
D
D
Figure 9.22 Distance protection with variable K
the fault point or busbar under consideration, as is evident from the two schematics
shown in Figure 9.22. In both cases the impedance that the relay sees for faults along
the line BC is a curve of varying slope.
If the infeed equation is split, then the impedance that the relay sees for a fault in
section BC of the two schemes in Figure 9.22 can be written as
Z
relay
=Z
line
+
¸
1+
I
D
+I
C
I
A
¸
Z
B
to fault (9.22)
192 Protection of electricity distribution networks
In this case, the ratio of I
D
/I
A
for a fault on the line BC is fixed since it
does not depend on the location of the fault. However, the ratio I
C
/I
A
is vari-
able and could even be negative for faults near to substation C when the current
I
C
changes direction. Taking into account the previous considerations, it should
be expected that, starting from busbar B, Z should increase as the fault point is
moved towards C until it reaches its maximum value, after which the value of Z
starts to fall. For a setting of Z calculated with a fixed value of K, the line can be
protected in one predetermined zone from busbar B by applying the infeed equa-
tion, and then in the other zone in front of busbar C because of the reduction in the
impedance.
If the settings are calculated using current values for faults on busbar C, it is
probable that applying the equation will not protect the line in zone 2 up to the
required 50 per cent. If the values are taken for a fault at busbar B, it is possible that
there will be cover in sections beyond 50 per cent, and even beyond 80 per cent,
which will produce overlap with zone 2 of the relay at B, as shown in Figure 9.23.
In these cases, therefore, it is recommended that the short-circuit values for faults at
the adjacent busbar are used (in this case busbar B) but excluding the infeed from the
parallel circuit, I
C
in this case.
B
Z
Z set with
data for fault
at C
Z
3
Z
1
Z
2
C
X
Overlap
zone
Figure 9.23 Overlap in a ring system. Z
1
=zone protected by zone 2; should be
greater than 50 per cent of BC. Z
2
=unprotected zone. Z
3
represents
overlap zone seen by the zones 2 of the distance relays at A and B
Distance protection 193
9.5 The effect of arc resistance on distance protection
For a solid fault, the impedance measured by the relay is equal to the impedance up
to the fault point. However, the fault may not be solid, i.e. it might involve an electric
arc or an impedance. With arc faults it has been found that the voltage drop in the
fault and the resultant current are in phase, indicating that the impedance is purely
resistive. When dealing with an earth fault, the fault resistance is made up of the arc
resistance and the earth resistance. Faults with arc resistance are critical when they
are located close to the limits of the relay protection zones since, although the line
impedance is inside the operating characteristic, the arc resistance can take the total
resistance seen by the relay outside this characteristic resulting in under reaching in
the relay. The situation for an impedance type relay, where the effect is particularly
critical, is given in Figure 9.24.
It can be seen from Figure 9.25 that, if the characteristic angle of the relay, φ,
has been adjusted to be equal to the characteristic angle of the line then, under
fault conditions with arcing present, the relay will under reach. For this reason it is
common practice to set φ a little behind (by approximately ten degrees) in order to
be able to accept a small amount of arc resistance without producing under reaching.
From Figure 9.25, and taking into account that an angle inscribed inside a
semicircle is a right angle:
Z=(Z
R
/K) cos(−φ) (9.23)
If the angle of the protected line is equal to that of the relay, the settings are
correct. However, if the angle of the line exceeds that of the relay by ten degrees, the
A
X
R
A
Z
L
Z
L
F
R
f
R
f
B
Figure 9.24 Under reach of an impedance relay due to arc resistance
194 Protection of electricity distribution networks
A
Z=Z
R
cos (Θ– )
R
F
10°
A
Z
B
X
B
R
f
R
f
Z
R

Θ
Figure 9.25 Mho relay setting for arc faults
relay characteristic will cover 98.5 per cent of its reach, which is acceptable when the
higher cover along the horizontal axis is taken into account since a greater resistance
coverage is achieved.
9.6 Residual compensation
Earth-fault units are supplied by line-to-earth voltages and a combination of phase
currents and residual currents that depend on the relationship between the positive-
and zero-sequence impedances of the line. If a line-to-earth fault occurs, say A-E, the
sequence networks are connected in series and therefore the voltage applied to the
relay is
V
A
=V
A1
+V
A2
+V
A0
=I
A1
(Z
L1
+Z
L2
+Z
L0
) (9.24)
and, on the basis that the positive and negative-sequence impedances of a line can be
assumed to be equal, eqn. 9.24 can be re-written as
V
A
=V
A1
+V
A2
+V
A0
=I
A1
(2Z
L1
+Z
L0
) (9.25)
Since the fault is line-to-earth, I
A1
=I
A2
=I
A0
, so that I
A
=3I
A1
. The ratio
V
A
/I
A
is therefore (2Z
L1
+Z
L0
)/3 which does not equal the positive-sequence
impedance Z
L1
.
Distance protection 195
The value of residual current to be injected is calculated so that a relay that is set to
the positive-sequence impedance of the line operates correctly. Therefore, applying
the line and residual currents to the relay
I
A
+3KI
A0
=I
A
(1+K) (9.26)
and
V
A
I
A
(1+K)
=Z
L1
(9.27)
Replacing V
A
/I
A
gives
2Z
L1
+Z
L0
3
=Z
L1
(1+K) (9.28)
from which
K=
Z
L0
−Z
L1
3Z
L1
(9.29)
9.7 Impedances seen by distance relays
Distance relays are designed to protect power systems against four basic types of
fault – three-phase, phase-to-phase, phase-to-phase-to-earth, and single-phase faults.
In order to detect any of the above faults, each one of the zones of distance relays
requires sixunits –three units for detectingfaults betweenphases (A-B, B-C, C-A) and
three units for detecting phase-to-earth faults (A-E, B-E, C-E). A complete scheme
would have one set of these six units for each zone, although so-called switching
schemes use one set for one or more zones. The setting of distance relays is always
calculated on the basis of the positive-sequence impedance. Given the impossibility
of selecting exactly the right voltages and currents to cover all types of faults, each
unit receives a supply that is independent of the others in order to obtain the required
relay operation.
9.7.1 Phase units
Phase units are connected in delta and consequently receive line-to-line voltages and
the difference of the line currents. The impedances that they measure are a ratio of
voltages and currents as follows:
Z
AB
=
V
AB
I
A
−I
B
Z
BC
=
V
BC
I
B
−I
C
Z
CA
=
V
CA
I
C
−I
A
(9.30)
196 Protection of electricity distribution networks
9.7.2 Earth-fault units
As mentioned earlier in Section 9.5, earth-fault units are supplied by line-to-earth
voltages and a combination of phase and residual currents. From the previous calcu-
lations, the impedances measured by the earth-fault units of distance relays for the
three phases are
Z
A
=
V
A
I
A
¸
1+
Z
0
−Z
1
3Z
1
¸ =
V
A
I
A
+
Z
0
−Z
1
Z
1
I
0
Z
B
=
V
B
I
B
¸
1+
Z
0
−Z
1
3Z
1
¸ =
V
B
I
B
+
Z
0
−Z
1
Z
1
I
0
Z
C
=
V
C
I
C
¸
1+
Z
0
−Z
1
3Z
1
¸ =
V
C
I
C
+
Z
0
−Z
1
Z
1
I
0
(9.31)
9.8 Power system oscillations
Power system oscillations can occur, for example, after a short-circuit has been
removed from the system, or when switching operations are carried out that involve
the connection or disconnection of large quantities of load. During this phenomenon
the voltage and current that feed the relay vary with time and, as a result, the relay
will also see an impedance that is varying with time, which may cause it to operate
incorrectly.
To illustrate the situation involving a distance relay during such oscillations,
consider the equivalent circuit of the power system shown in Figure 9.26. Assume
that there is a transfer of power fromthe source of supply, S, to the most distant load at
R. The current, I
S
, which flows from S towards R causes a voltage drop in the system
elements in accordance with the vector diagram shown in Figure 9.27. The value of

S
, the phase difference between E
S
and E
R
, increases with the load transferred.
The impedance measured by the distance relay situated at A is Z=V
A
/I
S
; the
expression for this impedance can be obtained starting from the voltage V
A
which
S
A
A
Z
S
Z
L
B
B
R
Z
R
Figure 9.26 Equivalent circuit for analysis of power system oscillations
Distance protection 197
I
S
Z
S
I
S
Z
L
I
S
Z
R
S
O
R
V
A
E
R
E
S
V
B
I
S
Θ
S
Figure 9.27 Vector diagram for power system oscillation conditions
S
I
S
E
S
I
S
E
R
I
S
V
A
I
S
V
B
O
=Z
A
Z
S Z
L
Z
R
Θ
S
B
R
Figure 9.28 Impedance diagram for system in Figure 9.26
supplies the relay:
V
A
=I
S
Z
L
+I
S
Z
R
+E
R
V
A
/I
S
=Z
L
+Z
R
+E
R
/I
S
(9.32)
The last equation can be easily drawn by dividing the vectors in Figure 9.27 by
the oscillation current I
S
. In this way the diagram of system impedances, which is
shown in Figure 9.28, is obtained in which all the parameters can be assumed to be
constant except I
S
and
S
, which are variable and depend on the power transfer. The
increment of load transferred brings with it an increase in I
S
and
S
. This results in
a reduction in the size of the vector V
A
/I
S
(see Figure 9.28) and, if the increment of
load is sufficiently large, the impedance seen by the relay (V
A
/I
S
) can move into the
relay operating zones, as shown in Figure 9.29.
Figure 9.29 is obtained by constructing an R-Xplane over the locus of the relay A,
and then drawing over this the relay operating characteristic and the diagramof system
198 Protection of electricity distribution networks
Increase in Θ
S
when E
S
=E
R
S
A
V
A
/I
S
Impedance seen
by the relay
O

X
R
B
R
I
S
E
R
I
S
E
S
Z
R
Z
S
Z
L
Θ
S
Figure 9.29 Impedance seen by the relay during power system oscillations
impedances. The relayat Awill measure the value of the impedance Z
L
for a solidfault
to earth at B and continuously measure the impedance represented by AO. If a severe
oscillation occurs then the load angle
S
increases and the impedance measured by
the relay will decrease to the value AQ

, which can be inside the relay operating
characteristic. The locus of the impedance seen by the relay during oscillations is a
straight line when E
S
=E
R
, as in Figure 9.29. If E
S
>E
R
, the locus is a family of
circles centred on the SR axis. A typical trajectory which delineates the impedance
in the R-X plane during a power oscillation is shown in Figure 9.30. Consequently,
the trajectory passes inside the relay operating characteristic, indicating that there
will be a possibility for the associated breaker to be tripped in the presence of system
oscillations.
In order to prevent the operation of the relay during oscillations, a blocking
characteristic is used (see Figure 9.30). The trajectory of the impedance crosses the
characteristics of the measuring and blocking units. If the measuring units operate
within a given time, and after the blocking unit has operated, tripping of the breaker
is permitted. On the other hand, if the measuring units have not operated after a
predetermined time delay, the breaker will not be tripped. Thus, under fault conditions
when the blocking and measuring units operate virtually simultaneously, tripping
takes place. However, under power oscillation conditions, when the measuring units
operate some time after the blocking unit, tripping is prevented.
Distance protection 199
Power oscillation
with E
s
>E
r
Blocking relay
characteristic
Load characteristic
Zone 1
Zone 2
Zone 3
Measuring unit
Figure 9.30 Blocking characteristic to prevent relay operation during power system
oscillations
To prevent operation of the relay during oscillations, a power-swing blocking unit
is added. The diameter, or reach, of its characteristic for mho relays is generally 1.3 or
more times the diameter of the outermost zone of the relay, which is usually zone 3.
During fault conditions the displacement of the impedance value seen by a distance
relay is much faster than during power swings. This fact is used to set the power
swing blocking unit, which is then inhibited if there is a time elapse of typically 0.1 s
or less, to enable the impedance trajectory to move from the power-swing blocking
characteristic into zone 3 or outermost relay characteristic. Manufacturers will usually
supply recommendations for setting this unit, when provided, depending on the actual
relay types being used, and the values given above should therefore be used as general
guidelines only.
9.9 The effective cover of distance relays
In interconnected power systems in which there are supply infeeds, the effective reach
of distance relays does not necessarily correspond to the setting value in ohms. It is
possible to calculate the ratio between both, using infeed constants that have been
defined earlier. The setting value of distance relays for zones 2 and 3 is determined
200 Protection of electricity distribution networks
I
L1
I
L2
I
A
I
B
I
L3
Figure 9.31 Power system with multiple infeeds
by the following expressions:
Z
2
=Z
L1
+(1+K
1
)X
2
Z
L2
(9.33)
and
Z
3
=Z
L1
+(1+K
2
)X
2
Z
L2
+(1+K
3
)X
3
Z
L3
(9.34)
where X
2
and X
3
, the percentage of effective cover as defined in Section 9.3, are
50 per cent and 25 per cent respectively. However, in some cases, principally because
of limitations in the reach of the relays, it is not possible to set the calculated values
of Z
2
and Z
3
in the relay, and it is therefore necessary to assess the effective cover
given by the relay over adjacent lines against the actual setting value.
From the previous equations, the expression for calculating the cover of zone 2
over adjacent lines can be calculated from the following (see Figure 9.31).
X
2
=
Z
2
−Z
L1
Z
L2
(1+K
1
)
(9.35)
where: Z
2
=setting for zone 2 in ohms; X
2
Z
L2
=effective cover over adjacent line
in ohms; Z
L1
=impedance of line associated with relay; K
1
=infeed constant for the
adjacent line.
From the above equations it can be found that the expression for calculating the
effective cover of the relay over remote lines is
X
3
=
Z
3
−Z
L1
−(1+K
2
)Z
L2
Z
L3
(1+K
3
)
(9.36)
where: Z
3
=zone 3 setting in ohms; X
3
Z
L3
=effective cover over remote line in
ohms; Z
L1
=impedance of line associated with relay; K
2
=infeed constant for the
longest adjacent line; Z
L2
=impedance of the longest adjacent line; K
3
=infeed
constant for the remote line.
Using these equations and the appropriate infeed constants, it is possible to
calculate the effective reach of the relay over any of the lines adjacent to the
protected line.
9.10 Maximum load check
This check is made to ensure that the maximum load impedance will never be inside
the outermost characteristic, which is normally zone 3. To fulfil this, the distance
Distance protection 201

Z
a
D
1
.
1
Z
a
R
0
.5
5
Za

30°
Z
X
Z
C
X

Figure 9.32 Check of maximum load for mho relay
between the characteristic of zone 3 and the maximum load point should be at least
25 per cent of the distance between the origin and the maximum load point for single
circuit lines, and 50 per cent for double circuit lines.
Mho relays
Typically the mho relay operating characteristic for zone 3 has a displacement (offset)
of 10 per cent of the setting value, as illustrated in Figure 9.32. The maximum load
point is defined as
Z
c
=
V
2
S
max

30

(9.37)
In the diagram, Z
a
and φ are the setting and the characteristic angle of the relay,
respectively. From Figure 9.32
D=Z
a

1.1 Z
a
2
=0.45Z
a
(9.38)
Applying the sine theory
sin β/sin(φ −30

) =(0.45Z
a
)/(0.55Z
a
) (9.39)
from which
sin β =0.818sin(φ −30

) (9.40)
From the previous expression the value of α can be obtained. Also, from
Figure 9.32
α =180

−β −(φ −30

)
202 Protection of electricity distribution networks
and
sin α/sin(φ −30

) =Z
x
/(0.55Z
a
) (9.41)
Therefore:
Z
x
=0.55Z
a
sin α/sin(φ −30

)
For all cases, it is possible to calculate the reach of the relay in the direction of
the load by applying the last equation above. The check consists of verifying that
Z
c
−Z
x
Z
c
×100%≥P (9.42)
where P =0.5 for double circuit lines, and 0.25 for single circuit lines, as mentioned
earlier.
Relays with a polygonal characteristic
Zone 3 will be determined by the reactive and resistive settings, X
3
and R
3
,
respectively. The situation is shown in Figure 9.33.
From Figure 9.33 it can be seen that:
φ =tan
−1
(X
3
/R
3
), r =

R
2
3
+X
2
3
(φ −30

) +120

+β =180

β =90

−φ
(9.43)
Applying the sine theory
sin β
sin 120

=
Z
x
r
Z
x
=r
sin β
sin 120

(9.44)
30°
X
3
X
R R
3
120°
r
Z
X
Z
C

Figure 9.33 Check of maximum load for polygonal relay
Distance protection 203
The above equation enables the reach of polygonal relays in the direction of the
load to be determined. The distance Z
x
should satisfy the condition defined by the
inequality check, given above.
9.11 Drawing relay settings
The setting of distance relays can be represented in diagrams as time against reach
in ohms for the item of equipment being protected. The reach clearly depends on
the settings having been defined in accordance with the methodology set out in the
previous paragraphs. It should be noted that the settings calculated using the equations
are subject to two restrictions:
1. Limitations for the particular relay, when the calculated value is excessively high
and it is impossible to set the relay.
2. Limitations for the load, when the value for the reach of zone 3 approaches the
maximum load point too closely.
When the first type of restriction applies, the reach is adjusted to the maximum
available on the relay.
Example 9.1
The following case study illustrates the procedure that should be followed to obtain
the settings of a distance relay. Determining the settings is a well-defined process,
provided that the criteria are correctly applied, but the actual implementation will
vary, depending not only on each relay manufacturer but also on each type of relay.
For the case study, consider a distance relay installed at the Pance substation in the
circuit to Juanchito substation in the system shown diagrammatically in Figure 9.34,
which provides a schematic diagram of the impedances as seen by the relay. The
numbers against each busbar correspond to those used in the short-circuit study, and
shown in Figure 9.35. The CT and VT transformation ratios are 600/5 and 1000/1
respectively.
From the criterion for setting zone 1
Z
1
=0.85Z
10–11
=0.85(7.21

80.5

) =6.13

80.5

primary ohms
and for zone 2
Z
2
=Z
10–11
+0.5(1+K
1
)Z
11–9
In this case the infeed constant is defined as:
K
1
=
I
14–11
+I
17–11
+I
5–11
+I
18–11
I
10–11
for a fault at busbar 11.
From the short-circuit values in Figure 9.35
K
1
=
1333.8

−85.54

+0+5364.6

−85.88

+449.9

−86.34

2112.6

(−85.55

)
204 Protection of electricity distribution networks
1
1
J
u
a
n
c
h
i
t
o

1
1
5
0
.
7
1
2



8
2
.
9
0
°
Ω
J
u
a
n
c
h
i
t
o

2
2
0
9
1
0
6
.
7
9



8
0
.
5
9
°
Ω
8
.
0
7



6
8
.
7
1
°
Ω
Z
0
=
3
6
.
9
5



7
5
.
6
6
°
Ω
P
a
n
c
e

1
1
5
Z
1
=
7
.
2
1



8
0
.
5
0
°
Ω
D
i
e
s
e
l

1
1
5
1
8
5
.
9
2



8
0
.
4
7
°
Ω
5
Y
u
m
b
o

1
1
5
M
e
l
e
n
d
e
z

1
1
5
1
4
1
7
C
a
n
d
e
l
a
r
i
a

1
1
5
1
1
.
9
5



9
0
.
0
°
Ω
2
2
0
k
V
1
1
5
k
V
1
1
5
k
V
1
1
5
k
V
1
1
5
k
V
1
1
5
k
V
1
1
5
k
V
F
i
g
u
r
e
9
.
3
4
I
m
p
e
d
a
n
c
e
d
i
a
g
r
a
m
f
o
r
E
x
a
m
p
l
e
9
.
1
s
h
o
w
i
n
g
i
m
p
e
d
a
n
c
e
s
s
e
e
n
b
y
a
r
e
l
a
y
o
n
t
h
e
J
u
a
n
c
h
i
t
o
c
i
r
c
u
i
t
a
t
P
a
n
c
e
s
u
b
s
t
a
t
i
o
n
Distance protection 205
1
1
J
u
a
n
c
h
i
t
o

1
1
5
2
.
2
3
2
8

9
0
.
0
8
°
k
A
J
u
a
n
c
h
i
t
o

2
2
0
9
1
0
P
a
n
c
e

1
1
5
D
i
e
s
e
l

1
1
5
1
8
5
Y
u
m
b
o

1
1
5
M
e
l
e
n
d
e
z
-
1
1
5
1
4
1
7
C
a
n
d
e
l
a
r
i
a

1
1
5
1
0
.
4
2
4
9

8
6
.
2
5
°
k
A
2
.
1
1
2
6

8
5
.
5
5
°
k
A
1
.
3
3
3
8

8
5
.
5
4
°
k
A
0
.
0


0
.
0
°
k
A
5
.
3
6
4
6

8
5
.
8
8
°
k
A
0
.
4
4
9
9

8
6
.
3
4
°
k
A
F
i
g
u
r
e
9
.
3
5
F
a
u
l
t
c
u
r
r
e
n
t
c
o
n
t
r
i
b
u
t
i
o
n
f
o
r
E
x
a
m
p
l
e
9
.
1
206 Protection of electricity distribution networks
In making this calculation it should be noted that the values of current are referred
to the receiving busbar.
This gives
K
1
=
7148.27

−85.87

2112.6

−85.5

=3.38

−0.37

so that 1+K
1
=4.38.
Therefore, the setting for zone 2 is
Z
2
=7.21

80.50

+(4.38×0.356

82.90

) =8.77

80.93

primary ohms
and the setting for zone 3 is
Z
3
=Z
10–11
+(1+K
2
)Z
11–17
+0.25(1+K
3
)Z
transformer
In this case, the infeed constant K
3
will be taken to be the same as K
2
since under
reach on that section is not significant. This approach is common when determining
zone 3 settings.
For a fault on busbar 11, the infeed constant is defined as
K
2
=
I
9–11
+I
14–11
+I
5–11
+I
18–11
I
10–11
Thus:
K
2
=
2232.8

−90.08

+1333.8

−85.54

+5364.6

−85.88

+449.9

−86.34

2112.6

−85.55

i.e.
K
2
=
9376.72

−86.86

2112.6

−85.5

=4.44

−1.36

so that
1+K
2
=5.44

−1.10

Therefore the setting for zone 3 is
Z
3
=7.21

80.50

+(5.44

−1.10

×8.07

68.71

)
+(1+4.44

−1.36

)11.95

90

=114.35

80.20

The relay settings, in primary ohms, can be summed up as follows:
Z
1
=6.13

80.5

Z
2
=8.77

80.93

Z
3
=114.35

80.20

The secondary ohms are calculated using the following expression:
Z
sec
=Z
prim
×
CTR
VTR
In this case CTR/VTR=120/1000=0.12, and, therefore, Z
1
=0.736, Z
2
=
1.052, and Z
3
=13.72.
Distance protection 207
Starting unit settings
The starting unit is set by taking 50 per cent of the maximum load impedance.
Given that the power transferred from Pance to Juanchito is S =30.4+j13.2MVA=
33.14MVA, then
Z
c
=
V
2
P
=
115
2
33.14
=399.03 primary ohms =47.88 secondary ohms
Residual compensation constant setting
K
1
=
Z
0
−Z
1
3Z
1
where K
1
=residual compensation constant and Z
1
, Z
0
=line positive- and zero-
sequence impedances giving
K
1
=
36.95

75.66

−7.21

80.50

3(7.21

80.50

)
=1.377

−6

K=1.4
Time setting
Time delay for zone 2=0.4 s
Time delay for zone 3=1.0 s
Load check
The setting of the unit that determines the longest operating characteristic of the relay
should be checked to make sure that it does not overlap the load zone.
The reach of the relay in the direction of the load is determined as follows:
sin β =0.818sin(φ −30

)
where the relay setting φ =75

.
sin β =0.818sin(45

)
i.e.
β =35.34

α =180

−β −(φ −30

)
giving α =99.66

so that the reach will be
Z
x
=
0.55 Z
3
sin 99.66

sin 45

=
0.55(114.35) sin 99.66

sin 45

Z
x
=87.68 primary ohms
The distance to the load point, expressed as a percentage, is
%=
399.03−87.68
399.03
×100%=78.03%
Therefore, it is concluded that the setting is appropriate and does not require adjusting
in reach as a result of the load.
208 Protection of electricity distribution networks
Determination of the effective cover
In accordance with the calculated settings, in zone 2 the relay covers 50 per cent of
the line 11-9. However, it is important to determine the cover of this setting along the
Juanchito-Yumbo 115 kV line (11-5) and, for this, the following expression is used:
X
2
=
Z
2
−Z
L1
Z
L2
(1+K
1
)
The infeed constant K
1
at substation Juanchito (no. 11) is given by
K
1
=
I
9-11
+I
14-11
+I
17-11
+I
18-11
I
10-11
for a fault on busbar 11.
Since it is known that the angles of the infeed constants are close to zero, their
values can be calculated using magnitudes only:
K
1
=
2232.8+1333.8+0+449.9
2112.6
=1.90
Therefore, the effective cover along the Juanchito-Yumbo line is
X
2
=
8.77−7.21
6.79(1+1.90)
=0.079=7.9 per cent
As expected, the reach of zone 2 is less than 50 per cent of the Juanchito-Yumbo
line, since the shortest line is the Juanchito 115 to Juanchito 220 circuit. The reaches
of the relays are given in Figure 9.36.
Example 9.2
For the power system in Figure 9.37, calculate the zone 2 setting for the relay at San
Antonio with the infeed constant based on the results for a fault in the Chipichape
substation, taking into account the criteria stated earlier in this chapter.
Determine the actual impedances that the relay sees for a fault on the Chipichape-
Yumbo line and on the busbar at Yumbo and, using these, determine whether zone 2
of the relay operates for these faults.
Incalculatingthe zone 2reach, checkthe amount of line coveredif the contribution
from the Yumbo-Chipichape line is neglected. Work with primary impedances.
Solution
The second zone of the relay extends to the Chipichape-Yumbo line, which is the
shortest adjacent line seen by the relay at San Antonio.
In the process, the initial fault point was taken as being at Chipichape, and then
moved in 10 per cent steps along one of the Chipichape-Yumbo circuits. With the aid
of a computer it was possible to determine the value of K for each point by making
a busbar called TEST the fault point. In this way the value of K for each case was
calculated using the following expression:
K=
¸
(I
SA-Ch
/2) +I
BA-Ch
+I
D-Ch
+I
Y-Ch
I
SA-Ch
/2
¸
Distance protection 209
5
.
8
T i m e i n s e c o n d s
1
.
2
1
.
0
0
.
4
1
=
2
3
.
3
8



6
9
.
5
2
°
Ω
=
2
.
0
8



6
7
.
1
8
°
Ω
=
8
.
0
7



6
8
.
7
0
°
Ω
=
1
0
.
2
5



8
0
.
2
8
°
Ω
Z
L
C
h
i
p
i
c
h
a
p
e

B
.

A
n
c
h
i
c
a
y
a
Z
L
J
u
a
n
c
h
i
t
o

1
1
5

J
u
a
n
c
h
i
t
o

2
2
0
Z
L
Y
u
m
b
o

C
h
i
p
i
c
h
a
p
e
Z
L
J
u
a
n
c
h
i
t
o

Y
u
m
b
o
Z
L
P
a
n
c
e

J
u
a
n
c
h
i
t
o
P
a
n
c
e

1
1
5
M
e
l
e
n
d
e
z
J
u
a
n
c
h
i
t
o

1
1
5
6
.
0
J
u
a
n
c
h
i
t
o

2
2
0
0
.
3
C
a
n
d
e
l
a
r
i
a
=
5
.
7
9



6
9
.
2
5
°
Ω
=
6
.
7
9



8
0
.
5
9
°
Ω
=
7
.
2
1



8
0
.
5
0
°
Ω
=
2
7
.
5
9


6
7
.
1
6
°
Ω
=
0
.
7
1
2


8
2
.
9
0
°
Ω
1
.
0
4
.
9
C
h
i
p
i
c
h
a
p
e
Y
u
m
b
o

1
1
5
S
.

A
n
t
o
n
i
o
S
t
a
.

B
a
r
b
a
r
a
1
.
0
B
u
g
a
Z
L
J
u
a
n
c
h
i
t
o

1
1
5

C
a
n
d
e
l
a
r
i
a
Z
L
C
h
i
p
i
c
h
a
p
e

S
.

A
n
t
o
n
i
o
Z
L
Y
u
m
b
o

S
a
n
t
a

B
a
r
b
a
r
a
Z
L
Y
u
m
b
o

B
u
g
a
L
i
n
e

t
o

B
a
j
o

A
n
c
h
i
c
a
y
á
1
9
.
9
6
2
.
1
F
i
g
u
r
e
9
.
3
6
R
e
a
c
h
o
f
d
i
s
t
a
n
c
e
p
r
o
t
e
c
t
i
o
n
r
e
l
a
y
s
f
o
r
E
x
a
m
p
l
e
9
.
1
210 Protection of electricity distribution networks
Chipichape
6
27.56 67.16° Ω
13
Bajo
Anchicayá
San Antonio
12
2.0 67.18° Ω
Diesel Yumbo
5.79 69.25° Ω
Test
5
Figure 9.37 Power system for Example 9.2
It can be seen that, in this case, I
inf
+I
relay
=I
Ch-TEST
.
Now
I
inf
+I
relay
I
relay
=K+1
and, therefore, the value of (1+K) can be calculated from the formula
I
Ch-TEST
(I
SA-Ch
)/2
The computer results for a fault at different locations between Chipichape and
Yumbo are given in Table 9.1. Knowing the value of K for each fault point, the value
of the actual Z seen by the relay can be calculated.
In Figure 9.38 the following curves are shown:
• line impedance as a function of the distance from Chipichape;
• actual impedance from Chipichape seen by the relay at San Antonio;
• value of (1+K).
The setting values have also been drawn for a fault at Chipichape with, and
without, the contribution from Yumbo which is variable. It should be noted that,
considering the contribution of the parallel Chipichape-Yumbo line, cover in zone 2
goes up to 68 per cent and then from 94 per cent to the end of the line. This means
that overlap could occur at the end of zone 2 of the relay at Chipichape.
Distance protection 211
T
a
b
l
e
9
.
1
V
a
l
u
e
s
o
f
c
u
r
r
e
n
t
a
n
d
i
m
p
e
d
a
n
c
e
f
o
r
f
a
u
l
t
s
o
n
t
h
e
C
h
i
p
i
c
h
a
p
e
-
Y
u
m
b
o
l
i
n
e
L
o
c
a
t
i
o
n
o
f
t
h
e
f
a
u
l
t
(
%
)
I
r
e
l
a
y
=
(
I
S
A
-
C
h
)
/
2
I
C
h
-
f
a
u
l
t
I
Y
-
C
h
i
n
p
a
r
a
l
l
e
l
l
i
n
e
I
n
f
e
e
d
c
u
r
r
e
n
t
1
+
K
Z
r
e
a
l
f
r
o
m
C
h
=
(
1
+
K
)
Z
L
0
(
C
h
)
1
5
3
4
.
8


8
7

7
7
4
8
.
6


8
3

3
2
4
3
.
5


8
2

6
2
2
7
.
1


8
3

5
.
0
4
8
0
1
0
1
2
9
5
.
7


8
7

6
1
1
7
.
8


8
5

2
1
7
5
.
9


8
3

4
8
2
2
.
1


8
5

4
.
7
2
1
2
.
7
3
3
2
0
1
2
1
3
.
0


8
6

5
5
9
1
.
5


8
5

1
8
5
8
.
8


8
3

4
3
7
8
.
5


8
5

4
.
6
0
9
5
.
3
3
7
3
0
1
1
3
5
.
0


8
6

5
0
8
4
.
6


8
5

1
5
4
6
.
6


8
4

3
9
4
9
.
9


8
5

4
.
4
7
9
7
.
7
8
0
4
0
1
0
6
1
.
2


8
5

4
5
9
4
.
5


8
5

1
2
3
7
.
8


8
4

3
5
3
3
.
5


8
4

4
.
3
2
9
1
0
.
0
2
5
5
0
9
9
0
.
7


8
5

4
1
1
4
.
6


8
5

9
2
8
.
1


8
5

3
1
2
3
.
9


8
5

4
.
1
5
3
1
2
.
0
2
2
6
0
9
2
2
.
7


8
4

3
6
3
8
.
2


8
5

6
1
2
.
7


8
7

2
7
1
5
.
2


8
5

3
.
9
4
3
1
3
.
6
9
7
7
0
8
5
5
6
.
2


8
4

3
1
5
8
.
9


8
5

2
8
8
.
2


9
2

2
3
0
2
.
3


8
5

3
.
6
8
9
1
4
.
9
5
1
8
0
7
9
0
.
6


8
3

2
6
6
9
.
0


8
4

7
0
.
8

+
1
3
2

1
8
7
8
.
4


8
4

3
.
3
7
5
1
5
.
6
3
3
9
0
7
2
5
.
5


8
2

2
1
6
4
.
6


8
4

4
2
2
.
0

+
1
0
1

1
4
3
9
.
0


8
4

2
.
9
8
3
1
5
.
5
4
4
1
0
0
(
Y
)
5
7
6
.
8


8
0

1
1
1
5
.
8


8
1

1
1
1
5
.
8

+
9
9

5
3
9
.
8


8
3

1
.
9
3
4
1
1
.
1
9
7
212 Protection of electricity distribution networks
20%
I
m
p
e
d
a
n
c
e

i
n

o
h
m
s

(
Ω
)
C
o
e
f
f
i
c
i
e
n
t

(
1
+
k
)
0
1
2
Chipichape
2
1
0
0 10%
6
3
5
4
2.73
3
4
5
8.53Ω
11
8
7
10
9
14.84
5.059
16
15
13
12
14
4.72
15.64
3.38
Coefficient (1 +k)
Real impedance seen by
relay at San Antonio
80% 50%
Line impedance as a function
of the distance x
30% 40%
5.34
60% 70%
Setting without the infeed
of the parallel circuit
4.61
4.48
7.78
4.33
10.03
4.15
3.94
3.69
13.70
12.02
14.95
5.79 Ω
Yumbo
90%
X
1.93
2.98
11.20Ω
15.54Ω
Setting considering the
infeed of the parallel circuit
Figure 9.38 Impedance curves for Example 9.2
9.12 Intertripping schemes
Used on high voltage lines for many years, distance protection has proved to be a
very reliable form of protection, notwithstanding the limitation that its first zone of
coverage does not protect the whole of the circuit, comparedtounit protectionschemes
such as differential protection. With rapid, reliable and economic communication
links now available, this limitation can be overcome by providing links between the
relays at each end of the line and adding intertripping to distance protection schemes.
In addition, distance protection has the advantage of being able to act as back up
to protection located at substations further along the line. Although there is a wide
Distance protection 213
variety of intertripping schemes, only the more common ones will be considered here.
As with any protection system, the selection of a particular scheme depends on the
criteria adopted by individual utilities, the communication links available, and the
importance of the circuits being protected. Ultimately it is the fact that these schemes
can rapidly clear faults at the far end of the line, outside zone 1, which favours the
use of intertripping.
9.12.1 Under reach with direct tripping
In this case, the settings of the distance relay that protects a line follow the criteria
referred to in Section 9.3. When the zone 1 units of the relays operate, they initiate a
signal that is sent along the communications linktotrigger animmediate trippingat the
other end of the line. The scheme is simple and has the advantage of being extremely
fast; however, it has the disadvantage that it may set off undesirable circuit breaker
tripping if there is any mal-operation of the communication equipment. Figure 9.39
illustrates the operation of the scheme with the distance relay located at one end of
the line.
9.12.2 Permissive under reach intertripping
This scheme is similar to the one described in the previous subsection, but differs
in that the zone 2 unit at the receiving end has to detect the fault as well before the
trip signal is initiated. The advantage of this scheme is that spurious trip signals are
blocked. Thus, proper account is taken of what has happened to the zone 2 unit of the
relay that received the signal from the relay located at the other end. In some cases
it is necessary to include a time delay to the tripping command from the remote end,
particularly when there are double circuit lines fed by one source located at one end
of the lines. Figure 9.40 shows the operation of the scheme with the distance relay
located at one end of the line only.
(a) (b)
To remote
terminal
Trip
relay
Receive relay
contact
85
TC

Send
circuit
+ +
Z
3
Z
3t
Z
2t
Z
2
Z
2t
Z
1
Z
1
Z
3t
Figure 9.39 Under reach with direct trip: (a) contact logic of trip circuit; (b) contact
logic of signalling channel send scheme
214 Protection of electricity distribution networks
(a) (b)
To remote
terminal
Trip
relay
85
TC

Send
circuit
+ +
Z
3
Z
3t
Z
2t
Z
2
Z
2t
Z
1
Z
1
Z
3t
Z
2
Figure 9.40 Permissive under reach intertripping: (a) contact logic of trip circuit;
(b) contact logic of signalling channel send scheme
(a) (b)
To remote
terminal
Trip
relay
85
TC

Send
circuit
+ +
Z
3
Z
3t
Z
2t
Z
2
Z
2t
Z
1
Z
2
Z
3t
Z
2
Figure 9.41 Permissive over reach intertripping: (a) contact logic of trip circuit;
(b) contact logic of signalling channel send scheme
9.12.3 Permissive over reach intertripping
With this arrangement, the operation is very similar to that referred to above, except
that sending the trip signal from a relay to the other end is carried out as a result of the
zone 2, and not the zone 1, unit operating. Again, tripping of the relay that receives
the signal is dependent on its zone 2 unit having seen the particular fault as well. The
schematic diagram for this arrangement is given in Figure 9.41.
9.13 Distance relays on series-compensated lines
The compensation of lines using series capacitors has proved to be an effective method
of increasing the efficiency of the power transfer along the circuit. The main reasons
Distance protection 215
for using line compensation are:
• improvement in reactive power balance;
• reduced system losses;
• improved voltage regulation;
• improved system transient stability performance;
• increased power transfer capability.
As regards the last point, the active power transfer from one system (1) to another (2)
is given by the expression P ={V
1
V
2
sin(φ
1
−φ
2
)}/X. In the case of a line, the intro-
duction of a series capacitor reduces the overall line reactance and therefore increases
the amount of active power that can be transferred. The amount of compensation is
usually quoted as the percentage of the line inductive reactance that is compensated
by the series capacitor. Values for line compensation are normally within the range
of 20 to 70 per cent .
The addition of series compensation can have serious effects on the performance
of the protection system, especially distance relays, related to voltage and/or current
inversion and the change of impedance seen by the relay. Figure 9.42 shows the
apparent impedance seen by a relay at A when 50 per cent series compensation is
applied at the middle of the line. Faults beyond the series capacitor appear closer and,
therefore, the zone 1 setting should be adjusted to have a smaller impedance setting
to avoid over reach. Figure 9.43 corresponds to the case where 70 per cent series
compensation is applied at the near end of the line at A. In this case, the relay can
see the fault in the reverse direction so that the setting must rely on the memory and
biasing of the healthy phases to guarantee proper operation of the protection.
9.14 Technical considerations of distance protection in tee circuits
In the application of distance relays in tee circuits, special attention should be paid
to the infeed effect due to the terminals that make up the tee. There can be infeeds at
two or three of the terminals, which will require special attention in each case.
9.14.1 Tee connection with infeeds at two terminals
The situation is illustrated in Figure 9.44 where it is assumed that there is no generation
at busbar C. The infeed current I
B
results in the distance relay at the busbar A seeing
an apparent impedance that is greater than the true impedance to the fault point. For
a fault at F, the relay at A is supplied by the following voltage:
V
A
=I
A
Z
1
+(I
A
+I
B
)Z
3
(9.45)
Thus, the apparent impedance seen by the relay at A is:
Z
A
=
V
A
I
A
=Z
1
+
¸
1+
I
B
I
A
¸
Z
3
Z
A
=Z
1
+(1+K
A
)Z
A
(9.46)
where K
A
is defined as the system infeed constant.
216 Protection of electricity distribution networks
X
A
21
A
ACDB normal condition
ACDЈBЈ gap operated
D

C
R
B
GAP
X
AC
=50%
X
CD
=50%
X
DB
=50%
C

D
B
Figure 9.42 Apparent impedance with series compensation at the middle of the line
Since under normal conditions K
A
is larger than one, the apparent impedance
seen by the relay at A, Z
A
, is greater than the actual fault impedance and therefore
the relay under reaches the intended cover along the line OC if the infeed is not taken
into account. Equally, the relay at B is seeing an apparent impedance given by:
Z
B
=Z
2
+
¸
1+
I
B
I
A
¸
Z
3
Z
B
=Z
2
+(1+K
B
)Z
3
(9.47)
On the other hand, the relays at A and B should be set in such a way that their
zone 1 reaches do not go beyond busbars B and C for the relay at A, and busbars A
and C for the relay at B. If not, faults on the transformer in substation C could initiate
tripping of the line AB. Zone 1 of the relay at A should be adjusted with the smaller
Distance protection 217
70%
D
A–C
X
A
ACDB normal condition
ABЈ gap operated
100%

R
GAP

B
21
C D
X
CD
=70%
X
DB
=100%
B
Figure 9.43 Apparent impedance with series compensation at the beginning of
the line
of the following values:
Z
1
=0.85Z
AB
Z
1
=0.85(Z
AO
+Z
OC
)
(9.48)
This guarantees maximumcover over sections OBand OC, without the possibility
of over reach from the relay at substation A when the infeed current I
B
disappears.
However, this will initiate an under reach of the relay under normal conditions when
218 Protection of electricity distribution networks
I
A
I
B
A
F
C
Z
1
Z
2
Z
3
O
B
Figure 9.44 Tee circuit with infeeds at A and B
current I
B
is present. One alternative would be to amend the equation to
Z
1
=0.85
¸
Z
AB
+

1+
I
B
I
A

Z
OC
¸
(9.49)
Although this guarantees effective cover over the line OC, eqn. 9.49 has the
disadvantage of allowing over reaches if the circuit is open at B. If the impedance
Z
OC
is much smaller in comparison to the impedance Z
OB
, then the cover of the relay
at A over section OB will be much less reduced compared to that which would be
achieved by the relay at A if the tee-off did not exist at O. In such a case it would be
essential to provide an intertrip in order to accelerate the zone protection to protect
this section of line adequately.
9.14.2 Tee connection with infeeds at all three terminals
If a supply infeed exists at the end of the tee-off, as illustrated in Figure 9.45, the
infeed effect is still obtained for faults on the line AB resulting in under reach of the
relays at A and B. In such a situation, the relay at A would see the following apparent
impedance for a fault F, as indicated in Figure 9.45.
Z
A
=
V
A
I
A
=Z
1
+
¸
1+
I
C
I
A
¸
Z
C
(9.50)
The value of Z
A
is greater than the value of the actual fault impedance (Z
1
+Z
2
),
resulting in under reach in relay A. The settings at A should therefore be calculated
on the basis of the actual system impedances, without considering the infeed effect,
in order to avoid over reaches when one or more of the terminals of the tee are open.
Distance protection 219
I
A
I
B
I
C
A
Z
1
Z
3
Z
2
C
F
B
Figure 9.45 Tee circuit with infeeds at A, B and C
With this criterion, optimumco-ordination will be obtained but the reach of the relays
is reduced by the effect of the infeed at the terminals.
9.15 Use of distance relays for the detection of the loss of
excitation in generators
An excitation fault on a generator produces loss of synchronism with a consequent
reduction in power generated and overheating in the windings. The quantity that varies
most when a generator loses synchronism is the impedance measured at the terminals
of the stator. Under loss of excitation, the voltage at the terminals will begin to fall and
the current will increase, resulting in a reduction in this impedance and a change in
power factor. The generator and associated power systemcan be represented as shown
in Figure 9.46. The voltage vector diagramis given in Figure 9.47, and the impedance
vector diagram in Figure 9.48, for a relay for detecting loss of synchronism located at
point A. The impedance seen by the relay, when there are variations in the magnitudes
of E
G
, E
S
, I and δ, are circles with their centres along the straight line CD.
Whena generator that is operatingsynchronouslyloses excitation, the ratio E
G
/E
S
decreases and the angle δ increases. This condition, when drawn in the impedance
plane, represents a movement of the load point (or of the impedance seen by the
relay) in the direction shown in Figure 9.49. A relay with mho characteristics that has
only two settings – the offset and the diameter – is used for detecting this condition
as shown in Figure 9.50. The offset is designed to prevent operations during system
oscillations when excitation has not been lost and protection against asynchronous
operation is required. The diameter setting should enable power to be supplied to
220 Protection of electricity distribution networks
E
G
=I(X
G
+X
T
+Z
S
) +E
S
E
G X
G
X
T
Z
S
A
I
E
S
Figure 9.46 Equivalent system to analyse loss of field
V
A
=E
S
+I(Z
S
+X
T
)

E
G
E
S
I
I(X
G
+X
T
+Z
S
)
IX
S
IX
T
IX
G
Figure 9.47 Voltage vector diagram for system of Figure 9.46
0.7
C
0.6
0.4
X
T
–j
X
A
j
X
Z
S
Load
point
3.0
D
2.0
R
E
G
E
S
=1.5
E
G
E
S
=1.0
Figure 9.48 Impedance vector diagram for Figure 9.46
Distance protection 221
Trajectory for oscillations
without loss of excitation
Trajectory for oscillations
with loss of excitation
C
A
D
Load point
R
j
X
E
G
E
S
=1.5
Figure 9.49 Load point movement
Offset
Diameter
j
X
R
Figure 9.50 Mho relay offset and diameter settings
222 Protection of electricity distribution networks
loads with leading power factors. A diameter setting value of 50 to 100 per cent of
X
d
will typically guarantee protection against asynchronous operation.
9.16 Exercises
Exercise 9.1
Demonstrate that the operating characteristic of a mho relay, better known as an
admittance relay, is a straight line in an admittance diagram. The crossing values of
the line should be indicated in each one of the axes.
Exercise 9.2
For the power system shown in Figure 9.51 calculate:
(i) The fault resistance, if the fault current is 200 A.
(ii) The value of the residual compensation constant.
(iii) The secondary impedance that the relay sees if it is used with a residual
compensation constant equal to 1.0 (100 per cent).
Note: The CT ratio is 800/1, and the VT ratio is 11800/

3: 110/

3.
Exercise 9.3
For the power system shown in Figure 9.52 determine the reach in secondary ohms
for zone 3 of the distance relay installed in the Juanchito substation, on the line that
goes to Pance substation.
Make a check of the proximity of the maximum load.
Calculate the infeed constants to cover the adjacent and remote lines, considering
the intermediate infeeds associated with busbar 7 only.
A
Earthing transformer
Z
0
= j6Ω
Z
T1
=Z
T
=j3Ω
Z
L1
=2.3 +j5.7Ω
Z
L0
=3.5 +j24Ω
19 Ω
45 MVA
Phase to
earth fault
R
Fault current =200A
B
Figure 9.51 Power flow for Exercise 9.2
Distance protection 223
J11.61 Ω
21
Juanchito–220
11.40 83.48° Ω
9
7
27.84 82.45° Ω
Alto
10
*
Pance–115
25.05 83.51° Ω
27.64 82.45° Ω
13.72 82.42° Ω
Pance–220
8
* Equivalent of two
autotransformers
Salvajina–220 2
Yumbo–220
1
Anchicaya–220
220 kV
220 kV
115 kV
220 kV
220 kV
220 kV
Figure 9.52 Power system for Exercise 9.3
21
Juanchito–220 9
7
Alto
10 Pance–115
Pance–220 8
Salvajina–220 2
Yumbo–220
1
Anchicaya–220
0.9071 –86.58° kA
0.9071 –86.58° kA
8.3504 –86.77° kA
2.4467 –86.18° kA
1.2107 –87.60° kA
1.6012 –87.04° kA 0.3815 –89.74° kA
1.2779 –87.07° kA
Figure 9.53 Fault current contribution for Exercise 9.3
224 Protection of electricity distribution networks
Note:
• transformation ratios of the CTs and VTs are 800/5 and 2000/1, respectively;
• three-phase short-circuit values for a fault on busbar 7 are given in Figure 9.53;
they correspond to line values (not per circuit);
• relay setting angle is 75

;
• maximum load per circuit for the Juanchito-Pance line is 40 MVA with an angle
of ±30

.
Chapter 10
Protection of industrial systems
With the increase in size of industrial plant electrical systems, and the high
short-circuit levels encountered on electricity power systems, it is essential that the
electrical protection arrangements in any industrial installation are correctly designed
and have the appropriate settings applied to ensure the correct functioning of the plant
and continuity of supply within the installation. The importance of maintaining con-
tinuity of supply to industrial installations cannot be over emphasised and, in this
respect, the interconnectors to the public supply system play a vital role. It is crucial
that correct co-ordination is maintained between the protection on the main industrial
supply infeeds and the power system supply feeders.
10.1 Protection devices
In addition to the overcurrent relay, which has been covered in Chapter 5, moulded-
case circuit breakers (MCCBs) and thermal relay-contactor and fuse devices are
frequently used to protect elements of the industrial electrical system and these will
be covered in more detail in this chapter.
10.1.1 Overcurrent relays
This type of relay is usually equipped with an instantaneous and/or time-delay unit.
This latter unit can be inverse, very inverse or extremely inverse, and when it is
necessary to prevent the relay from operating in one direction it should be provided
with a directional element. This is required for ring systems or networks with several
infeed sources, the latter being very common in industrial systems.
10.1.2 Direct acting devices in power and moulded-case circuit breakers
As their name indicates, these are devices that act directly on power breakers and
therefore do not require AC or DC coils for tripping. They are especially common
226 Protection of electricity distribution networks
for operating breakers up to 600 V, but are sometimes used on breakers of a higher
voltage and, in these cases, are generally fed by current transformers.
These devices can be operated by:
1. An armature attracted by the electromagnetic force that is produced by the fault
current which flows through a trip coil.
2. A bimetallic strip that is actuated by the heat produced by the fault current.
The characteristic curves of these mechanisms are generally the result of com-
bining the curves of instantaneous relays with long or short time-delay relays. The
starting current of the long delay units can normally be set to 80, 100, 120, 140 and
160 per cent of the nominal value. The calibrations of short time-delay relays are
typically 5, 7.5 and 10 times the nominal value.
It is common to use breakers to protect circuits of low current capacity with
a combination of a single time-delay element plus an instantaneous unit normally
having a bimetallic or magnetic element. In this case, the curves are normally set in
the factory based on the nominal values.
10.1.3 Combined thermal relay contactor and fuse
The combined thermal relay contactor and fuse is used extensively for protection,
mainly in low power systems. In this case the fuse provides protection against short-
circuits and the thermal relay gives protection against overloads. Given that the ther-
mal relay acts directly on the contactor, special care should be taken to prevent a relay
operation for values of fault current that exceed the capacity of the contactor. If this
should be the case, a more rapidly acting fuse should be selected in order to guarantee
that it will operate for any current greater than the breaking capacity of the contactor.
10.2 Criteria for setting overcurrent protection devices associated
with motors
The criteria normally used for the selection of the nominal values and the range of
settings of low voltage overcurrent devices such as thermal relays and moulded-case
breakers that are usedfrequentlyinindustrial plants are similar tothose for overcurrent
protection included in Chapter 5. An important consideration for these devices is that,
as for overcurrent relays, the selected settings can vary depending upon the criteria
adopted by the particular utility or plant operators, providing that the resultant settings
guarantee appropriate protection to the machines and the elements of the systemunder
analysis. Thus, the settings should be higher than the motor-locked rotor current and
below the motor thermal limit.
10.2.1 Thermal relays
A thermal relay basically has three parameters that can be adjusted: the rating of the
coil, the range of taps in the thermal element, and the range of the instantaneous
element.
Protection of industrial systems 227
Coil rating
In order to determine the coil rating, manufacturers provide a range of maximum and
minimum current values for which the thermal relay has been designed. The rating of
the coil is somewhat above the maximum value of the motor secondary rated current.
The range of a thermal relay for a motor should overlap the motor manufacturer’s
value of the motor nominal secondary current, in amperes. Using a thermal relay that
has a maximum current rating very close to the motor nominal secondary current (in
amperes) is not recommended; in this case it is better to use the next highest range
available.
Example 10.1
Consider a motor with the following characteristics:
power: 100HP, p.f. =0.8
voltage: 440V
efficiency: 100 per cent
The thermal protection consists of three single-phase relays, fed from a set of
current transformers with 200/5 ratios. The setting range of the relays is given in
Table 10.1, and the operating characteristics are shown in Figure 10.1.
I
nom
=122.36A, and I
nom(sec)
=122.36×5/200=3.06A. In accordance with
the data in Table 10.1, a thermal coil with a rating of 3.87 A, which has a current
range of 3.10 to 3.39 A, can be selected. A relay with a range of 2.82–3.09 A should
not be used since the maximum value is very close to the I
nom(sec)
of the motor.
Range of taps in the thermal element
The tripping current of the thermal element of the relay is normally specified with
a range of 90/95/100/105/110 per cent of the current rating of the coil.
Range of instantaneous elements
The instantaneous elements are specified at ten times the nominal current of the motor.
Typically this range is 6 –150 A. For this example:
I
inst
=10×I
nom
=10×122.36×5/200=30.59A (secondary)
This confirms that the 6–150 A range is appropriate.
10.2.2 Low voltage breakers
The low voltage breakers used to protect motors usually have two elements: a time-
delay unit for long-time overloads and an instantaneous element for short-circuits.
The short-time element is optional and recommended only for the more powerful
motors, or when the possibility exists of losing co-ordination with other breakers
located nearer to the source.
The values that should be specified for a breaker are as follows: the nominal
current, and the setting ranges for the time-delay unit, the instantaneous unit, and
also the short-time unit if this is fitted.
228 Protection of electricity distribution networks
T
a
b
l
e
1
0
.
1
S
e
t
t
i
n
g
s
f
o
r
t
y
p
e
T
M
C
2
1
B
a
n
d
T
M
C
2
3
B
r
e
l
a
y
s
A
d
j
u
s
t
e
d
m
o
t
o
r
f
u
l
l
-
l
o
a
d
c
u
r
r
e
n
t
,
A
m
i
n
m
a
x
T
h
e
r
m
a
l
c
o
i
l
r
a
t
i
n
g
,
A c
o
n
t
i
n
u
o
u
s
T
a
r
g
e
t
u
n
i
t
,
A 0
.
2
/
2
I
n
s
t
a
n
t
a
n
e
o
u
s
u
n
i
t
,
A 6
-
1
5
0
M
o
d
e
l
n
u
m
b
e
r
s
H
a
n
d
r
e
s
e
t
t
h
e
r
m
a
l
u
n
i
t
C
a
s
e
s
i
z
e
S
-
1
A
p
p
r
o
x
.
w
t
.
l
b
(
k
g
)
n
e
t
s
h
i
p
1
2
(
5
.
4
)
1
6
(
7
.
3
)
5
0
/
6
0
H
z

6
0
H
z

s
t
a
n
d
a
r
d
t
i
m
e
s
h
o
r
t
t
i
m
e
1
.
3
2
1
.
4
5
1
.
6
4
1
2
T
M
C
2
1
B
1
A
1
2
T
M
C
2
3
B
1
A
1
.
4
6
1
.
5
9
1
.
8
2
B
2
A
B
2
A
1
.
6
0
1
.
7
5
2
.
0
0
B
3
A
B
3
A
1
.
7
6
1
.
9
3
2
.
2
0
B
4
A
B
4
A
1
.
9
4
2
.
1
1
2
.
4
2
B
5
A
B
5
A
2
.
1
2
2
.
3
3
2
.
6
5
B
6
A
B
6
A
2
.
3
4
2
.
5
5
2
.
9
2
B
7
A
B
7
A
2
.
5
6
2
.
8
1
3
.
2
0
B
8
A
B
8
A
2
.
8
2
3
.
0
9
3
.
5
2
B
9
A
B
9
A
3
.
1
0
3
.
3
9
3
.
8
7
B
1
0
A
B
1
0
A
3
.
4
0
3
.
7
4
4
.
2
5
B
1
1
A
B
1
1
A
3
.
7
5
4
.
1
1
4
.
8
8
B
1
2
A
B
1
2
A
4
.
1
2
4
.
4
7
5
.
1
5
B
1
3
A
B
1
3
A
4
.
4
8
4
.
9
7
5
.
1
5
B
1
4
A
B
1
4
A

T
h
i
s
r
e
l
a
y
h
a
s
a
t
i
m
e
/
c
u
r
r
e
n
t
c
u
r
v
e
s
u
i
t
a
b
l
e
f
o
r
u
s
e
w
i
t
h
m
o
s
t
g
e
n
e
r
a
l
p
u
r
p
o
s
e
A
C
m
o
t
o
r
s
,
5
0
/
6
0
H
z
.

T
h
i
s
r
e
l
a
y
h
a
s
a
f
a
s
t
t
r
i
p
p
i
n
g
t
i
m
e
s
u
i
t
a
b
l
e
f
o
r
a
p
p
l
i
c
a
t
i
o
n
s
w
h
e
r
e
t
h
e
m
o
t
o
r
h
e
a
t
s
r
a
p
i
d
l
y
u
n
d
e
r
s
t
a
l
l
e
d
c
o
n
d
i
t
i
o
n
s
.
Protection of industrial systems 229
300
200
400
500
600
700
Average time/current curves
based on 40°C ambient 50 Hz
60 Hz
60 Hz
300
200
600
700
500
400
7
0
0
2
0
3
0
5
0
7
0
1
0
0
2
0
0
3
0
0
5
0
0
5
3
0
2
0
1
0 7 8
3
0
0
2
0
0
1
0
0
7
0
5
0
6
0
0
5
0
0
Time (s)
1
0
1
0
0
0
Time (s)
4
8
0
0
100
C
u
r
r
e
n
t

(
A
)
800
C
u
r
r
e
n
t

(
A
)
100
800
(a)
(b)
Figure 10.1 Range of settings and operating characteristics of GE thermal relays
TMC. (a) Typical time/current characteristic curve for type TMC 23B
relays. (b) Average time/current curves based on 40 degrees ambient for
relays type TMC21B, 24B and 24D. Source: Protection of electricity
distribution systems. Reproduced by permission of General Electric
Company
230 Protection of electricity distribution networks
Nominal current
The nominal value of the breaker is selected using the next higher available setting
to the value obtained from the following expression:
I
breaker
=1.05×SF×I
nom.motor
where SF is the so-called service factor, which is an overload margin permitted by
the manufacturer.
In the case of breakers associated with motor control centres (MCCs), the nominal
current is selected using the following expression:
I
nom
=1.2×I
FL
where I
FL
is the full-load current taking into account the nominal power of all the
motors plus the other loads that are supplied by the feeder. To calculate this, the
nominal current of the largest motor, and the current for the rest of the load multiplied
by the demand factor, are added together.
Setting of the time band
When selecting the time band there should be a co-ordination margin of 0.2 s between
one breaker and the breaker acting as back up. In those breakers associated with MCC
feeders, the time setting should be checked to confirm that the largest motor can be
started satisfactorily while the rest of the load is taking nominal power.
The current of a locked-rotor motor should be estimated in accordance with a stan-
dard code of practice such as the National Electric Code of USA(NEC) Article 430-7
using the code letter in those cases where this information is available. Where this is
not the case, then the current should be taken as being six times the full load current
in accordance with Table 430-151 of NEC Article 430-7. Starting from the value of
current with a locked rotor, the starting current for each motor should be calculated
taking into account the particular method of starting, and the starting time should be
based on manufacturer’s data for the motor. In addition, the operating characteristic
of the breaker should be checked to ensure that it provides complete cover over the
thermal-capacity characteristic of the associated conductors. Table 10.2 compares
some of the factors associated with motor starting.
Range of short-time unit
The short-time element is specified taking into account the fact that the operating
current includes the motor locked-rotor current. This current is normally of the order
of six or seven times the motor nominal current. The range of settings of this element
is usually expressed as a multiple of the starting current selected for the long time-
delay unit.
Setting of instantaneous element
The instantaneous element provides protection against short-circuits, cutting down
the tripping time of a breaker when there are severe faults on the associated circuit. The
setting of the instantaneous element is calculated using the expression I
inst
=10×I
FL
,
where I
FL
is the full load current of the associated feeder.
Protection of industrial systems 231
Table 10.2 Comparison of motor starting methods (fromIndustrial Power
Systems Handbook, by D. Beeman, 1985; reproduced by
permission of McGraw-Hill Publishing Company)
Type of starter

Motor voltage Starting torque Line current
Line voltage Full voltage
starting torque
Full voltage
starting current
Full-voltage starter 1.0 1.0 1.0
Autotransformer
80% tap 0.80 0.64 0.68
65% tap 0.65 0.42 0.46
50% 0.50 0.25 0.30
Resistor starter, single step
(adjusted for motor voltage
to be 80% of line voltage)
0.80 0.64 0.80
Reactor
50% tap 0.50 0.25 0.50
45% tap 0.45 0.20 0.45
37.5% tap 0.375 0.14 0.375
Part-winding starter (low
speed motors only)
75% tap 1.0 0.75 0.75
50% tap 1.0 0.50 0.50
line voltage = motor rated voltage

The settings given are the more common for each type.
Example 10.2
Determine the settings for the thermal relay and the 200 and 600 A breakers, which
protect the system indicated in Figure 10.2, using the information given.
Induction motor
500HP, 2400V, power factor =0.8
Service factor: 1.0
Code letter: G; thermal limit with locked rotor: 5.5s
Direct start, duration: 1.0s
Thermal relay
(See Figure 10.1.)
Coil: 3.87A
Taps: 90–110 per cent of coil rating
CT ratio:150/5
232 Protection of electricity distribution networks
500 HP
M
Contactor
Thermo-magnetic
interrupter
200 A
2.4 kV
Thermo-magnetic
interrupter
600 A
1500 kVA
13.8/2.4 kV
49
150/5
Figure 10.2 System for Example 10.2
Breakers
The setting values and characteristic curve are shown in Table 10.3 and Figure 10.3
respectively.
Solution
Nominal motor current
I
N
=
500×0.746kW

3×0.8×2.4kV
=112.16A
Locked rotor current
From NEC Table 430-7(b), a motor with code G is assessed at 6.29 kVA/HP, and so:
I
LR
=
500×6.29kVA

3×2.4kV
=756.57A
Protection of industrial systems 233
T
a
b
l
e
1
0
.
3
R
a
n
g
e
o
f
s
e
t
t
i
n
g
s
o
f
S
i
e
m
e
n
s
A
l
l
i
s
b
r
e
a
k
e
r
B
r
e
a
k
e
r
t
y
p
e
a
n
d
f
r
a
m
e
s
i
z
e
T
r
i
p
p
i
n
g
X
F
M
R
r
a
t
i
n
g
(
p
r
i
m
a
r
y
A
)
L
o
n
g
-
t
i
m
e
e
l
e
m
e
n
t
c
a
l
i
b
r
a
t
e
d
p
i
c
k
-
u
p
s
e
t
t
i
n
g
s
(
A
)
M
a
x
i
m
u
m
c
o
n
t
i
n
u
o
u
s
r
a
t
i
n
g
(
A
)
G
r
o
u
n
d
-
e
l
e
m
e
n
t
c
a
l
i
b
r
a
t
e
d
p
i
c
k
-
u
p
s
e
t
t
i
n
g
s
(
A
)
A
B
C
D
E
F
G
1
5
%
2
5
%
5
0
%
1
0
0
%
L
A
-
6
0
0
A
8
0
4
0
5
0
6
0
7
0
8
0
9
0
1
0
0
1
0
0
×
×
4
0
8
0
a
n
d
2
0
0
1
0
0
1
2
5
1
5
0
1
7
5
2
0
0
2
2
5
2
5
0
2
5
0
3
0
5
0
1
0
0
2
0
0
L
A
-
8
0
0
A
4
0
0
2
0
0
2
5
0
3
0
0
3
5
0
4
0
0
4
5
0
5
0
0
5
0
0
6
0
1
0
0
2
0
0
4
0
0
6
0
0
3
0
0
3
7
5
4
0
0
5
2
5
6
0
0
6
7
5
7
5
0
6
0
0

9
0
1
5
0
3
0
0
6
0
0
L
A
-
8
0
0
A
o
n
l
y
8
0
0
4
0
0
5
0
0
6
0
0
7
0
0
8
0
0
9
0
0
1
0
0
0
8
0
0
1
2
0
2
0
0
4
0
0
8
0
0
L
A
-
1
6
0
0
A
2
0
0
1
0
0
1
2
5
1
5
0
1
7
5
2
0
0
2
2
5
2
5
0
2
5
0

5
0
1
0
0
2
0
0
4
0
0
2
0
0
2
5
0
3
0
0
3
5
0
4
0
0
4
5
0
5
0
0
5
0
0
6
0
1
0
0
2
0
0
4
0
0
8
0
0
4
0
0
5
0
0
6
0
0
7
0
0
8
0
0
9
0
0
1
0
0
0
1
0
0
0
1
2
0
2
0
0
4
0
0
8
0
0
1
6
0
0
8
0
0
1
0
0
0
1
2
0
0
1
4
0
0
1
6
0
0
1
8
0
0
2
0
0
0
1
6
0
0
2
4
0
4
0
0
8
0
0
1
6
0
0
L
A
-
3
2
0
0
A
2
0
0
0
1
0
0
0
1
2
5
0
1
5
0
0
1
7
5
0
2
0
0
0
2
5
0
2
5
0
0
2
5
0
0
3
0
0
5
0
0
1
0
0
0
2
0
0
0
3
2
0
0
1
6
0
0
2
0
0
0
2
4
0
0
2
8
0
0
3
2
0
0
3
6
0
0
4
0
0
0
3
2
0
0
#
4
8
0
8
0
0
1
6
0
0
3
2
0
0
L
A
-
4
0
0
0
A
4
0
0
0
2
0
0
0
2
5
0
0
3
0
0
0
3
5
0
0
4
0
0
0
4
5
0
0
5
0
0
0
4
0
0
0
6
0
0
1
0
0
0
2
0
0
0
4
0
0
0
*
7
5
0
o
n
L
A
-
8
0
0
A
#
3
0
0
0
f
o
r
L
A
F
-
3
0
0
0
A
×
m
a
y
n
o
t
t
r
i
p
234 Protection of electricity distribution networks
30
Earth fault current in per cent
of tripping XFMR rating
10 20
Current in multiples of pick-up
100 50 70 7 2 1 3 5
Maximum
100 30 20 10 50 70
Interrupting time
Instantaneous
curve (I)
Maximum
Intermediate
Minimum
time band
T
i
m
e

(
s
)
T
i
m
e

b
a
n
d

4
Intermediate
Maximum
Minimum
time band
Time band
Time band
T
i
m
e

b
a
n
d
2
Earth fault curves (G)
T
i
m
e

b
a
n
d

1
T
i
m
e

b
a
n
d

3
Short time
curves (s)
T
i
m
e

b
a
n
d

6
T
i
m
e

b
a
n
d
5
0.01
0.02
0.06
0.04
0.08
0.1
0.2
0.6
0.4
0.8
1
2
4
10
6
8
20
400
40
80
60
100
200
800
600
1000
Long time curves (T)
Long time pick-up
is continuously adjustable
from 0.5 to 1.25 times tripping
XFMR rating, with calibrated
points per table
Time delay
Reference points
per device
faceplate
Earth fault
pick-up
adjustment
range
Instantaneous
pick-up
adjustment range
Short time pick-up
adjustment
range
Figure 10.3 Characteristic operating curves for Siemens-Allis Interrupter
Setting of thermal relay
I
start
=1.05×1.0×112.16A=117.8A
Setting=117.8×
5
150
×
1
3.87
×100%=101.46%
With a setting of 100 per cent,
I
start
=1.0×3.87A×(150/5) =116.10A (3.5% overload)
Protection of industrial systems 235
Setting of 200 A breaker
(i) Long-time element
Range: 100/125/150/175/200/250 A
I
start
=1.05×1.0×112.16A=117.8A
Set at 125 A (11.45 per cent, overload)
Selection of time band:
The long-time element should permit the motor to start:
Start point: I =756.57A locked rotor current, and t =1.0s
Operation of breaker at the start:
I/I
start
=756.7A/125A=6.05 times.
Require 1.4 s tripping time. From Figure 10.3, the intersection of 1.4 s and 6.05 times
lies above the lower curve of Band 1. Therefore, Band 2 is chosen in this case to
guarantee the required discrimination margin of 0.4 s.
(ii) Short-time element
Range: 3/5/8/12 times long-time pick-up current
I
start1
=6×(112.16A) =672.96A
Given that this value is less than the motor starting current, it is necessary to increase
the setting value. Try eight times, then
I
start2
=8×(112.16A) =897.28A and setting=897.28A/125A=7.18
Setting selected: 8×I
pickup
=8×125A=1000A
Time band: in this case the intermediate band was chosen to provide the necessary
discrimination margin with the instantaneous unit (see Figure 10.3).
(iii) Instantaneous element
Range: 3/5/8/12 times long-time pick-up current
I
start
=12×(112.16A) =1345.92A
Setting=1345.92A/125A=10.77
Setting selected: 11×I
pickup
(1375 A)
Setting of 600 A breaker
(i) Long-time element
Range: 300/375/400/525/600/750 A
Nominal current of transformer, I
n
=
1500kVA

3×2.4kV
=360.84A
With a setting of 400 A, the overload=400/360.84=1.108, i.e. 10.8 per cent, which
is acceptable.
I
start
=1.10×(360.84A) =396.92A
Selection of time band:
Operation of motor breaker within the limit of the long-time element:
I/I
start
=1000A/125A=8.0 times. At eight times, and with Band 2, ⇒t =1.5s
236 Protection of electricity distribution networks
Operation of transformer breaker:
I/I
start
=1000A/400A=2.5 times. At 2.5 times, and with t =1.7s, Band 1 is chosen
since this is the lowest available. Notice that a margin of 0.2 s has been applied here
since the co-ordination is between two lowvoltage breakers where the curves include
the opening time.
(ii) Short-time element
I
start
=6×(360.84A) =2165.04A
Setting=2165.04A/400A=5.41⇒6 times
B C A
A: Inom. motor Motor 600HP, 2.4 kV 112A
B: Ilocked rotor Motor 500HP, 2.4 kV 756 A
C: Inom.trfr 1500kVA Transformer, 2.4kV Side
1
T
i
m
e

(
s
)
Motor 500 HP, 2.4kV
Starting characteristic
4
Motor Siemens LA-200A
Interrupter long time element
Setting = 125 A
Band = 2
Short time element
Setting = 8 × I
pickup
Band = Intermediate
Instantaneous
element = 11 × I
pickup
Siemens l A-600 A interrupter
Long time element
Setting = 400 A
Band = 1
Short time element
Setting = 6 × I
pickup
Band = Intermediate
Instantaneous element cancelled
3
2
Motor 600 HP, 2.4 kV
TMC thermal relay
CTR = 150/5
Tap = 100%
Current (A) (×10) – 2.4 kV busbar
1
3 2
0
.
6
0
.
8 1
0
.
5
5
0
3
0
1
0
2
0 7 5
5
0
0
1
0
0
3
0
0
2
0
0
7
0
7
0
0
7
0
0
0
5
0
0
0
3
0
0
0
2
0
0
0
1
0
0
0
1
0
0
0
0
0.01
0.02
0.06
0.04
0.08
0.1
0.2
0.6
0.4
0.8
2
4
10
6
8
20
400
40
80
60
100
200
800
600
1000
2
4
3
1
Figure 10.4 Co-ordination curves for Example 10.2
Protection of industrial systems 237
Since there is no overlap with the short-time element of the motor, the intermediate
band is also selected.
(iii) Instantaneous element
The instantaneous element is cancelled in order to maintain co-ordination. The
co-ordination curves are shown in Figure 10.4.
It should be emphasised that, although the values selected for the protective
devices in this example ensure proper co-ordination, they are not unique. Other
settings could be chosen provided that the curves so obtained guarantee adequate
reliability and good selectivity.
Chapter 11
Industrial plant load shedding
All electrical power systems that contain generation are liable to be subjected to
a variety of abnormal operating conditions such as network faults, loss of some or
all generation, the tripping of circuits within the system and other disturbances that
can result in the reduction in the generation capacity available to the system. In these
situations a balance between the existing load and the remaining generation should be
re-established if possible before the reduction in frequency produced by the overload
affects the turbines and the generation auxiliary equipment, which could eventually
result in the total collapse of the system.
To assist in restoring the equilibrium, frequency relays are employed. These
disconnect the less important loads in stages when the frequency drops to a level that
indicates that a loss of generation or an overload has occurred. This type of scheme
is very useful in industrial plants where in-house generation is synchronised to the
public grid system. The concepts, criteria and conditions applicable to the design of
an automatic load shedding system for industrial plants, based on frequency relays,
are set out in this chapter.
11.1 Power system operation after loss of generation
When a total or partial loss of generation occurs within the system, the first indicators
are a drop in voltage and in frequency. However, given that voltage drops can also be
caused by system faults, it is generally recognised that a drop in frequency is a more
reliable indication of loss of generation. A sudden loss of generation in the system
will result in a reduction in the frequency at a rate of change that depends on the size
of the resultant overload and the inertia constant of the system.
The relationship that defines the variation of frequency with time, following
a sudden variation in load and/or generation, can be obtained, starting from the
equation for the oscillation of a simple generator:
GH
πf
0
×
d
2
δ
dt
2
=P
A
(11.1)
240 Protection of electricity distribution networks
where: G=nominal MVA of machine under consideration; H =inertia constant
(MWs/MVA=MJ/MVA); δ =generator torque angle; f
0
=nominal frequency; P
A
=
net power accelerated or decelerated (MW).
The speed of the machine at any instant (W) can be given by the following
expression:
W =W
0
+

dt
=2πf (11.2)
in which W
0
is the synchronous speed, i.e. the nominal speed at rated frequency.
Differentiating eqn. 11.2 with respect to time:
dW
dt
=
d
2
δ
dt
2
=2π
df
dt
(11.3)
Replacing eqn. 11.3 in eqn. 11.1 gives
df
dt
=
P
A
f
0
2GH
(11.4)
Eqn. 11.4 defines the rate of the variation of the frequency in Hz/s, and can be
used for an individual machine or for an equivalent that represents the total generation
in a system. In such a case the inertia constant can be calculated from
H =
H
1
MVA
1
+H
2
MVA
2
+· · · +H
n
MVA
n
MVA
1
+MVA
2
+· · · +MVA
n
(11.5)
where the subscripts 1, 2. . . n refer to the individual generator units. It should be
emphasised that the constant H in eqn. 11.5 is expressed to an MVA base equal to
the total generation capacity of the system.
The accelerating power P
A
in eqn. 11.4 is responsible for the frequency variation.
It can be calculated from
P
A
=P
M
−P
E
(11.6)
where P
M
=the mechanical power entering the generator and P
E
=the electrical
power leaving the generator.
Under stable conditions P
A
=0 and there are no frequency variations. In the case
of overloads P
E
>P
M
. Thus, P
A
<0 and there will be a drop in the systemfrequency.
11.2 Design of an automatic load shedding system
In order to design an automatic load shedding system, a model that represents the
different generating machines should first be defined, and then the load parameters
and the criteria for setting the frequency relays.
11.2.1 Simple machine model
Within the scope of this book, a single machine has been used in the power system
model used to illustrate a load shedding system. This is equivalent to assuming that the
Industrial plant load shedding 241
generator units are electrically connected with negligible oscillations between them,
and with a uniform frequency across the whole of the system, ignoring the effect of
the regulating equipment. The load is represented as a constant power, which implies
that there is no reduction in load as a result of the voltage and frequency drops after
a contingency situation. This model provides a pessimistic simulation of the system
since the reduction of the load due to the frequency drops and the effect of the speed
regulators are neglected. In using this model the inertia constant of the system is
calculated using eqn. 11.5.
The rate of change of the frequency is calculated fromeqn. 11.4 with the following
assumptions:
• the mechanical power entering the generators does not vary and is equal in
magnitude to the prefault value;
• the magnitude of the load does not vary with time, voltage or frequency. It is
only reduced by disconnecting part of the load as a result of the automatic load
shedding system.
This simple machine model, with loads modelled as a constant power, is used to
determine the frequency relay settings and to verify the level of minimum frequency
attained before a contingency situation is reached, for the following reasons:
• the ease of using an iterative process to design the load shedding system;
• consideration of suitable security margins, since the fact that the load diminishes
with loss of voltage is neglected. This implies introducing much more severe rates
of frequency variation, thus achieving more rapid settings.
11.2.2 Parameters for implementing a load shedding system
The following aspects need to be defined in order to implement the load shedding
system.
Maximum load to be disconnected
Generally, one of the most drastic conditions corresponds to the total loss of inter-
connection between the public electricity supply network and the internal electrical
system within the industrial plant. In this case, the unbalance between generation
and load will be equal to the maximum import and should be compensated by the
disconnection of a similar amount of load from the plant system.
Starting frequency of the load shedding system
The disconnection system should be set so that it will initiate operation at a value of
frequency below the normal working system frequency. Taking into account varia-
tions in frequency caused by oscillations inherent in the public system, this value is
normally selected at approximately 93 per cent of nominal system frequency. How-
ever, if it is thought that there is a possibility of more severe oscillations occurring
on the system, then it is recommended that a supervisory control arrangement using
242 Protection of electricity distribution networks
Table 11.1 Typical times for the operation
of turbines (full load)
% of rated frequency
at full load
Maximum permissible
time (minutes)
99.0 continuously
97.3 90
97.0 10
96.0 1
Note: The operating times at low frequency and full load
are cumulative.
overcurrent relays, which can detect the outages of circuits connecting the industrial
plant to the public system, should be installed to avoid incorrect operations.
Minimum permissible frequency
A steam turbine is designed so that, when operating at nominal mechanical speed
and generating at nominal system frequency, excessive vibrations and stresses in its
components, e.g. resonance of turbine blades, are avoided. However, when running
below normal speed at a reduced system frequency, cumulative damage could be
produced by excessive vibration. It is recommended, therefore, that the time limits
given in Table 11.1 should not be exceeded. However, during transient operation and
with load below nominal, in the majority of cases reduction of frequency down to
93 per cent of rated frequency can be permitted without causing damage either to the
turbine or to the turbogenerator auxiliary lubrication and cooling systems.
11.3 Criteria for setting frequency relays
The determination of the frequency relay settings is an iterative process, and is carried
out in such a way that the final settings satisfy the requirements of both speed and
co-ordination. In this process, the co-ordination between relays that trip successive
stages of load should be checked in order to ensure that the least amount of load is
shed, depending on the initial overload condition.
11.3.1 Operating times
When selecting the settings, it is necessary to consider the time interval between the
system frequency decaying from the relay pick-up value to the point in time when
the load is effectively disconnected. The relay pick-up time is included in this time
interval, plus the preset time delay of the relay, if this is required, and the breaker
opening time.
The following values are typically used for industrial systems: relay pick-up time:
50 ms and breaker opening time: 100 ms.
Industrial plant load shedding 243
11.3.2 Determination of the frequency variation
The frequency variation required to calculate the settings is obtained by using a simple
machine model for the system, and a constant power model for the load. This assumes
that the load connected to the generators is the same before, and after, the contingency,
neglecting any form of damping. Given this, the calculated rate of loss of frequency
in the system is pessimistic, and the settings determined on this basis thus provide an
arrangement that rapidly restores the frequency to its normal value, thereby ensuring
a secure system.
11.4 Example of calculating and setting frequency relays in an
industrial plant
The procedure for calculating the settings of the frequency relays at a typical industrial
plant is given next. The single-line diagram is shown in Figure 11.1, together with
the principal data for the study.
11.4.1 Calculation of overload
The initial load conditions are summarised as: total load: 24.0 MW; in-house
generation: 8.0 MW; total import: 16.0 MW; GH constant: 35.43.
In the eventuality of the loss of the incoming grid supply, the in-house generators
will experience the following overload:
%overload=
24.0−8.0
8.0
×100%=200%
11.4.2 Load to be shed
With the loss of the grid supply, 16 MW of capacity is lost, which will have to be
borne initially by the in-house generators. Aload equal to, or greater than, this amount
must be disconnected in order to relieve the overload. Therefore, the load to be shed
is 24−8=16MW. It should be noted that there are high priority loads totalling
8.02 MW that cannot be disconnected.
11.4.3 Frequency levels
Disconnection of load is initiated each time the systemfrequency falls to 59 Hz, which
indicates that a loss of generation has taken place. A minimum frequency of 56 Hz is
acceptable.
11.4.4 Load shedding stages
Assume that three stages of shedding have been set up that will off load the incoming
circuit by 15.98 MW, as detailed in Table 11.2. With this arrangement, the in-house
generation can then supply the 8.02 MW of high priority load.
244 Protection of electricity distribution networks
13.2kV busbar
4.16kV busbar
Priority
loads
Load 2 G–1 G–2 Load 3 Load 2 Load 2
Priority
loads
Priority
loads
Load 1
115kV busbar
Distribution
network
Priority
loads
Load 3
Figure 11.1 Systemarrangement for example of calculating frequency relay settings
11.4.5 Determination of the frequency relay settings
The settings of the frequency relays shown in Table 11.2 are determined in such
a way that each stage is disconnected only when the system frequency falls to
a predetermined value. This value is obtained by calculating the reduction in the
Industrial plant load shedding 245
Table 11.2 Load shedding stages
Priority Description MW
1 load 1 6.03
2 load 2 4.73
3 load 3 5.22
Total load to be shed=15.98MW
system frequency due to an overload equal to the stage considered, as described
below.
First-stage setting
The first stage is disconnected when the frequency reaches 59 Hz.
Second-stage setting
An overload equal to the first stage is considered and the subsequent rate of frequency
drop is determined from
df
dt
=

−6.03
2GH

×60=−5.106Hz/s
The frequency as a function of time is given in Figure 11.2 and f =(60−5.106t ).
The opening time for the first stage is
t
trip
=t
pick-up
+t
breaker
+t
relay
t
pick-up
=(60−59)/5.106=0.196s
t
trip
=0.196+0.100+0.05=0.346s
The frequency drop up to the operation of the first stage is f =[60−5.106(0.346)] =
58.233Hz. The second stage is set below this value, i.e. at 58.15 Hz.
Third-stage setting
These are set so that they will not operate for overloads below (6.03+4.73) =
10.76MW, which are disconnected by stages 1 and 2.
df
dt
=

−10.76
2GH

×60=−9.111Hz/s
f =60−(9.111)t
The pick-up time for the first stage is t =(60−59)/9.111=0.110s, and the tripping
time for this stage is t
trip1
=0.110+0.05+0.100=0.260s. Thus, the frequency drop,
f =60−9.111(0.26) =57.631Hz.
246 Protection of electricity distribution networks
Notes:
D1: 1st stage of load disconnection
P1: Pick-up of 1st stage relay
0.1 0.2 0.3 0.4 0.5
Time (s)
4.73 MW overload
9.97 MW overload
D2 (0.35,57.27)
5.24 MW overload
D2 (0.29,56.5)
D3 (0.36,56.1)
P3
D1 (0.22,56.9)
P3
D1 (0.26,57.63)
P2
16 MW
overload
P2
D1 (0.34,58.23)
10.76 MW overload
P1
0.074
P1
6.03 MW overload
f
(Hz)
59
58.1
57.20
57
56
58
60
Figure 11.2 Calculation of settings of frequency relays
Industrial plant load shedding 247
The value that produces pick-up at the second stage is t =(60−58.15)/9.111=
0.203s, so that t
trip2
=0.203+0.05+0.100=0.353s.
The frequency equation shows that in t
trip1
=0.260s there is a variation of slope
as a consequence of the disconnection of the first stage (6.03 MW); see Figure 11.2.
After this, the accelerating power is only 4.73 MW.
Therefore
df
dt
=
−4.73
2GH
×57.631=3.847Hz/s
and, from the frequency equation:
f −57.632=−3.847(t −0.26), t >0.26s
f =58.632−3.847t
The frequency drop up to the stage 2 tripping is given by f =58.632−3.847t
trip2
,
from which f =57.274Hz. Therefore a third stage setting of 57.20 Hz is selected.
The final settings are given in Table 11.3.
The operation of the system in the presence of a total loss of connection with the
distribution network is represented by the lower characteristic in Figure 11.2.
11.4.6 Verification of operation
In order to verify the operation of the proposed system, the reduction in frequency
during the process of load shedding is studied by two different methods: modelling
the load as a constant power, and modelling the load with damping as a result of the
voltage drop. For the latter case a transient stability program is used.
Modelling load as a constant power
For this case it is assumed that the magnitude of the load is constant and therefore
does not depend on the voltage level. This consideration is pessimistic with regard
to the actual situation and therefore provides some margin of security. The frequency
analysis is carried out using eqn. 11.4, the settings of the frequency relays given in
Table 11.3 and assuming a maximum circuit breaker opening time of 100 ms.
Figure 11.3 shows the behaviour of the frequency fromt =0s, when loss of supply
occurs at a total systemload of 24 MW. Under such conditions the sequence of events
Table 11.3 Frequency relay settings
Stage Frequency setting
(Hz)
Delay time
(s)
1st 59.00 instantaneous
2nd 58.15 instantaneous
3rd 57.20 instantaneous
248 Protection of electricity distribution networks
Notes:
P1: Pick-up of 1st stage relay
D1: 1st stage of load disconnection
0.1 0.2 0.3 0.4 0.6 0.5 0.7
Time (s)
D1 (0.27,57.8)
D2 (0.38,57.3)
D2 (0.29,56.5)
D3 (0.36,56.1)
Load modelled as
a constant power
D1 (0.22,56.9)
P3
D3 (0.55,56.5)
P2
Load modelled as
a constant impedance (P=kV
2
)
P1
f
(Hz)
56
57.20
57
58
58.15
59
60
Figure 11.3 Variation of frequency during load shedding
is as given in Table 11.4. The drastic initial overload is eventually eliminated by the
load shedding system in 0.357 s. During this time, the frequency will fall to 56.17 Hz,
a value which more than meets the standards accepted for industrial systems operating
separately. After t =0.357s, the system frequency then starts to recover.
Industrial plant load shedding 249
Table 11.4 Disconnection of the grid supply; sequence of events
Time Frequency
(Hz)
Event Rate of
change of
frequency
(Hz/s)
Rate of
change of
load shed
(MW)
Overload
remaining on
generators
(MW)
0.000 60.00 Disconnection of grid supply −13.46 – 16.00
0.074 59.00 Pick-up of 1st stage relay −13.45 – 16.00
0.137 58.15 Pick-up of 2nd stage relay −13.45 – 16.00
0.207 57.20 Pick-up of 3rd stage relay −13.45 – 16.00
0.224 56.90 1st stage of load disconnection −8.00 6.03 9.97
0.287 56.50 2nd stage of load disconnection −4.20 4.73 5.24
0.357 56.10 3rd stage of load disconnection 0.00 5.22 0.02
To obtain the values in Table 11.4, it must be borne in mind that, since the initial
overload is equal to 16 MW, then
df
dt
=
−16
2GH
×60=−13.45Hz/s
Therefore, the pick-up time for the first stage is
t
p
=
60−59
13.45
=0.074s
The opening time of the breakers associated with the first stage is t
d
=0.074+
0.100+0.05=0.224s and the frequency at that moment is f =60−13.45t
d
=
56.9Hz. The values for the other stages can be calculated in a similar way, and
are given in Table 11.4.
Modelling load with damping as a result of the voltage drop (P =kV
2
)
The stability program should be used to set the initial condition of the load defined
for this study, as this can simulate the disconnection of the infeed at t =0s, and
subsequently the disconnection of the loads at the time at which the frequency relays
are set to operate.
Figure 11.3 also illustrates the frequency characteristic obtained by computer
modelling the loads by constant admittances. This method is less drastic than the
constant power model since, in this case, the power of the load is damped by the
voltage drop. As the overload is reduced, the frequency drop is less than that obtained
using the first model. As seen in Figure 11.3, the frequency relay settings, and there-
fore the tripping of the breakers, is much slower although the minimum value of the
frequency eventually reaches 56.55 Hz, which is above the predetermined minimum
limit of 56 Hz. From Figure 11.3 and the results of the stability program it can be
seen that the frequency after the loss of the third stage gives recovery at a rate of
0.67 Hz/s. This implies that it should take approximately 2.345 s for recovery to the
normal level of 60 Hz.
250 Protection of electricity distribution networks
The modelling of the loads by these two methods makes it possible to obtain
graphs of the frequency variation as a function of time that correspond to the most
adverse and most favourable extremes in the system. In this way, the curves obtained
for the two models delineate the operating area of the system under study, before the
loss of connection with the grid supply network.
Analysis of voltage with total loss of infeed supply and operation of the load
shedding system
The loss of generation in the system not only causes loss of frequency but also a
drop in voltage. The automatic load shedding scheme should prevent systemvoltages
falling to such a level as to cause tripping of the contactors on the motors serving the
plant.
A check should be made on the voltage levels on the system as follows:
1. Determine the initial voltage at each busbar, using a load flow program.
2. Determine the variations in voltage at each busbar after the loss of connection
with the grid supply, and while the load shedding scheme is in operation. For
this a transient stability program is used to obtain the voltages at each busbar for
each stage of the analysis.
3. Produce the curves of voltage versus time for the busbars feeding the priority
loads.
Chapter 12
Protection schemes and substation design
diagrams
Previous chapters have detailed the make up and operating characteristics of various
types of protection relays. This chapter considers the combination of relays required
to protect various items of power system equipment, plus a brief reference to the
diagrams that are part of substation design work. A general knowledge of these
diagrams is important in understanding the background to relay applications.
12.1 Protection schemes
It is difficult to define precisely the protection schemes that should be adopted for an
electricity distribution system, given the large number of valid alternatives for each
situation, but some schemes will be presented as a guide for protecting the various
elements that make up a power system. However, any protection scheme should
strike a balance between the technical and economic aspects so that, for example,
sophisticated protection devices are not used for small machines or for less important
power system elements.
12.1.1 Generator protection
Generator protection should take into account the importance of the generator and its
technical characteristics such as power, voltage and earthing arrangement, plus any
economic considerations. Acomplex protection scheme can ensure that the generator
is protected against whatever faults may occur. However, it is unlikely that such a
cost could be justified for every generating station, especially those with small units.
It is, therefore, necessary to define a protection scheme that is adequate for the size
of the machine.
Two generator protection schemes are given below, based on suggestions by
manufacturing companies.
252 Protection of electricity distribution networks
Small generators
For small generators, typically up to 5 MVA, it is considered necessary to have:
• protection against internal faults;
• back-up protection for external faults using overcurrent relays with voltage
restraint;
• reverse-power protection;
• earth-fault protection, using an overcurrent relay;
• protection against overloads by means of thermal relays.
This scheme is illustrated in Figure 12.1 (for relay identification, see section 3.1.4).
Large generators
For large generators, say over 5 MVA, the protection, which is shown in Figure 12.2,
should normally comprise:
• differential protection to cover internal faults;
• earth-fault protection using high impedance relays;
• back-up protection by means of distance or overcurrent protection with voltage
restraint;
• reverse-power protection;
• negative-phase sequence protection;
• protection against loss of excitation;
• protection against overload using thermal relays.
• out of step
• inadvertent energisation (50/27)
• stator earth protection (59N and 27N)
• over/under frequency
• over/under voltage
12.1.2 Motor protection
The amount of protection, and its type, used for a motor is a compromise between
factors such as the importance of the motor, the potential dangers, the type of duty,
and the requirements of protection co-ordination against the cost of the protection
scheme. The schemes that are illustrated represent common practice and international
recommendations for the protection of motors of different powers, and are divided
into four categories:
• protection of low power motors (less than 100 HP);
• protection of motors up to 1000 HP;
• protection of motors greater than 1000 HP;
• additional protection for synchronous motors.
In the diagrams the starting equipment of the respective motors has not been
represented.
Protection schemes and substation design diagrams 253
R
Three-phase
connection point
51G
Field
87 86
41
G
49
52
32 51V
Busbar
46 40
Figure 12.1 Protection schematic for small generators
Protection of low power motors
Low power motors are normally protected by fuses associated with thermal overload
relays incorporating bimetallic elements (Figure 12.3) – the fuses protecting against
short circuits – or low voltage breakers plus thermal overload relays (Figure 12.4)
254 Protection of electricity distribution networks
52
BF
87 86
49
46 21 32 40 51V 78 24 27 59 60 81 50/27
G
64
41
Field
Three phase
connection point
R 59N 27TN
87B
Other
busbar CTs
Busbar
Figure 12.2 Protection schematic for large generators
when the breaker should have a magnetic element to trip instantaneously under short-
circuit conditions.
Protection of motors up to 1000 HP
The protection arrangements should include thermal protection against overloads
and short-circuits (49/50), protection for locked rotor (51) and earth-fault protection
(50G), as indicated in Figure 12.5.
Protection schemes and substation design diagrams 255
Contactor
Thermal
overload
relay
G
Fuse
Busbar
Figure 12.3 Schematic of fuse protection for low power motors
G
Thermo-magnetic
interrupter
Contactor
Busbar
Thermal
overload
relay
Figure 12.4 Schematic of low voltage breaker protection for low power motor
256 Protection of electricity distribution networks
M
Toroidal
type CT
52
Busbar
49 46
50G
50
51
Figure 12.5 Protection schematic for motors up to 1000 HP
Protection of motors greater than 1000 HP
The scheme shown in Figure 12.6 includes unbalance protection (46), thermal protec-
tion against overloads (49), protection for a locked-rotor situation (51), differential
protection for internal faults (87), back up for short-circuits (50), and earth-fault
protection (50G).
Additional protection for synchronous motors over 1000 HP
In addition to the protective devices indicated in Figures 12.4 and 12.5, a large syn-
chronous motor requires protection for the field winding, plus a low power factor
relay (55) and undervoltage protection (27), and a high/low frequency relay (81)
Protection schemes and substation design diagrams 257
87
Three-phase
connection point
M
Toroidal
type CT
50
86 52
Busbar
49 46
50
51
Figure 12.6 Protection schematic for motors over 1000 HP
to prevent the motors running under conditions of low frequency operation. The
schematic diagram for the protection of synchronous motors over 1000 HP is given
in Figure 12.7.
Protection for the field winding
This would require an earth-fault protection relay (64), and a field relay (40) to deal
with loss of excitation current.
258 Protection of electricity distribution networks
87
64
M
Three-phase
connection point
41
40
Toroidal
type CT
50
41
86
46 49
52
51
50
55 81 27
Busbar
Figure 12.7 Protection schematic for synchronous motors over 1000 HP
12.1.3 Transformer protection
Transformer protection should take into account the power, voltage, vector group,
and the importance of the unit within a particular system. Depending on these factors,
the transformers can be assigned to one of the two groups, as below.
Protection schemes and substation design diagrams 259
MV busbar
LV busbar
52
49
26
63
52
86 87
51
50
51
50
51
50
51N
50N
N
Figure 12.8 Protection for MV/LV transformer
MV/LV transformers
The protection of these units should include overcurrent protection for both the MV
and LVwindings, plus devices such as over-pressure protection (e.g. Buchholz surge),
and thermal protection, as indicated in Figure 12.8. Typical ratios for inter-busbar
MV/LV transformers at substations are 33/11, 34.5/13.2, and 13.2/4.16 kV.
260 Protection of electricity distribution networks
HV busbar
Other
busbar CTs
MV busbar
52
LV busbar
52
26
49
87B
52
63 86 87T
BF 24
51
50
51
50
51
50
51
50
N
51
50
N
Figure 12.9 Protection schematic for HV/MV/LV transformer
HV/MV/LV transformers and autotransformers
In addition to the protection listed for the MV/LV transformers, the protection for
transformers in this group should include overall differential protection, which is
essential because of its reliability and high speed of operation. In this case, shown
in Figure 12.9, since the transformer has three windings a three terminal type of
differential protection is required. The diagram also includes the differential busbar
Protection schemes and substation design diagrams 261
protection, which is usually installed on large transformers connected to HV busbars.
As the transformer has an LV winding, overcurrent protection for this winding has
been included as well. Typical ratios for HV/MV transformers are 132/33, 145/11,
132/11, 115/34.5 and 115/13.2 kV.
It is common to use autotransformers where large powers are involved and the
voltage ratio is around 2: 1. Typical ratios for autotransformers are 275/132 and
230/115 kV. The protection schemes for autotransformers are very similar to those for
HV/MV/LVtransformers since autotransformers can be treated as three winding units
for protection purposes. The protection for this type of transformer is essentially the
same as that quoted in the previous paragraph, taking into account the modifications
to the overcurrent relay connections (see Figure 12.9).
12.1.4 Line protection
Line protection generally consists of overcurrent, distance, and directional overcur-
rent relays and, depending on the voltage level at which the line is operating, schemes
indicated below are in general use.
Medium voltage lines
These circuits should be protected with overcurrent relays, and directional overcur-
rent relays should be used on an MV ring, while MV radial circuits should include
reclose relays within the protection scheme. The schematic diagrams for these two
cases are shown in Figure 12.10. It is common practice on MV feeders for the trip-
ping of instantaneous relays to be routed via the reclosing relays. This ensures that
the reclosing relay is energised from a protection relay trip to start a reclose opera-
tion. Time-delay units are often used to produce a definite time trip, i.e. without the
possibility of a reclose operation occurring.
High voltage lines
HV lines would normally have distance and directional overcurrent protection, plus
carrier wave receive and reclosing relays. Duplication of the main protection can
sometimes be justified on important lines, or the use of some other type such as
differential protection. When radial circuits are involved, directional overcurrent
relays can be replaced by nondirectional overcurrent relays. The schematic diagram
for a typical protection arrangement for HV lines is illustrated in Figure 12.11.
12.2 Substation design diagrams
Aprotection engineer should have a good understanding, not only of the performance
of relays and the criteria for setting these as set out in the previous chapters, but also
of the relationship between protection and the other equipment in the substations
and the distribution system. This section, although not intended to cover substation
design, includes some basic information on substation equipment layout, and on
other diagrams that a protection engineer should be able to handle without difficulty,
262 Protection of electricity distribution networks
52
To radial
circuit
51
50
50N
79
51N
94
52 94
51
50
51
50
67 67N
Busbar (a) (b) Busbar
N
Figure 12.10 Protection arrangements for feeders and MV ring lines: (a) feeder
protection; (b) MV ring protection
in order to ensure a better appreciation of protection schemes and relay settings, and
operational procedures.
Apart from the pure electrical aspects, the design of a substation incorporates
several engineering fields, among them civil, mechanical and electronic. Within
the electrical design function, the basic diagrams used are the single-line diagram,
substation equipment layout drawings, diagrams of ACand DCconnections, and sec-
ondary wiring plus logic diagrams. A brief mention of these is given in the following
paragraphs.
12.2.1 Single-line diagrams
Asingle-line diagramshows the disposition of equipment in a substation, or network,
in a simplified manner, using internationally accepted symbols to represent various
Protection schemes and substation design diagrams 263
52
67 67N
Z2
21
87B
86
Other
busbar CTs
79
85
Busbar
25
BF
Z1
21
Z3
21
Figure 12.11 Typical protection arrangement for HV lines
items of equipment such as transformers, circuit breakers, disconnectors etc., gen-
erally with a single line being used to represent three-phase connections. Often the
main data for the HVequipment is included on the diagram. More detailed single-line
diagrams include such items as the instrument transformers, and the protection,
measurement and control equipment and their associated secondary wiring.
12.2.2 Substation layout diagrams
Substation layout diagrams provide scale drawings of the location of each piece
of equipment in a substation, both in plan and elevation. While individual utilities
may have their own format, there is a high degree of standardisation of these types
264 Protection of electricity distribution networks
1 2 6 7
1
3
A A'
1 Surge diverter
2 VTs and line trap
3 Columns
4 CTs
5 Circuit breakers
6 Disconnect switches
7 Power transformer
Equipment identification
1 2 6 4 5 3 6 6 3 4 5 7
3
3
1
3
52
52
a
(b)
(c) (a)
6 4 5 3 4 5 3 6
Figure 12.12 General arrangement for two 115 kV bays: (a) general layout;
(b) elevation A-A

; (c) single-line diagram
of drawings worldwide, for contractual and tendering purposes. Figures 12.12a
and b show the plan and elevation drawings for a typical layout of two 115 kV bays,
one for a transmission line and the other for the HV side of a local transformer, con-
nected to a single 115 kV busbar. The equivalent single-line diagram is depicted in
Figure 12.12c at the top right-hand corner of the drawing.
Although protection engineers may not be directly involved with layout diagrams,
these drawings do showthe relationship between various items of primary equipment
and the location of those items associated with protection systems – for example,
current and voltage transformers that may be located separately from other items
of equipment or placed within high voltage equipment such as circuit breakers. The
protectionengineer is thus able toensure that he cansafelylocate protectionequipment
within the substation.
12.2.3 Diagrams of AC connections
A diagram of AC connections generally shows the three-phase arrangement of the
substation power equipment, and the AC circuits associated with the measurement,
control and protection equipment, in schematic form. The AC diagrams for a typical
Protection schemes and substation design diagrams 265
substation contain information corresponding to bays for incoming transmission lines,
bus section and bus couplers, power transformers, and MVfeeder circuits. In addition
there would also be diagrams containing information on such items as motors and
heating that operate on AC.
The layout of AC connections diagrams should be carried out observing the fol-
lowing points. Each diagram should include all equipment corresponding to a bay,
with breakers, disconnectors and transformers represented by schematic symbols. In
CT current circuits only the current coils of the measurement instruments and the
protection relays should be drawn, indicating clearly which coils are connected to
each phase and which to the neutral. The polarity of equipment should be indicated
on the drawings. It is useful to indicate equipment whose future installation can be
foreseen by means of dotted lines.
Solid-state relays should be represented schematically by squares, showing the
number of terminals and the method of connecting the wiring carrying the voltage and
current signals. The points where a connection to earth exists should also be indicated
in this diagram, for example when the neutral of the measurement transformers is
connected in star.
The main nominal characteristics should be marked close to each item of equip-
ment. For example, for power transformers, the voltage ratio, power rating, and
vector group should be provided; for power circuit breakers, the nominal and short-
circuit current ratings; the transformation ratios for voltage and current transformers,
and the nominal voltage of lightning arresters. Voltage transformer circuits should be
drawn physically separated from the rest of the circuits, and the connections to the
coils of the instruments that require a voltage signal should also be indicated. As a
minimum, the AC diagram of a transformer should include all the equipment in the
bay between the high voltage busbar and the secondary bushings of the transformer.
12.2.4 Diagrams of DC connections
Diagrams of DC connections illustrate the direct current circuits in a substation and
should clearly show the various connections to the DC auxiliary services. These
diagrams contain information corresponding to equipment such as breakers and
disconnectors, protection and control systems for transformers, busbars, transmis-
sion lines and feeders, annunciator systems, motor and heating circuits that operate
on DC, and emergency lighting and sockets. A diagram of connections for all
substation equipment that takes supplies from the DC system should be provided.
The positive infeeds are normally shown at the top of the diagram, and the nega-
tive ones at the bottomand, as far as possible, the equipment included in the diagrams
should be drawn between the positive and negative busbars. Due to the considerable
amount of protection and control equipment within a substation it is generally con-
venient to separate out the DC connections into different functional groups such as
control and protection equipment, and other circuits such as motors, heating etc.
It is common practice to draw dotted horizontal lines to indicate the demarcation
between the equipment located in the switchgear and that located in the protection
relay panels. It is useful if the signalling and control equipment in the relay and control
266 Protection of electricity distribution networks
panel is located in one part of the diagram, and the protection equipment in another
part. Every terminal should be uniquely identified on the drawing. As far as possible,
the contacts, coils, pushbuttons and switches of each mechanism should be drawn
together and marked by a dotted rectangle so that it is easy to identify the associated
equipment and its role in the circuit.
The internal circuits of the protection equipment are not shown, since it is suf-
ficient to indicate the tripping contacts and the points of interconnection with other
equipment inside a dotted rectangle. Given the complexity of distance relays, it might
be necessary to make a separate diagram to indicate their connections to the DC sys-
temand the interconnection of the terminals. It is also possible that separate diagrams
may be required for transformer and busbar differential protection.
Each power equipment bay should have two DC circuits; one for feeding the pro-
tection equipment and a separate one for signalling purposes and controlling breakers
and disconnectors. The two supplies should be kept independent of each other and
care should be taken to avoid connecting any equipment across the two DC supplies.
12.2.5 Wiring diagrams
Wiring diagrams show the interconnection of the multicore cables, for example
between the switchgear and the associated control panels, and the routeing of individ-
ual wires to the equipment installed in the relay and control panels. These diagrams are
required to facilitate the wiring of the measurement, protection and control equipment
at the substation construction stage. The wiring should be carried out in accordance
with the layout shown in the AC and DC diagrams.
It is logical that the layout of the different devices on the wiring diagrams should
be as seen from the rear of the relay and control panels, as in practice. Each device
should be represented by its schematic, with every terminal located in accordance
with its actual position on the panel. Each conductor should be marked with the same
identification code as the terminal to which it is connected, and also marked at each
end with the location of the far end of the conductor, according to a predetermined
code. To make the wiring easier to install, the location of the wires on the wiring
diagram should correspond to their proposed location inside the relay and control
panel. In the wiring diagrams the following elements should be uniquely identified –
terminals and sets of terminals; multicore cables that go to the switchgear; conductors
that go fromindividual terminals to equipment located in the relay and control panels;
equipment installed in the relay and control panels.
Multicore cables
Each multicore cable should have an identification number; in addition every con-
ductor in each cable should be numbered. It is useful if the numbering of multicore
cables is carried out consecutively by voltage level. With this in mind, an ample range
of numbers should be provided, for example multiples of 100 for each voltage level,
thus ensuring that there are sufficient spare consecutive numbers available for any
additional cabling in the future. All the conductors in the wiring diagram should be
marked at each end with the location of the far end of the conductor.
Protection schemes and substation design diagrams 267
2
1

Z
1
o
p
e
r
a
t
e
s
6
7
o
p
e
r
a
t
e
s
6
7
N
o
p
e
r
a
t
e
s
2
1

Z
2
o
p
e
r
a
t
e
s
2
1

Z
3
o
p
e
r
a
t
e
s
2
1

Z
4
o
p
e
r
a
t
e
s
2
1
N

Z
1
o
p
e
r
a
t
e
s
2
1
N

Z
2
o
p
e
r
a
t
e
s
2
1
N

Z
3
o
p
e
r
a
t
e
s
2
1
N

Z
4
o
p
e
r
a
t
e
s
O
p
e
r
a
t
e
s

8
5
E
n
e
r
g
i
s
e
s

2
1

Z
2
Y
e
s
N
o
E
n
e
r
g
i
s
e
s

7
9
R
e
c
l
o
s
i
n
g
s
w
i
t
c
h

p
o
s
i
t
i
o
n
O
f
f
1

3

1

+
3

S
y
n
c
h
r
o
n
i
s
m
Y
e
s
N
o
N
o
r
e
c
l
o
s
i
n
g
Y
e
s
S
i
n
g
l
e
p
h
a
s
e
f
a
u
l
t
N
o
R
e
c
l
o
s
i
n
g
E
n
e
r
g
i
s
e
s

t
h
e

t
r
i
p

c
o
i
l

N
o
.
1

o
f

t
h
e

f
a
u
l
t
e
d

p
h
a
s
e
E
n
e
r
g
i
s
e
s

t
h
e

t
h
r
e
e

t
r
i
p

c
o
i
l
s

N
o
.
2
E
n
e
r
g
i
s
e
s

t
h
e

t
h
r
e
e

t
r
i
p

c
o
i
l
s

N
o
.
2
1
1
S
y
n
c
h
r
o

c
h
e
c
k

r
e
l
a
y

i
s

i
n
c
l
u
d
e
d

o
n
l
y

w
h
e
n

s
e
v
e
r
a
l

s
o
u
r
c
e
s

a
r
e

a
v
a
i
l
a
b
l
e
E
n
e
r
g
i
s
e
s

2
1

Z
3
E
n
e
r
g
i
s
e
s

2
1

Z
4
F
i
g
u
r
e
1
2
.
1
3
P
r
o
t
e
c
t
i
o
n
l
o
g
i
c
s
c
h
e
m
a
t
i
c
f
o
r
1
1
5
k
V
l
i
n
e
b
a
y
268 Protection of electricity distribution networks
12.2.6 Logic diagrams
These diagrams represent the protection schemes for the different substation bays
by means of normalised logic structures in order to show in a structured manner the
behaviour of the substation protection system for any contingency. An example of
such a diagram for a 115 kV line bay at a substation is shown in Figure 12.13.
12.2.7 Cabling lists
Cabling lists provide information on the multicore cables that run between various
items of equipment and help to make it easier to verify the substation wiring for
maintenance work. The lists should include the following information:
• number, length and type of multicore cable;
• colour or number of each conductor in the multicore cable;
• identification of each end of the conductor;
• identification of the equipment at each end of the conductor;
• the function of the conductor.
Chapter 13
Processing alarms
13.1 General
The various protection schemes described in earlier chapters in this book are provided
to disconnect faults on power systems promptly, with the minimum of disruption to
the network and to customers. Alarms are triggered by the operation of the protection
relays and send coded information to distribution control centres so that the control
room operators are aware of what is happening on the network. Other alarms indicate
the state of the power system, for example voltages at various locations and load
flows on the more important circuits.
These alarms provide one of the main sources of information flowing in real-time
into a distribution control centre and are normally channelled to one printer in the
control room where a hard copy can be produced. The alarm streams are also chan-
nelled to the operator’s or control engineer’s console where they can be displayed on
computer screens. A third avenue for alarm streams is for them to be stored in a data
logger where the principle function is to retain a history of the alarm streams, which
can be used for post-system fault analysis if a serious fault occurs in a system and it
is thought necessary to carry out such an analysis (see Figure 13.1).
It is worth nothing that, with this arrangement, the alarms are not processed. Sev-
eral events may occur simultaneously, or at a close time proximity with respect to
each other, and each incident may trigger many alarms resulting in a large number
of alarms flowing into the control centre in close succession. The operator would
then have to use his/her judgement based on experience to decide what exactly
has happened to the system. Subsequent telephone calls from customers may also
help to determine the exact location of these incidents such as a blackout in a cer-
tain district. The aim of an alarms processor is to help the operator to arrive at
a sensible conclusion speedily and to discard redundant information in the alarm
streams.
270 Protection of electricity distribution networks
SCADA
computer
Telemetering information from outstations in the system
Computer logger
for
alarm stream
Operator console
screen showing
unprocessed
stream
Alarm logging
printer
Figure 13.1 Alarm streaming
An alarm is a signal that may be analogue or digital in nature and each alarm
carries certain information, which might include for example:
• notification of the operation of protective equipment including tripping, reclosing
and warning messages;
• indication of the change of state of a particular piece of equipment – e.g. a circuit
breaker opening or closing;
• warning of potential danger with plant, lines or cables unless remedial action is
taken – e.g. cable low oil pressure or tripping battery low;
• high or low voltage levels and loads on plant items.
Different utilities or power companies may use various protocols to identify their
alarmmessages – for example the letter ‘N’ could indicate a specific type of alarm. In
addition the priority level of the alarm can be shown, if required. It is also necessary
to be able to indicate the time of day, and the name of the substation where the alarm
has originated. The type of alarm also needs to be identified, e.g. ‘AUTOTRIP’ can
mean automatic trip. Then codes such as ‘OPRT

’ may be used to indicate that some
equipment has operated, with the

indicating the alarm has automatically reset itself.
Some messages are quite brief and give little information beyond, say, the fact that at a
particular time an automatic trip has operated at a named substation, and that the alarm
has reset itself. There is no information about which trip has operated, and whether it
has gone from an open state to a closed state or vice versa. Other types of alarm may
give more information. For example I 1729 JOHNSTON 33 BSECT OPEN could be
used to indicate that the 33 kV busbar section circuit breaker at Johnston substation
has opened at 1729 hours.
13.2 Alarm processing methods
Two simple methods of processing alarms are used widely in many existing control
centres. In one, all the alarms are assigned a certain level of fixed priorities and are
Processing alarms 271
Alarm logging
printer
Operator console
screen showing
processed and
unprocessed
alarms
Network
database
Alarm
processor
SCADA
computer
Computer logger
for alarm streams
Figure 13.2 Extended alarm processor
then displayed to the operator once they have been activated in their pre-fixed priority
level. The alarms may also have a fixed route to a specific console, which is often
related more to the geographical location of the alarm than to priority levels. The
second method is more flexible than the first. Under normal conditions, all levels
of alarms will be displayed. However, under emergency conditions, the operator has
the option to suppress certain levels of alarms coming to the console to avoid being
overloaded with incoming data. This enables the operator to be less stressed and more
able to cope with the incoming information and reach a decision on systemconditions.
An additional path can be added to the control room alarm system, comprising
a network database and an alarms processor (see Figure 13.2). The former contains
the original network topology and also real-time updated topology. It also contains
updated information on each substation including the rating and voltage levels of
transformers, busbar couplers, protection equipment, auto-reclosure equipment, fault
throwing switches and other information that the alarms processor unit takes into con-
sideration when interpreting alarm messages from the system. The network topology
information is updated in real-time while other information may be updated in an off-
line mode. This information, together with the alarm streams, is fed into the alarms
processor which will automatically identify the actual events on the system. The pro-
cessed alarms, together with the unprocessed ones, will be on display to the control
room operator. The system acts as an aid to the operator who will then have to decide
whether the processor has arrived at a sensible conclusion about the events.
13.3 Expert systems
The term expert system is used in reference to a computer program that is assumed
to have the same knowledge as a human expert, and which will attempt to simulate
his/her reasoning process in order to arrive at a solution to a given problem or a given
set of conditions. In its simplest form the program will be structured in the form of
272 Protection of electricity distribution networks
a tree, with each option representing a node with only two branches connected to it.
The answer of ‘yes’ or ‘no’ to each option will allow the program to proceed along
one of these branches that leads to the next option. The decision to go along one
branch instead of the other is governed by so-called rules. In this case it is a simple
yes/no rule. Ultimately the systemshould arrive at a solution of the network problem.
This type of expert system is relatively easy to construct, but it is not suitable for
complex tasks such as alarm processing.
There are many techniques or approaches for alarm processing and each has
its own advantages and disadvantages. Each approach has its own form of present-
ing alarms to the operator and in general they can be classified into four principle
approaches. The expert system should embody the essence of all these approaches,
which are summarised below.
Alarm processing
This is an important approach for a large control roomthat covers a wide geographical
area, and where each operator is responsible for a specific region within this area.
Alarms in a particular region are only presented to the relevant operator’s console.
However, top-level alarms that are considered to have grave severity may also be
diverted and displayed simultaneously at the senior operator’s console.
Alarm prioritisation
In this approach, alarmmessages are displayed in a priority order list on the operator’s
console. This approach may be in addition to the alarm processing referred to in the
paragraph above.
Alarm filtering
This approach aims to reduce the number of alarms each operator receives. It elim-
inates and suppresses alarms that simply duplicate the information from alarms
received earlier. Finally, alarm reduction would be possible by suppressing all alarms
that are expected, for example, as a result of an operator’s actions or a correct
protection system operation.
Alarm summarising
In this approach, alarms relating to one perceived event are replaced by a simple
summary message stating the possible cause of all these alarms.
13.4 Equivalent alarms
Alarm systems in an electricity utility are often not installed at the same time, but as
required as the electrical network develops. In addition, they may have been installed
by different manufacturers using different parts, so that the alarms that arrive at
the control room for the same type of event may be constructed differently. For
example, the message ‘alarm suppressed’ may be displayed in a number of different
Processing alarms 273
ways, e.g. ALM.SUPP, ALMS SUP, ALMSUPP, ALRMSUP, etc. Examples of other
equivalent alarms are IT RECV or I/T REC for intertrip signal received, 33AR ALM,
33KV A/R, AR ALM33 for an alarm originating from a 33 kV substation, and CP
LO, LOP, LO OIL P, CABL LOP for a cable low oil pressure alarm.
13.5 Rules
The reasoning mechanism of an expert system is via the rules set up within the
program. Some of these rules are illustrated below:
• event – successful closure of {CB} under operator control, where {CB} is the
description of a given circuit breaker in a substation; expect that an operator
closes the circuit breaker – action OpAction Close {CB} – and that the indication
CLOSED {CB} is received;
• event – volts high on main busbar 1 at {StationName}; expect to receive alarm
MN.B/B.1 HIGH OPERATED {StationName};
• event – {StationName}; voltage on main busbar 1 returned to normal; expect to
receive alarm MN.B/B.1 HIGH_RESET {Station Name}.
The first alarm shown above is an event that indicates a successful closure of
a specific circuit breaker under operator instruction. In order for this event to be true,
two actions are needed. The first one is the operator action leading to the closing of the
{CB}, and the second one is the alarm that comes back from the substation indicating
that the {CB} has closed. The rules are simple but the different combinations and
the overlaying of these rules can be used to reason out complex situations. Thus
equivalent messages allow the system to interpret different descriptions of alarms
that have the same meaning, while rules are used to interpret the combined meaning
of alarms.
13.6 Finger printing approach
One approach that may be used to handle situations of wide variety and complexity is
the finger printing approach which, as the name suggests, involves matching current
alarms with the set of alarms that would be expected for a particular type of system
event. If there is an exact match then it is concluded that this event has occurred.
Consider the network in Figure 13.3 where, because of the indicated fault, the
line between substations A and C has tripped out. The fault will generate a series of
alarms as shown on Table 13.1.
The first alarm indicates that an automatic trip has operated in substation A while
the second alarmindicates that circuit breaker 5 in substation Ahas opened. Similarly
the third alarm indicates that an automatic trip has operated in substation C with the
subsequent two alarms indicating that circuit breakers 2 and 3 at substation C have
also opened. In a finger printing approach, such an alarm list is stored in the database
274 Protection of electricity distribution networks
cb2
T2
cb3
cb1
T1
cb2
T2
cb1
T1
cb3
cb3
T2 T1
cb5
cb2 cb4
cb1
Substation B Substation C
Substation A
Figure 13.3 Network example
Table 13.1 Alarms for network in Figure 13.3
Time Substation Circuit Action
1033 SUBSTN A AUTOTRIP OPRT*
1033 SUBSTN A CB5 OPEN
1033 SUBSTN C AUTOTRIP OPRT*
1033 SUBSTN C CB3 OPEN
1033 SUBSTN C CB2 OPEN
and, if an exact match occurs, then it can be concluded that there has been a fault in
line AC.
The finger printing approach works well under an ideal situation. However, in an
alarms processing situation, it is not the best choice. This is because it is unable to
cope with missing alarms or a piecemeal arrival rate of alarms, or to relate to different
system events occurring simultaneously.
Processing alarms 275
13.7 Hypothesis approach
An alternative to the fingerprinting method is the hypothesis approach. In this method
a database is set up identifying those alarms that should be received at the control
centre for any incident on the power system. As alarms arrive, a series of hypotheses
that might account for the alarms receivedsofar canbe generatedandthese hypotheses
can later be confirmed or discarded as more alarms arrive progressively at the control
centre. Figure 13.4 provides a simplified schematic diagram of the flow chart for a
hypothesis expert system developed at the Department of Electronic and Electrical
Engineering, University of Strathclyde, Glasgow, UK.
Yes
Processor
(1)
Incoming
alarm
Alarm
fits existing
hypotheses?
(2)
No
Hypothesis
generator
(4)
Printer
(6)
Yes
Full
score and/or
timed out?
(5)
No
+1
(3)
Figure 13.4 Flow chart for a hypothesis expert system
276 Protection of electricity distribution networks
When an alarm is fed into the input processing section of the expert system (1), it
is broken down into separate fields and converted for ease of manipulation. The next
stage is to check if the incoming alarm fits into any of the existing hypotheses (2). If
the result is positive, the point scores of those hypotheses containing this alarm are
increased (3) and the results are passed on to the evaluation stage (5) where each
hypothesis is checked to see if any have reached a full points score, and/or reached
the pre-determined time that could be one or two polling periods, depending on the
nature of the telemetering system.
If a full points score has been achieved then the programconcludes that a solution
has been found and prints out a summary message on the operator’s console (6). If
any hypothesis has timed out without all the expected alarms having arrived then the
most likely hypotheses are printed out, indicating those alarms that were expected
but had not been received. If a hypothesis has not reached a full score nor timed out,
then this is recycled back to the input section to repeat the checking process to see if
any other relevant alarms have been received.
If the incoming alarmdoes not fit into any of the hypotheses stored in the program
then newhypotheses have to be generated (4). It may be necessary to expand the rules
to form completely new hypotheses, and produce listings of expected alarms for each
new hypothesis. These alarms are then transferred to the evaluation stage (5) and
processed to check if a full points score has been reached and/or any of the hypotheses
have timed out, as referred to in the previous paragraph.
Referring to the alarms list in Table 13.1, relating to the system shown in
Figure 13.3, under the hypothesis approachthe arrival of the first alarm1033SUBSTN
A AUTOTRIP OPRT

would result in the generation of six hypotheses:
1. Fault in transformer 1 (T1) at substation A.
2. Fault in transformer 2 (T2) at substation A.
3. Fault on line AB.
4. Fault on line AC.
5. Fault on transformer 1 (T1) at substation B.
6. Fault on transformer 2 (T2) at substation C.
The arrival of the second alarm 1033 SUBSTN A CB5 OPEN helps to reject
hypotheses 1, 2, 3, and 5, since none of these faults would have required circuit
breaker 5 at substation A to open in order to isolate any one of these faults. The third
and fourth alarms do not help to eliminate either of the remaining two hypotheses
(4 and 6). However, they help to confirmthat these two hypotheses are still valid. The
arrival of the fifth alarm, 1033 SUBSTN C CB2 OPEN, does not alter the situation.
Hypotheses 4 and 6 are still valid. However, for hypothesis 6, (fault on transformer 2 at
substation C), one further alarmwould be expected – SUBSTNCTR2 TRANS PROT
OPRT – indicating that the protection for transformer 2 at substation C had operated.
Under such circumstances, the alarms processor may have to wait for a given period
before it can arrive at a decision. As no further alarm arrives, hypothesis 4 – fault on
line AC – becomes the selected solution. This example illustrates the basic principle
of the hypothesis approach, but it also raises several important points, such as how
Processing alarms 277
long the alarms processor should wait before arriving at a decision, how to deal with
missing alarms and with simultaneous events.
The following examples demonstrate different capabilities of the expert system
and the inference mechanism behind the system.
Example 13.1
The first example covers the substations at Iverley, NortonandPedmore inthe network
shown diagrammatically in Figure 13.5, where circuit breakers marked with a circle
operate in a normally open position.
The alarms received for this event are shown in Table 13.2.
From Table 13.2, the first alarm for this event arrives at 0821 stating 0821
IVERLEYAUTOTRIP OPRT

. The alarms processor recognises this as meaning that
there has been an automatic trip of one or more circuit breakers at Iverley substation
11kV
bb3
33kV
2
bb1
T2
Iverley
substation
T3
bb2
33 kV
4
5
6
T1 T2
3
Norton
substation
T1 T2
1 2
bb1 bb2
11kV
4
T1
1
11kV
bb4 bb5
Pedmore
substation
3
Figure 13.5 Network for Example 13.1
278 Protection of electricity distribution networks
Table 13.2 Alarms received for Example 13.1
Time Substation Circuit Action
0821 IVERLEY AUTOTRIP OPRT

0821 IVERLEY T3 33 KV CB OPEN
0821 NORTON AUTOTRIP OPRT

0821 NORTON NRTN-IVLY/PED 2 OPEN
0821 NORTON NRTN-IVLY 3 OPEN
0822 PEDMORE T2 11 KV CB OPEN
due to protection operation. As only one alarm has been received so far, the alarms
processor initiates hypotheses for a fault on each of the three circuits connected to
Iverley, as well as the 33 kV and 11 kV busbars and the three 33/11 kV transformers.
Some examples of the hypotheses that would be generated are:
1. Fault on transformers 1, 2 or 3 (T1, T2, T3) at Iverley.
2. Fault on the 33 kV busbars 1 or 2 (bb1, bb2) at Iverley.
3. Fault on the 11 kV busbars 3, 4 or 5 (bb3, bb4, bb5) at Iverley.
4. Fault on Norton-Iverley teed Pedmore lines 1 or 2.
5. Fault on Norton-Iverley line 3.
It can be seen that a large number of hypotheses can be generated. However, the
arrival of the next alarm 0821 IVERLEY T3 33 KV OPEN, which refers to the
33 kV circuit breaker 3 at Iverley, helps to pinpoint the situation more specifically by
eliminating a number of possible hypotheses.
Two likely events are:
A. A fault on Iverley 11 kV busbar 5, successfully isolated by protection. The
expected alarms are:
• Iverley T3 33 kV circuit breaker 3 open (received).
• Iverley autotrip operated (received).
Score: 2
B. A fault on one or more of the following circuits successfully cleared by
protection – the Norton-Iverley teed Pedmore line 2, Norton-Iverley line 3,
Pedmore transformer 2, Iverley 33/6.6 kV transformer 3.
The expected alarms are:
• Iverley autotrip operated (received).
• Norton circuit breaker 2 open.
• Norton circuit breaker 3 open.
• Norton autotrip operated.
• Iverley 33 kV circuit breaker 3 open (received).
Processing alarms 279
• Pedmore transformer 2 11 kV circuit breaker open.
• Pedmore autotrip operated.
Score: 2
The next alarm received – 0821 NORTON AUTOTRIP OPRT

– is expected
if hypothesis B is correct, so the score for this hypothesis is increased by a further
point to 3. The following three alarms also contribute to hypothesis B, which finally
reaches a score of 6.
One expected alarm – PEDMORE AUTOTRIP OPRT

– has not been received.
However, this does not invalidate hypothesis B as being the preferred solution and
the alarms processor prints out hypothesis B as the final solution. This would include
a list of nodes describing the fault area since it is not possible to be more specific
about the location of the fault. The network bounded by the Norton-Iverley line 3, the
Norton-Iverley teed Pedmore line 2, transformer 2 at Pedmore, busbar 2 at Iverley,
and transformer 3 at Iverley is defined as the blackout area.
Figure 13.6 shows the single line diagram for the next two examples. The first
one is centred around Enville substation, and the second one around Clent substation.
Prestwood
substation
4
T1
1
T2
2
4
1
3
T1
T1
bb1
2
T2
5
T1
11kV
Clent
substation
Stourton
substation
Enville
substation
33 kV
bb1 bb2 bb2
bb4 bb3
11 kV
3
33 kV
Figure 13.6 Network for Examples 13.2 and 13.3
280 Protection of electricity distribution networks
It should be noted that those circuit breakers marked with a circle normally operate
in the open position.
Example 13.2
Upon the arrival of the first alarm of this event – 1425 ENVILLE AUTOTRIP
OPRT

– the alarms processor initiated a number of hypotheses to cover all the
possible fault locations:
1. Fault on transformer 1 (T1) at Enville, 33 kV busbar 1 (bb1) at Enville, and the
Prestwood-Enville circuit.
2. Fault on transformer 2 (T2) at Enville, 33 kV busbar 2 (bb2) at Enville, the
Enville-Stourton circuit, transformer 1 (T1) at Stourton.
3. Fault on 11 kV busbar 3 (bb3) at Stourton.
4. Fault on 11 kV busbar 4 (bb4) at Stourton.
The arrival of the second alarm– 1425 ENVILLE 33 KVB.SECT OPEN

– indicates
that the 33 kV bus-section circuit breaker at Enville was opened and then reclosed.
With these two alarms having been received, which are relevant to the first two
hypotheses, these two hypotheses are thus awarded a score of 2.
The most likely hypotheses can now be narrowed down to:
A. Successful reclosure after a fault on Enville transformer T1, 33 kV busbar 1 at
Enville, or on the Prestwood-Enville circuit. For this hypothesis the expected
alarms are:
• ENVILLE AUTOTRIP OPRT

(received).
• ENVILLE T1 11 KV CB OPEN

.
• ENVILLE 33 KV B.SECT OPEN

(received).
• PRESTWOOD AUTOTRIP OPRT

.
• PRESTWOOD CB4 OPEN

.
Score: 2
B. Successful reclosure after a fault on Enville transformer T2, 33 kV busbar 2 at
Enville, Enville-Stourton circuit, Stourton transformer T1, or 33 kV busbar1
at Stourton. The expected alarms for this hypothesis are:
• ENVILLE AUTOTRIP OPRT

(received).
• ENVILLE T2 11 KV OPEN

.
• ENVILLE 33 KV BSECT OPEN

(received).
• STOURTON AUTOTRIP OPRT

.
• STOURTON T1 11 KV OPEN

.
• STOURTON 33 KV BSECT OPEN

.
Score: 2
Referring to the alarms listed in Table 13.3, the arrival of the third alarm – 1425
ENVILLE T1 11KV CB OPEN

– relates to T1 11 kV circuit breaker at Enville,
which has opened and successfully reclosed. Therefore the score for hypothesis A
Processing alarms 281
Table 13.3 List of alarms received for Example 13.2
Time Substation Circuit Action
1425 ENVILLE AUTOTRIP OPRT

1425 ENVILLE 33KV B.SECT OPEN

1425 ENVILLE T1 11KV CB OPEN

1425 PRESTWOOD PRSTWD-ENVIL OPEN

Table 13.4 Alarms for Example 13.3
Time Substation Circuit Action
1748 CLENT T1 33 KV CB OPEN
1748 CLENT T1 33 KV CB O/S
1750 CLENT T1 11 KV CB O/S
has its score increased by 1 to 3, whereas hypothesis B still has a score of 2, and the
other hypotheses are no longer viable.
The fourth alarm – 1425 PRESTWOOD PRSTWD-ENVIL OPEN

– indicates
that circuit breaker 4 at Prestwood, controlling the circuit to Enville, has opened and
subsequently reclosed. Since this message is expected for hypothesis A, its score is
increased incrementally once more to a value of 4. This leaves hypothesis A short of
one alarm – PRESTWOOD AUTOTRIP OPRT

. However, since the alarm PREST-
WOOD CB4 OPEN

has been received, it is not necessary to wait for this final
alarm before being able to arrive at a decision. If no more alarms relating to these
two hypotheses arrive within the next polling period then the alarms processor has
to weigh up these hypotheses and make a decision as to the most likely one. On this
case, hypothesis A has the highest score and is selected as the preferred event.
Example 13.3
For this event the alarms are shown in Table 13.4.
The first alarm indicates that the 33 kV circuit breaker at Clent substation has
opened. Such a message could indicate either of the following likely possibilities.
1. The circuit breaker has opened due to an autotrip initiated by protection operation.
Therefore, further alarms would be expected together with appropriate autotrip
alarms.
2. The circuit breaker has been opened prior to maintenance work being carried out
on associated equipment.
The next two alarms add weight to the probability that some form of outage is
being performed, with two hypotheses being presented to the operator.
282 Protection of electricity distribution networks
A. Fault on Prestwood-Stourton teed Clent circuit, Clent transformer 1, Stourton
33 kV busbar 2. For this hypothesis the expected alarms are:
• CLENT T1 33 KV CB OPEN (received).
• PRESTWOOD CB5 OPEN.
• PRESTWOOD AUTOTRIP OPRT

.
• STOURTON AUTOTRIP OPRT

.
• STOURTON 33 KV CB OPEN.
• CLENT T1 11 KV CB OPEN (received).
• CLENT AUTOTRIP OPRT

.
This results in a score of two alarms received out of a possible seven.
B. Suspected maintenance on Clent transformer 1, Prestwood-Stourton teed
Clent circuit, and Stourton 33 kV busbar 2. The expected alarms for this
hypothesis are:
• PRESTWOOD CB5 OPEN.
• PRESTWOOD CB5 O/S.
• STOURTON CB2 OPEN.
• STOURTON CB2 O/S.
• CLENT T1 33 KV CB OPEN (received).
• CLENT T1 33 KV CB O/S (received).
• CLENT T1 11 KV CB OPEN (received).
• CLENT T1 11 KV CB O/S.
Score: 3
The scores for both hypotheses are not high, but the fact that no AUTOTRIP
alarms have arrived strengthens the hypothesis that the event is due to maintenance
work rather than a fault event, which should have initiated protection operation.
After waiting for a further polling period to ensure that no more alarms related to this
event have arrived, all the competing hypotheses are weighed up, and the solution of
hypothesis B is chosen.
The three examples have illustrated three different conditions. The first one had
a full score and the system identified a blackout area with the names of the nodes
identifying that area. The second example illustrates the ability of the system to cope
with missing alarms, and detected a successful reclosure after a fault in a given area
identified by a set of node names. The third example shows how the alarms processor
copes when there are substantial missing alarms and the system can only conclude a
suspected situation.
An alarms processor programmed to operate on a hypothesis approach basis can
cope with a large number of incoming alarms from different network incidents, even
if these are time-coincidental, for example during a thunderstorm with multiple light-
ningstrikes onthe power system. The abilityof the process toassemble the appropriate
alarms together for each incident, and to reach a decision on each incident, provides
a powerful aid to the system control room operator when the alarm load is most
demanding and is also a most useful analytical tool for subsequent fault analyses.
Chapter 14
Installation, testing and maintenance of
protection systems
Although the aim of this book is to provide the basis to guarantee a suitable relay
setting procedure in distribution networks, it is felt that some reference should be
made to the installation, testing and maintenance of protection systems. No matter
how well the relay applications are carried out, a protection scheme is worthless if its
actual performance cannot be guaranteed. It is important toemphasise that a protection
scheme covers not only the relays but also the CTs and VTs that feed them, and the
circuit breakers that open the circuits on receipt of a trip signal from the relays when
a fault occurs.
14.1 Installation of protection equipment
The installation of protection relays should be carried out following the instructions
normally included by manufacturers in their service manuals. The relays are installed
on control panels that should preferably be situated in areas free of dust, damp,
excessive vibration, and heat. The older types of relays may be set into the panel
(flush fitted) or mounted on the face of the panel with a suitable casing. Multifunction
relays are generally mounted on standard 19-inch rack cases. The heights of the cases
are specified in terms of the number of standard rack units, which typically range
from 2 to 5 rack units. The depth of rack mount cases varies depending on the relay
model. Some relay models include built-in facilities for testing the relay in the case.
Other relay models require external provisions to test the relay in its case. Figure 14.1
shows typical layouts of protection and control panels for HV and MV substation
bays. The dimensions correspond to typical equipment from various manufacturers.
The electrical connections to the relay terminal should, as far as possible, be made
with flexible stranded copper wiring, and the terminals should be designed so that
a solid connection with effectively no resistance can be guaranteed. Great care must
be taken to ensure that the CT secondary wiring circuits are not open-circuited when
primary current is flowing since this could cause damage to the relays, wiring and the
284 Protection of electricity distribution networks
HV coupler Feeder circuits Transformer HV line HV line
86 86
Breaker fault
protection
67
86
86
86 67
N
79
67 67
51 51 51
86
50
86
N
51
50
87
50 50
87 87
21
79
21 21
51
N
50
50
51
50 50
51 51
51 51 51
50
51
67
86 86 86
51
N
50
67
50
N
51
N
51
50
51
50
51
50
50
50 50 50
87 87 87
50
51
51
Busbar II
87 87 87
Busbar I
50
51
(a)
(b)
(c)
Single busbar Double busbar
Disconnector
control switch
Double busbar
Push buttons
for tap changer
Tap position
indicator
Cooling
fan switch
A
MVAR
Circuit breaker
control switch
V
MW MVAR MW
V
A A A
HV line
A A
HV line
Temperature
indicator
C
V
Transformer
A A A
Single busbar Double busbar
Busbar circuit
Feeder circuits
V V
HV coupler
A A V
HV line HV coupler Transformer HV line
2
2
0
0
800
2
2
0
0
600
600
2
2
0
0
Feeder circuits
Figure 14.1 Layout of equipment on control panels: (a) front viewof panel; (b) rear
view of panel with electromagnetic relays; (c) rear view of panel with
numerical relays
Installation, testing and maintenance of protection systems 285
CTs due to the high voltage that can be generated. Before starting to test the relays,
any mechanical fastening that the manufacturer has placed on the moving parts to
prevent damage during transportation and installation should be removed.
14.2 Testing protection schemes
Testing protection equipment can be divided into three types:
• factory tests;
• pre-commissioning tests;
• periodic maintenance tests.
14.2.1 Factory tests
It is the responsibility of the relay manufacturer to carry out suitable tests on all protec-
tion equipment before it is delivered and put into service. Since the protection relays
are required to operate correctly under abnormal system conditions, it is essential
that their operation be guaranteed under such conditions. The simulation of opera-
tional conditions for test purposes is normally carried out during the manufacture and
certification of the equipment. These tests should be the most rigorous possible in
order to ensure that the protection will operate correctly after transportation to, and
installation at, the substation site.
The factory operating tests can be divided into two main groups – those tests in
which the relay parameters are reproduced, and those involving the simulation of
conditions such as temperature, vibration, mechanical shock, electric impulse, etc.,
which could affect the correct operation of the relay. In some cases, tests from both
groups can be carried out simultaneously.
14.2.2 Precommissioning tests
The most important precommissioning tests and in-service checks can be summarised
as below:
• analysis of the wiring diagrams to confirm the polarity of connections, positive-
and negative-sequence rotation, etc.;
• a general inspection of the equipment, physically verifying all the connections,
at both the relay and panel terminations;
• measurement of the insulation resistance of the protection equipment;
• inspection and secondary injection testing of the relays;
• testing current transformers;
• checking the operation of the protection tripping, and alarm circuits.
The precommissioning tests need to be carefully programmed so that they take
place in a logical and efficient order, in order that no equipment is disturbed again
during subsequent tests. Before starting the tests it is essential to ensure that the
assembly of the particular item being tested has been completed and checked. In
286 Protection of electricity distribution networks
addition, the list of the tests to be carried out should be arranged in a chronological
order together with any precautions that need to be taken into account. Some of the
more usual tests are briefly described below.
Insulation resistance measurement
This test should be carried out with the aid of a 1000 volt Megger. It is difficult to be
precise as to the value of resistance that should be obtained. The climate can affect
the results – a humid day tends to give lower values, whereas on a dry day much
higher values may be obtained.
Secondary injection tests
These tests are intended to reproduce the operating conditions for each relay and are
limitedsolelytothe protectionas such, sothat it is important toreadandunderstandthe
relay instruction manual (application, operation, technical characteristics, installation
and maintenance). In order to carry out these tests it is necessary to electrically
isolate the relay by means of test plugs or physically withdraw the relay from its case.
Although the relays should have been carefully tested in the manufacturer’s works, it
is necessary to make some checks on site after they have been mounted on the panels
in order to be sure that they have not been damaged in transit to the installation. The
actual tests carried out on the relays depend largely on the type of relay.
Secondary injection tests are required to ensure that the protective relay equip-
ment is operating in accordance with its preset settings. Relay inputs and outputs must
be disconnected prior to performing these tests. The test equipment supplies the relay
with current and voltage inputs that correspond to different faults and different oper-
ating situations. Pick-up values are reached by gradually changing the magnitudes of
these inputs while simultaneously measuring the relay operating time. Tripping con-
tacts and targets must be monitored during these tests in order to ensure that the relay is
working according to the manufacturer’s specifications and the settings that have been
implemented. If the curves and characteristics of a relay are to be tested at many points
or angles, it is convenient to use test equipment that can conduct a test automatically.
Modern protective relay test equipment has the option of performing automatic
tests aided by software programs, for which the testing process is much faster and
more precise. In addition, the time during which the relay is out of operation is
minimised. Figure 14.2 shows a typical layout of a protective relay test equipment.
This equipment is able to provide current and voltage injection as well as phase
shift when testing directional protection. It thus permits testing of a wide variety of
relay types such as overcurrent, directional overcurrent, reverse power, distance and
under/overvoltage units.
It is very important to record all the test results, preferably on special forms for
each type of relay. For example, a typical pro-forma for overcurrent relay tests, as
shown in Figure 14.3, could have the following information recorded:
• basic data about the circuit supplying the relay;
• the settings used before any tests commenced, which had been applied in accor-
dance with the protection co-ordination study. This information should include
the pick-up current, time dial, and instantaneous settings;
Installation, testing and maintenance of protection systems 287
INT INT
STOP START
0
1
0
2
0
3
0
4
0
50
6
0
7
0
8
0
9
0
1 0 0
%
SERIAL
RELAY TEST UNIT
SWERKER 750 Programma
V
A
I
0 1 I >
U>
I >
I >
I >
I >>
I >>
I < I >>
Trip

+
U
U
NI
NI
RI
RI
Figure 14.2 Application example of a relay testing unit (Programma Sverker
750). (Reproduced by permission of GE Power Systems-Programma
Electric AB)
• operating times for different multipliers, as measured by calibration tests. These
should be checked against the data provided by the manufacturer;
• test data for the instantaneous units;
• finally, the equipment used in the test should be recorded together with
any relevant observations, plus details of the personnel who participated in
the test.
It is important to note that the tests referred to beforehand correspond to steady-
state conditions, and the equipment to carry them out is rather conventional. As
a consequence of technological developments, more sophisticated equipment is now
available to undertake tests using signals very similar to those that exist during fault
conditions. Since relays are required to respond to the transient conditions of disturbed
power systems, their real response can be assessed by simulating the signals that
go to the relays under such conditions. Several manufacturers offer equipment to
carry out dynamic-state and transient simulation tests. A dynamic-state test is one
288 Protection of electricity distribution networks
GERS
Consulting Engineers
PROJECT
1. SETTINGS
2. PHASE UNIT TESTS
2.1. Overcurrent Pick up
2.2. Check of Time Operation Curve
2.3. Operation Curve (Extremely Inverse)
2.4. Instantaneous Pick Up
Parameter
Parameter
Parameter
Multiples of Pick Up Current
Phase
A
B
C
Theoretical Result
4.30
NA
4.31
0.00%
0.23%
% Error
Current Transformer (CT prim. Amps)
Phase
600 600
Current Transformer (CT sec. Amps) 5 5
51 Relay Curve U4 (Extremely Inverse) U3 (Very Inverse)
51 Primary Pickup (Amps) 520 535
51 Tap 4.3 4.5
51 Time Dial 2.5 2
50 Primary Pickup (Amps)
Pick up current [A]
Injected current [A] Fixed
2.0 2.5 3.0 3.5 4.0 4.5
8.60 10.75 12.90 15.05 17.20 19.35
Operation Time [s] Theoretical 4.813 2.788
Operation Time [s] Phase A Measured 4.90 2.75
% Error-Phase A 1.81% 1.36%
Operation Time [s] Phase B Measured NA NA
% Error-Phase B NA NA
Operation Time [s] Phase C Measured 4.79 2.78
% Error-Phase C 0.48% 0.29%
1.860 1.348 1.033 0.824
4.30
Phase
A
B
C
Theoretical Parameter Result % Error
Instantaneous Pick Up [A] Aprox NA
2.5. Instantaneous Pick Up
Phase
A
B
C
Theoretical Parameter Result % Error
Operation Time [cycles] NA
2.6. Signaling Tests
3. TEST EQUIPMENT USED
4. REMARKS
Phase Phase
A
B
C
A
B
C
NA
NA
NA
Signaling Test Test Signaling
Time Overcurrent
Instantaneous
Overcurrent
OK
OK
OK
Not Used Not Used
50 Tap Not Used Not Used
50 Time Delay (cy) Not Used Not Used
Neutral/Ground
TEST REPORT
DATE
TESTED BY
APPROVED BY
50/51-50/51N PHASE AND GROUND OVERCURRENT RELAY
MANUFACTURER TYPE
551 SEL
LOCATION
ER-1
SERIAL NUMBER
2000263013
CIRCUIT
R-MSW1-M-B
10.00
1.00
0.10
1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
T
i
m
e

(
s
e
c
s
)
Multiples of Pick Up Current
Real Theoretical
Reviewed Approved
Initials/Signature
Figure 14.3 Typical test report sheet for overcurrent relays
Installation, testing and maintenance of protection systems 289
in which phasor test quantities representing multiple power system conditions are
synchronously switched between states. Power system characteristics, such as high
frequency and DC decrement, are not, however, represented in this test. A transient
simulation test signal can represent in frequency content, magnitude, and duration,
the actual input signals received by a relay during power system disturbances.
Current transformer tests
Before commissioning a protection scheme, it is recommended that the following
features of current transformers are tested.
Overlap of CTs
When CTs are connected in order that a fault on a breaker is covered by both protection
zones, the overlap connections should be carefully checked. This should be carried
out by a visual inspection. If this is not possible, or difficult, a continuity test between
the appropriate relay and the secondary terminals of the appropriate CT should be
carried out.
Correct connection of CTs
There are often several combinations of CTs in the same bushing and it is important to
be sure that the CTs are correctly connected to their respective protection. Sometimes
all the CTs have the same ratio but much different characteristics, or the ratios are
different but the CTs are located close together, which can cause confusion.
Polarity
Each CT should be tested individually in order to verify that the polarity marked on
the primary and secondary windings is correct. The measuring instrument connected
to the secondary of the CT should be a high impedance voltmeter or a moving coil
ammeter, with centre zero. A low voltage battery is used in series with a push-button
switch in order to energise the primary. When the breaker is closed the measuring
instrument should make a small positive deflection, and on opening the breaker there
should be a negative deflection, if the polarity is correct.
Primary injection test
This test checks out all the protection system, including the current transformers.
The principal aims of this test are to verify the CT transformation ratios and all the
secondary circuit wiring of both the protection and the measurement CTs so that the
operation of the tripping, signalling and alarm circuits is confirmed.
Figure 14.4 shows the schematic diagramof a typical equipment used for carrying
out primary injection tests. The test current is usually between 100 and 400 A. The
two high current terminals of the primary injection equipment marked <➀> and
<➁>should be temporarily connected to the terminals of the CT that is being tested.
290 Protection of electricity distribution networks
(N)
240V
a.c
(L)
Variac
2
High current
test transformer
1
Metering CT and
temperature
measurement
A
Figure 14.4 Equipment arrangement for primary injection test
14.2.3 Periodic maintenance
Protection equipment can be inactive for months; however, it is required to operate
swiftly and accurately if a fault occurs on its associated primary equipment. Perhaps
one of the most difficult requirements of the protection is that it should remain inop-
erative for external and close faults. Periodic maintenance on protection equipment
should be carried out in order to be sure that the protection is always able to function
correctly as and when required.
When a maintenance programme has been produced, the frequency of mainte-
nance will depend upon the history of each item of equipment and any tendency
to faults. Excessive testing should be avoided so that the maintenance programme
will show up faults, not cause them. The frequency of maintenance, however, will
vary greatly with the type of equipment. Certain items should be checked con-
tinuously, while others should be tested weekly, monthly and even once a year
or more.
The classes of equipment that should be maintained are:
• protection relays;
• auxiliary control relays;
• annunciators and alarm systems;
• additional control devices (knobs, keys, interlocks, etc.);
• fault recorders if they are stand-alone devices.
It should be noted that, for an electrical utility where rationing energy could
have grave consequences, taking equipment out of service for maintenance should
be minimised by suitably programming the work in such a way that supplies are
not jeopardised. Wherever possible, maintenance of protection equipment should
be arranged to take place at the same time as the primary equipment is out of
service.
Installation, testing and maintenance of protection systems 291
Methodology
In order to achieve the maintenance objectives, a co-ordinated programme of work,
evaluation and re-scheduling is required, as indicated below:
• set up a maintenance programme on a six month or one year basis, with tentative
dates for carrying out work on individual items of equipment;
• evaluate the performance of the programme at the end of each six month period
to obtain results for analysis;
• take into account the results of the previous period when preparing the new
programme, and in particular the maintenance frequency for each item of
equipment;
• re-schedule the preventative maintenance programme when immediate and
corrective maintenance actions are required;
• co-ordinate with other maintenance groups beforehand, in the case of other
equipment that may also require maintenance;
• file and analyse the results obtained with the aim of improving the performance
indices, and reducing the costs.
Maintenance criteria
Within the maintenance programme, various categories of work can be grouped in
general terms such as the following. Preventative maintenance is those jobs that
are carried out on the basis of the maintenance programme. Corrective maintenance
covers those jobs that are not programmed and require immediate attention.
Additionally, there are two different ways of achieving the maintenance work. On
line: when, in order to carry out the work, it is not necessary to take out of service the
high voltage equipment associated with the relays or instruments that need checking.
Off line: when it is necessary to take the associated high voltage equipment out of
service.
The principal sources of establishing the maintenance criteria are:
• recommendations made by the manufacturers;
• analysis of fault statistics;
• network voltage level;
• types of relay or instrument used;
• relative importance of the protected equipment;
• maintenance experience;
• previous behaviour of the equipment;
• type of fault, if a fault recorder operates.
Starting fromthe fact that the operating time of relays is small in comparison to the
operating time of the power equipment, the fundamental principle is to carry out the
least testing to give the best operation. This philosophy tends towards extending main-
tenance intervals, dependingonthe circumstances. The inspectioninterval depends on
various factors, and principally on the history of the installation under test, the type of
292 Protection of electricity distribution networks
electrical protection used (electromagnetic, solid state, microprocessor or numerical),
the importance of the equipment in relation to the power systemand the consequences
of an incorrect relay operation, and environmental conditions. On the basis of these
factors, criteria for the frequency of maintenance in protection and control can be
established, depending on whether this is preventative or corrective work.
14.3 Commissioning numerical protection
As indicated in Section 3.4, numerical protection is based on a number of micropro-
cessors, which can be programmed to reproduce virtually any type of protection and
relay. It represents important advantages in reliability, space, and economics com-
pared to conventional protection arrangements. The method of calculating settings on
numerical protection does not differ greatly from that used for conventional protec-
tion, which has been described in earlier chapters. However, there are big differences
in the procedures to be followed during testing and commissioning this type of pro-
tection. When numerical protection is used, most of the wiring of a traditional relay
arrangement is substituted by the programmable logic, with the great advantage of
having multiple configurations and settings available as the numerical protection per-
mits it. Consequently, programmable logic should be included along with all settings
in the documentation for construction, commissioning, and testing purposes.
14.3.1 Setting the parameters
The method of setting numerical protection differs considerably from that used for
electromechanical relays. Setting the parameters of a numerical protection, also
known as parameterisation, involves selecting the functions of system configura-
tion, protection, control, communication, alarms, and reporting features. As with
conventional relays, settings for non-unit protections are determined on the basis of
short-circuit and co-ordination studies. For unit protection, settings do not depend on
co-ordination studies; therefore, machine constants, system parameters, and opera-
tion criteria are required in order to define the appropriate settings. Parameter settings
are achieved by carrying out the following operations:
• selecting the required power system configuration and parameters of general
operation;
• selecting the required protection functions;
• determining the setting groups;
• determining the dynamic or adaptive schemes;
• determining the relay logic and number of digital inputs and outputs;
• determining the configuration of reporting features and alarms.
Figure 14.5 represents the set-up system summary in a numerical protection.
The capability of manually or automatically changing the setting groups to meet
the needs of the system is one of the great advantages of using numerical protection.
Asetting group may be changed immediately when the systemtopology changes with
no compromise on the reliability of the system.
Installation, testing and maintenance of protection systems 293
Figure 14.5 Setup systemconfiguration of a numerical protection system(Beckwith
M-3425). (Reproduced by permission of Beckwith Electric Co.)
14.3.2 Performance tests
Numerical protection devices are usually not allowed to be field calibrated since
they are calibrated by specialised and certified process calibration equipment in the
manufacturer’s factory. Protective relay manufacturers perform numerous tests to
each device before they are dispatched to customers after they pass the tests.
Most numerical relays, if not all, can be accessed through the front panel (Human
Machine Interface – HMI) or through a serial port using a personal computer (PC).
By accessing the relay with a PC it is possible to easily perform steady-state and
dynamic tests. In some cases it is the only method to program logic schemes.
During the commissioning of numerical relays the following primary test methods
should be carried out in order to check relay performance and operation values within
the manufacturer’s specifications.
(i) Acceptance tests
The traditional acceptance tests are used to determine that the relay’s systemconfigu-
ration, protection, control, metering, communication, alarms, and reporting features
294 Protection of electricity distribution networks
are functional and that their response is in accordance with the manufacturer’s spec-
ifications. They are also used to check that the relay has been correctly installed. As
with traditional relays, acceptance tests include calibration or adjustment verification
for any parameter measured with secondary current and voltage injection. Current,
voltage, frequency, and phase angle relationship applied to the device under test
should be controlled and properly monitored. Nowadays, state-of-the-art protective
relay test systems are available with graphical test software programs, pre-configured
test routines, and the capability of personalising test routines that can be later stored
for testing similar relays. Most of these software programs do not require computer
programming skills to develop automatic relay tests.
A typical scope of acceptance routine tests for numerical protection is as follows:
• isolate all the relay inputs and outputs signals;
• verify the correct polarity and supply voltage value;
• verify that the wiring of all analogue and digital inputs and outputs is correct;
• electrical insulation test;
• functional test of all hardware components:
– binary input/output;
– LCD display, LEDs, and any key pad;
– communication interface;
– other hardware.
• setting all the parameters according to a well-documented setting report;
• tests of all enabled protection functions with secondary current and voltage injec-
tion as referred to before. Certain elements should be disabled or temporarily
modified while testing;
• verification of operating measured-values;
• tests of the programmable logic scheme(s);
• test of targets and output contacts;
• verification of reporting features and alarms;
• reload/verify the in-service settings into the relay.
(ii) Functional tests
After the acceptance tests have been performed and it has been verified that the relay
can meet the intended specifications, functional tests should be performed. The spe-
cific procedure for each relay depends on the type of protection involved and the logic
scheme implemented. Ageneral procedure for carrying out these tests for a numerical
protection arrangement is given below:
• verify external AC and DC input signals;
• verify external input contacts;
• verify tripping and signalling;
• verify remote/transferred tripping;
• verify interaction with the SCADA system, if applicable;
• verify setting group change according to the logic setting.
Installation, testing and maintenance of protection systems 295
Functional tests can be easily achieved if the advanced tools for metering and
reporting of the numerical protection are used. Some of these features are logic status
report, in-service readings, metering data, event report, and oscillographies. The two
last features are useful not only for commissioning, but also for troubleshooting.
(iii) Dynamic tests
In a dynamic test the value of the applied phasors that represent the power systemcon-
ditions should be properly adjusted when analysing pre-fault load, fault and post-fault
states. In this test there are no typical characteristics of a power system, such as high
frequency and DC decrement. Changes in phasor values (i.e. the magnitude and/or
angle) should be controlled in such manner that they do not present any significant
slip during the whole test.
(iv) Transient simulation tests
The test signal of the transient simulation represents, for the relay, the actual external
signals of frequency, magnitude, and duration when a disturbance occurs in the power
system. These signals may involve offset (DC displacement), the saturation effects
of CTs and the response of transient voltage surge suppressors.
(v) End-to-end tests
End-to-end testing methodology is appropriate to verify the communication scheme
and the protection system in an electrical line at the same time. Standard end-to-end
testing methods use the following resources on each side of the line:
• three-phase protection relay test equipment with appropriate communication and
transient simulation capabilities;
• for the time synchronisation of the two, test equipment used could be one of the
following methods: GPS-satellite receiver (currently the most common method),
pilot wire, fibre optic, or with the power system;
• fault simulation such as the Electromagnetic Transients Program (EMTP) or real
transient files taken from disturbance recorders.
The Electromagnetic Transients Program is a computer program for simulating
high speed transient effects in electric power systems. It includes a wide range of
modelling capabilities including oscillations ranging in duration from microseconds
to seconds.
Disturbance recorders generate oscillographic records in IEEE Standard
COMTRADE – Common Format for Transient Data Exchange – files. These files
permit a comprehensive analysis of a particular event to be achieved by using appro-
priate software and are fed into relay test equipment which can reproduce the original
waveform to be injected into a relay under test.
Bibliography
ANDERSON, P. M.: ‘Analysis of faulted power systems’ (The Iowa State University
Press, Ames, Iowa, USA, 1995)
ANDERSON, P. M., and MIRHEYDAR, M.: ‘An adaptive method of setting under-
frequency load shedding relays’, IEEE Trans., 1992, PWRS-7(2), pp. 647–655
ANSI/IEE standard C57.13: ‘Standard requirements for instrument transformers’,
1978
ANSI/IEEE standard 141: ‘Recommended practice for electric power distribution for
industrial plants’, 1990
ANSI/IEEE standard 242: ‘Recommended practice for protection and co-ordination
of industrial and commercial power systems’, 1986
ANSI/IEEE standard 399: ‘Recommended practice for power system analysis’, 1990
BLACKBURN, J. L.: ‘Protective relaying’ (Marcel Dekker, New York, 1998,
2nd edn.)
BURNETT, J.: ‘IDMT relay tripping of main incoming circuit breakers’, Power
Eng., J., 1990, 4, pp. 51–56
CLOSSON, J., and YOUNG, M.: ‘Commissioning numerical relays’, XIV IEEE
Summer Meeting, Acapulco, Mexico, July 2001
COOK, V.: ‘Analysis of distance protection’ (John Wiley & Sons, New York, 1985)
DAVIES, T.: ‘Protection of industrial power systems’ (Pergamon Press, Cleveland,
USA, 1983)
‘Developments in power system protection, 6th international conference’, IEE Conf.
Publ., 434, 1997
‘Developments in power system protection, 7th international conference’, IEE Conf.
Publ., 479, 2001
‘Distribution protection’, ABB buyer guide, vols. I, II and III, 1989–1990
‘Electrical distribution system protection manual’ (Cooper Power Systems, USA,
1990, 3rd edn.)
ELMORE, W. A.: ‘Protective relaying theory and applications’ (ABB, New York,
1994)
FEENAN, J.: ‘The versatility of high-voltage fuses in the protection of distribution
systems’, Power Eng. J., 1987, 1, pp. 109–115
298 Bibliography
GERS, J. M.: ‘Aplicación de protecciones eléctricas a sistemas de potencia’
(‘Application of electrical protection to power systems’), (Universidad de Valle,
Cali, 1993)
GERS, J. M.: ‘Enhancing numerical relaying performance with logic customization’,
XV IEEE Summer meeting, Acapulco, Mexico, July 2002
GROSS, C.A.: ‘Power system analysis’ (John Wiley & Sons, New York, 1986,
2nd edn.)
‘Guides and standards for protective relaying systems’ (IEEE, New York, 1995)
HARKER, K.: ‘Power system commissioning and maintenance practice’ (Peter
Peregrinus, London, 1998)
HEADLEY, A., BURDIS, E. P., and KELSEY, T.: ‘Application of protective devices
to radial overhead line networks’, IEE Proc. C, Gener. Transm. Distrib., 1986,
133, pp. 437–440
HORAK, J.: ‘Pitfalls and benefits of commissioning numerical relays’ (Neta World,
Colorado, 2003)
‘Instruction manual for generator protection system BE1-GPS100’, Basler Electric,
Publication 9 3187 00 990, Revision C, November 2001
‘Instruction manual for overcurrent BEI-951’, Basler Electric, Publication
9 3289 00 990, Revision F, August 2002
‘Instruction Book M-3311 Transformer Protection Relay’, Beckwith Electric Co.,
Publication 800-3311-IB-01MC1, February 2003
‘Instruction Book M-3425 Generator Protection Relay’, Beckwith Electric Co.,
Publication 800-3425-IB-02MC2, January 2003
LAKERVI, E., and HOLMES, E. J.: ‘Electricity distribution network design’ (Peter
Peregrinus, 1995, 2nd edn.; revised 2003)
LAYCOCK, W. J.: ‘Management of protection’, Power Eng. J., 1991, 5, pp. 201–207
LO, K. L., McDONALD, J. R., and YOUNG, D. J.: ‘Expert systems applied to alarm
processing in distribution control centres’, in Conference proceedings, UPEC 89,
Belfast, 1989
LO, K. L., and NASHID, L.: ‘Expert systems and their applications to power systems,
Part 3 Examples of application’, Power Eng. J., 1993, 7(5), pp. 209–213
LONG, W., COTCHER, D., RUIU, D., ADAM, P., LEE, S., and ADAPA, R.:
‘EMTP – a powerful tool for analysing power system transients’, Computer
applications in power, IEEE, 1990, 3(3), pp. 36–41
‘Microprocessor relays and protection systems’, IEEE Tutorial Course, 88EH0269-
1-PWR, 1987
MOORE, P. J., and JOHNS, A. T.: ‘Adaptive digital distance protection’, IEE Conf.
Publ., 1989, 302, pp. 187–191
PHADKE, A. G., and HOROWITZ, S. H.: ‘Adaptive relaying’, IEEE Comput. Appl.
Power, 1990, 3(3), pp. 47–51
PHADKE, A. G., and THORP, J. S.: ‘Computer relaying for power systems’ (John
Wiley & Sons, New York, 1988)
‘Power system protection’ (Peter Peregrinus, 1997)
‘Protection and control devices standards, dimensions and accessories’, Basler
Electric, Product Bulletin SDA-5 8-01, August 2001
Bibliography 299
‘Protective relays application guide’, GEC Measurements (Baldini and Mansell,
1987, 3rd edn.)
‘Quadramho static distance protection relay’, GEC Measurements, publication
R-5580B, London
‘Relay Testing Unit Sverker 750’, Programma Electric ABB, User’s Manual,
Ref. No. ZP-CD01E R01A, 1994
SIDHU, T. S., SACHDEV, M. S., and WOOD, H. C.: ‘Microprocessor-based relay
for protecting power transformers’, IEE Proc. C, Gener. Transm. Distrib., 1990,
137, pp. 436–444
STEVENSON, W. D.: ‘Elements of power systemanalysis’ (McGrawHill, NewYork,
1982, 4th edn.)
‘Switching, protection and distribution in low voltage networks’, Siemens, (Publicis
MCD Verlag, 1994, 2nd edn.)
TEO, C. Y., and CHAN, T. W.: ‘Development of computer-aided assessment for
distribution protection’, Power Eng. J., 1990, 4, pp. 21–27
URDANETA, A. J., RESTREPO, H., MARQUEZ, J., and SANCHEZ, J.:
‘Co-ordination of directional overcurrent relay timing using linear programming’,
IEEE PAS Winter meeting, New York, February 1995
WHITING, J. P., and LIDGATE, D.: ‘Computer prediction of IDMT relay settings
and performance for interconnected power systems’, IEE Proc. C., Gener. Transm.
Distrib., 1983, 130(3), pp. 139–147
WRIGHT, A., and NEWBERRY, P. G.: ‘Electric fuses’ (Peter Peregrinus, 1994,
2nd edn.)
WRIGHT, A.: ‘Application of fuses to power networks’, Power Eng. J., 1991, 4,
pp. 293–296, ibid. 1991, 5, pp. 129–134
YOUNG, D. J., LO, K. L., McDONALD, J. R., HOWARD, R., and RYE, J.: ‘Devel-
opment of a practical expert system for alarm processing’, IEE Proc. C, Gener.
Transm. Distrib., 1992, 139(5), pp. 437–447
ZHANG, Z. Z., and CHEN, D.: ‘An adaptive approach in digital distance protection’,
IEEE Trans., 1991, PWRD-6(1), pp. 135–142
Appendix
Solutions to exercises
To assist in the understanding of the solutions, the original diagrams included in the
Exercises have been redrawn, and where necessary, additional diagrams are included
to illustrate the calculations.
Solution to Exercise 1.1
Fault F
1
was correctly cleared by the tripping of breakers 2 and 5; therefore breakers
3 and 4 operated incorrectly and should be entered in column 3.
For fault F
2
, breakers 25 and 26 should have opened to clear the fault. These
are not shown in column 2; therefore they should be inserted in column 3 as having
mal-operated. Breakers 21 and 22 should have been tripped by back-up protection
and shown in column 5. Breakers 23 and 24 should not have tripped as the flow of
fault current was against their directional settings and these two breakers should be
placed in column 3 as having mal-operated. With the failure of breaker 25 to trip, the
fault was finally cleared by breaker 27 tripping on back-up protection, which should
be entered in column 5 also.
Fault F
3
should have been tripped by opening breakers 11 and 17. These are
shown as having operated in column 2. Therefore, the tripping of breakers 10 and 19
was unnecessary, and these two breakers should be placed in column 3.
The completed table is shown in Table A.1.
Solution to Exercise 5.1
The equivalent impedances of all the circuits are calculated, referred to the same volt-
age level, to obtain the diagram of the positive-sequence network and thus calculate
the short-circuit levels.
302 Protection of electricity distribution networks
Table A.1 Relay/breaker operation for Exercise 1.1
Case Breakers
that
operated
Breakers
that
mal-operated
Tripped by
primary
protection
Tripped by
back-up
protection
F
1
2, 3, 4, 5 3, 4 2, 5 –
F
2
21, 22, 23, 24, 27 25, 26, 23, 24 – 21, 22, 27
F
3
10, 11, 17, 19 10, 19 11, 17 –
10MVA 10MVA 10MVA
8
5
.
3
5
Ω
(
1
1
5
k
V
)
1
C
2
B
Yy0
3
115/34.5kV
58.45MVA
Z%=4.8
1400MVASC
A
4
Figure A.1 Schematic diagram for Exercise 5.1
Solutions to exercises 303
9.45Ω
Z
source
Z
transf
Z
line
3
115kV
10.86 Ω
A B
85.35 Ω
C
Figure A.2 Equivalent circuit of the system shown in Exercise 5.1
Z
source
=
V
2
P
sc
=

115×10
3

2
1400×10
6
=9.45, referred to 115 kV
Z
transf
=Z
pu(transf )
×Z
base(transf )
=0.048×

115×10
3

2
58.45×10
6
=10.86, referred to 115 kV
Z
line
=85.35, referred to 115 kV
The equivalent circuit is shown in Figure A.2.
Calculation of short-circuit currents
Busbar A:
I
sc(A)
=
115×10
3

3×Z
source
=
115×10
3

3×9.45
=7025.96A, referred to 115 kV
Busbar B:
I
sc(B)
=
115×10
3

3×(Z
source
+Z
transf
)
=
115×10
3

3×(9.45+10.86)
=3269.09A, referred to 115 kV
=3269.09×
115×10
3
34.5×10
3
=10896.97A, referred to 34.5 kV
Busbar C:
I
sc(C)
=
115×10
3

3×(Z
source
+Z
transf
+Z
line
)
=
115×10
3

3×(9.45+10.86+85.35)
=628.39A, referred to 115 kV
=628.39×
115×10
3
34.5×10
3
=2094.62A, referred to 34.5 kV
304 Protection of electricity distribution networks
Calculation of nominal currents
Relay 1: I
nom1
=
P
nom1

3×V
1
=
10×10
6
VA

3×34.5×10
3
V
=167.35A, referred to 115 kV
Relay 2: I
nom2
=3×I
nom1
=502.04A, referred to 34.5 kV
Relay 3: I
nom3
=
P
nom3

3×V
3
=
58.45×10
6
VA

3×34.5×10
3
V
=978.15A, referred to 34.5 kV
Relay 4: I
nom4
=
P
nom4

3×V
4
=
58.45×10
6
VA

3×115×10
3
V
=293.44A, referred to 115 kV
Selection of CT transformation ratios
The CT ratio is chosen from the larger of the two following values:
(i) the nominal current.
(ii) the maximum short-circuit current without saturation occurring (0.05×I
sc
).
The values of the CT ratios calculated in accordance with the abovementioned
criteria are given in Table A.2.
Setting of instantaneous units
The instantaneous units have settings from 6 to 144 A in steps of 1 A. The setting of
the instantaneous units on the feeders are based on 0.5×I
sc
at busbar C.
Relay 1:
I
inst
=0.5×I
sc(C)
=0.5×2094.62=1047.31 primary amps
=1047.31×(5/200) =26.18 secondary amps.
Set at 27 secondary amps, equivalent to 1080 primary amps.
Table A.2 Summary of calculations for CTratios for Exercise 5.1
Relay number P
nom
I
s
0.05×I
sc
I
nom
CT ratio
(MVA) (A) (A) (A)
1 10.00 2094.62 104.78 167.35 200/5
2 30.00 10896.97 544.85 502.04 600/5
3 58.45 10896.97 544.85 978.15 1000/5
4 58.45 7025.96 351.30 293.44 400/5
Solutions to exercises 305
Relay 2:
The setting is made by taking 125 per cent of the value of the current for the maximum
fault level which exists at the following substation.
I
inst
=1.25×I
sc(C)
=1.25×2094.62=2618.28 primary amps
=2618.28×(5/600) =21.82 secondary amps.
Set at 22 secondary amps, equivalent to 2640 primary amps.
Relay 3:
The instantaneous unit is overridden in order to avoid lack of co-ordination with the
transformer.
Relay 4:
The setting is made by taking 125 per cent of the value of the short-circuit current
existing on the 34.5 kV side, referred to the 115 kV side.
I
inst
=1.25×I
sc(B)
, referred to 115 kV, =1.25×3269.09
=4086.36 primary amps
=4086.36×(5/400) =51.08 secondary amps.
Set at 52 secondary amps, equivalent to 4160 primary amps.
Selection of pick-up settings
From eqn. (5.5), pick-up setting (PU) =OLF×I
nom
×(1/CTR).
Using an overload factor of 1.5, the pick-up settings are:
Relay 1: PU
1
=5×167.35×(5/200) =6.28; set at 7.
Relay 2: PU
2
=1.5×502.04×(5/600) =6.28; set at 7.
Relay 3: PU
3
=1.5×978.15×(5/1000) =7.34; set at 8.
Relay 4: PU
4
=1.5×293.44×(5/400) =5.5; set at 6.
Time dial settings
Relay 1:
The minimum time dial setting is chosen for this relay, since it is at the end of the
circuit and therefore does not need to be co-ordinated with any other protective device.
Use a time dial setting of 1/2.
The operating time of relay 1 should be calculated from just before the operation
of its instantaneous unit. Based on the primary setting of 1080 Aand a pick-up setting
of 7, the plug setting multiplier (PSM) is
PSM=I
inst.prim.1
×(1/CTR
1
) ×(1/PU
1
) =1080×(5/200) ×(1/7)
=3.86 times.
With a time dial setting of 1/2, and PSM=3.86 times, from the relay curves,
t
1
=0.16s.
Relay 2:
Calculate the back-up time to relay 1; t
2
=0.16+0.4=0.56s.
306 Protection of electricity distribution networks
Calculate the multiplier on the basis of the 1080 primary amps for the CT associated
with relay 1.
PSM
2a
=1080×(1/CTR
2
) ×(1/PU
2
) =1080×(5/600) ×(1/7) =1.29 times.
With PSM
2a
=1.29 times, and t
2a
=0.56s, from the relay curves, the time dial
setting=1/2.
The operating time of relay 2 should be calculated from just before the operation of
its instantaneous unit:
PSM
2b
=I
inst.prim.2
×(1/CTR
2
) ×(1/PU
2
) =2640×(5/600) ×(1/7)
=3.14 times.
With a time dial setting of 1/2, and PSM
2b
=3.14 times, from the relay curves,
t
2b
=0.25s.
Relay 3:
Calculate the back-up time to relay 2; t
3a
=0.25+0.4=0.65s.
Calculate the multiplier on the basis of the 2640 primary amps for the CT associated
with relay 2.
PSM
3a
=2640×(1/CTR
3
) ×(1/PU
3
) =2640×(5/1000) ×(1/8)
=1.65 times.
With PSM
3a
=1.65 times, and t
3a
=0.65s, from the relay curves, time dial
setting=1/2.
As the instantaneous unit is overriddenthe multiplier is obtainedusingthe short-circuit
current:
PSM
3b
=I
sc.3
×(1/CTR
3
) ×(1/TAP
3
) =10896.97×(5/1000) ×(1/8)
=6.81 times.
With a time dial setting of 1/2, and PSM=6.81 times, fromthe relay curves, t
3
=0.07
seconds.
Relay 4:
Calculate the back-up time to relay 3; t
4
=0.07+0.4=0.47s.
Calculate the multiplier on the basis of the 10896.97 primary amps for the CT
associated with relay 3, but referred to 115 kV:
PSM
4
=10896.97×(34.5/115) ×(1/CTR
4
) ×(1/PU
4
)
=10806.97×(34.5/115) ×(5/400) ×(1/6) =6.81 times.
With PSM
4
=6.81 times, and t
4
=0.47s, from the relay curves, time dial setting=4.
The settings are summarised in Table A.3.
Solutions to exercises 307
Table A.3 Summary of relay settings for Exercise 5.1
Relay CT ratio Pick-up Time Instantaneous units
number (A) dial primary (A) secondary (A)
1 200/5 7 1/2 1080 27
2 600/5 7 1/2 2640 overridden
3 1000/5 8 1/2 overridden overridden
4 400/5 8 4 4160 52
Solution to Exercise 5.2
Figure A.3 gives the single line diagram of the substation.
Calculation of short-circuit currents
13.2 kV busbar:
From Figure A.3, which repeats Figure 5.28, it can be noted that the short-circuit
level at the 13.2 kVbusbar is 21900 amperes, and fromthis the equivalent impedance
of the system behind the busbar can be obtained:
I
sc(13.2kV)
=21900A, referred to 13.2 kV
P
sc
=

3×13200V×21900A=500.70125MVA
Z
source
=
V
2
P
sc
=

13.2×10
3

2
500.70125×10
6
=0.348 referred to 13.2 kV
=0.348×

115×10
3
13.2×10
3

2
=26.414 referred to 115 kV
115 kV busbar:
In order to obtain this short-circuit fault level it is necessary to know the equivalent
impedance behind this busbar:
Z

source
+Z
transf
=Z
source
where Z

source
is the equivalent impedance behind the 115 kV busbar and Z
transf
is
the equivalent impedance of the two transformers.
Z
transf 1
=0.0963×

115×10
3

2
20×10
6
=63.6784 referred to 115 kV
Z
transf 3
=0.101×

115×10
3

2
41.7×10
6
=32.0318 referred to 115 kV
The equivalent impedance of the twotransformers is Z
transf
=63.6784||32.0318=
21.3116 referred to 115 kV. Thus, Z

source
+Z
transf
=26.414 referred to 115 kV.
Z

source
=26.414−21.3116, =5.1024 referred to 115 kV.
308 Protection of electricity distribution networks
2
4
k
A
2
1
.
9
k
A
Fault
115/13.2 kV 115/13.2 kV
1
6
.
0
9
k
A
1
4
.
7
k
A
2000/5A
Z%=101
41.7 MVA
13.2 kV Busbar
(5 MVA)
Typical
feeder
CO-11
50–600/5A
W W
CO-11
3 3
CO-11 CO-11
800/5A
CO-11
W
Z%=9.63
20 MVA
CO-11
W
1
.
6
k
A
MITSU
T3
300/5A
0
CO-11 CO-11
ASGEN
T1
150/5A
W
CO-11
115 kV Busbar
3
50
51
50
51
3
1 1
3
50N
51N
50
51
50N
51N
50N
51N
51
50N
51N
51
1
0
.
8
2
k
A
0
7
.
2
1
k
A
7
.
9
1
k
A
Figure A.3 Single-line diagram of substation for Exercise 5.2
Solutions to exercises 309
With this value of equivalent impedance the short-circuit level at the 115 kV
busbar can be obtained:
I
sc(115kV)
=
115×10
3

3×5.1024
=13012.56A referred to 115 kV
Calculation of nominal currents
Relay on the high voltage side of transformer T1:
I
nom(HV.T1)
=
P
nom.trans1

3×V
=
20×10
6
VA

3×115×10
3
V
=100.41A referred to 115 kV
Relay on the low voltage side of transformer T1:
I
nom(LV.T1)
=
P
nom.trans1

3×V
=
20×10
6
VA

3×13.2×10
3
V
=874.77A referred to 13.2 kV
Relay on the high voltage side of transformer T3:
I
nom(HV.T3)
=
P
nom.trans3

3×V
=
41.7×10
6
VA

3×115×10
3
V
=209.35A referred to 115 kV
Relay on the low voltage side of transformer T3:
I
nom(LV.T3)
=
P
nom.trans3

3×V
=
41.7×10
6
VA

3×13.2×10
3
V
=1823.9A referred to 13.2 kV
Relay on a typical feeder:
I
nom(feeder)
=
P
nom.feeder

3×V
=
5×10
6
VA

3×13.2×10
3
V
=218.69A referred to 13.2 kV
Setting of instantaneous units
Relay on typical feeder:
The setting of the instantaneous units on relays associated with feeders are calculated
using the mid value of the short circuit current seen by the relay.
I
inst
=0.5×I
sc(13.2kV bus)
=0.5×21900 primary amps =10950A
Relay on the high voltage side of transformer T1:
I
inst
=1.25×I
sc(13.2kV bus)
, referredto115kV, =1.25×7210×(13.2/115) =1034.48
primary amps, ⇒1034.48×(5/150) =34.8 secondary amps. Set relay to 35 sec-
ondary amps, equivalent to 1050 primary amps.
Relay on the low voltage side of transformer T1:
The instantaneous element is overridden to avoid the possibility of loss of discrimi-
nation with the transformer secondary protection.
310 Protection of electricity distribution networks
Relay on the high voltage side of transformer T3:
The setting is based on 125 per cent of the short circuit current that exists at the
13.2 kV busbar seen by the relay in question, referred to the 115 kV side.
I
inst
=1.25×I
sc(13.2kV)
, referred to 115kV, =1.25×14700×(13.2/115) =
2109.13 primary amps, ⇒2109.13×5/300=35.15 secondary amps. Set at 36
secondary amps, equivalent to 2160 primary amps.
Relay on the low voltage side of transformer T3:
The instantaneous element is overridden to avoid the possibility of loss of discrimi-
nation with the transformer secondary protection.
Selection of pick-up settings
Relay on typical feeder:
PU
feeder
=1.5×218.69×(5/300) =5.47; PU
feeder
set at 6.
Relay on high voltage side of transformer T1:
PU
HV.T1
=1.5×100.41×(5/150) =5.02; PU
HV.T1
set at 6.
Relay on low voltage side of transformer T1:
PU
LV.T1
=1.5×874.77×(5/900) =7.28; PU
LV.T1
set at 8.
Relay on high voltage side of transformer T3:
PU
HV.T3
=1.5×209.35×(5/200) =5.23; PU
HV.T3
set at 6.
Relay on low voltage side of transformer T3:
PU
LV.T3
=1.5×1823.9×(5/2000) =6.84; PU
LV.T3
set at 7.
Time dial settings
Relay on typical feeder:
Use the minimumtime dial setting, in order to deal with relay at the end of the circuit;
time dial setting=1/2.
It is necessary to calculate the relay operating time, just before the instantaneous unit
operates:
PSM
feeder
=I
inst.prim
×(1/CTR) ×(1/PU
feeder
) =10950×(5/300) ×(1/6)
=30 times.
With time dial setting of 1/2, and PSM
feeder
=30, from the relay operating curve
t
feeder
=0.024s.
Relay on low voltage side of transformer T1:
The back-up time over the feeder relay is now calculated; t
LV.T1
=0.024+0.4=
0.424s. The multiplier for this relay is based on the 10950 primary amps for the CT
associated with the feeder:
PSM
LV.T1
=10950×(1/CTR
LV.T1
) ×(1/PU
LV.T1
) =10950×(5/900) ×(1/8)
=7.6 times.
Solutions to exercises 311
With PSM
LV.T1
=7.6 and the relay back-up time =0.424s, from the relay curves,
time dial setting=4.
As the instantaneous unit is overridden, the short-circuit current seen by the relay
and multiplied by 0.86 is used for the delta-star transformer.
PSM=0.86×I
sc.LV.T1
×(1/CTR
LV.T1
) ×(1/PU
LV.T1
)
=0.86×7210×(5/900) ×(1/8) =4.3 times.
With a time dial setting of 4, and PSM=4.3, from the relay curve, t
HV.T1
=1.4s.
Relay on the high voltage side of transformer T1:
Required operating time, as back up to the low voltage side relay, =1.4+0.4=1.8s.
The multiplier is calculated on the basis of the 7210 primary amps (referred to 115 kV)
which is seen by the CT associated with the relay on the low voltage side of T1:
PSM
HV.T1
=7210×(13.2kV/115kV) ×(1/CTR
HV.T1
) ×(1/PU
HV.T1
)
=7210×(13.2kV/115kV) ×(5/150) ×(1/6) =4.6 times
With PSM
HV.T1
=4.6, and the back-up time =1.8s, from the relay curve, time dial
setting=6.
Relay on low voltage side of transformer T3:
The back-up time over the feeder relay is 0.024+0.4=0.424s.
The multiplier is calculated on the basis of the 10950 primary amps which is seen by
the CT associated with the relay on the low voltage side of T1:
PSM
LV.T3
=10950×(1/CTR
LV.T3
) ×(1/PU
LV.T3
) =10950×(5/2000) ×(1/7)
=3.9 times.
With PSM
LV.T3
=3.9, and the back-up time =0.424s, from the relay curve time
dial setting=2.
As the instantaneous unit is overridden, the short-circuit current seen by the
relay on the low voltage side of T3 and multiplied by 0.86 is used for the delta-star
transformer.
PSM=0.86×I
sc.LV.T3
×(1/CTR
LV.T3
) ×(1/PU
LV.T3
)
=0.86×14700×(5/2000) ×(1/7) =4.5 times.
With a time dial setting of 2, and PSM=4.5, from the relay curve, t
LV.T3
=0.6s.
Relay on the high voltage side of transformer T3:
Required operating time, as back up to the low voltage side relay, =0.6+0.4=1.0s.
The multiplier is calculated on the basis of the 14700 amps (referred to 115 kV) which
is seen by the CT associated with the relay on the low voltage side of T3:
PSM
HV.T3
=14700×(13.2kV/115kV) ×(1/CTR
HV.T3
) ×(1/PU
HV.T3
)
=14700×(13.2kV/115kV) ×(5/300) ×(1/6) =4.7 times
With PSM
HV.T3
=4.7, and the back-up time =1.0s, from the relay curve, time dial
setting=4.
312 Protection of electricity distribution networks
Solution to Exercise 5.3
The single diagram for the network being studied is shown in Figure A.4.
Maximum short-circuit currents
The equivalent impedances of all the circuits are calculated, referred to the same
voltage level, in order to produce the diagram of positive-sequence impedances for
calculating the short-circuit levels.
Z
source
=
V
2
P
sc
=
(115000)
2
2570.87
=5.144 referred to 115 kV,
=5.144×

34.5
115

2
=0.463 referred to 34.5 kV
Z
transf1
=0.117×
(34500)
2
V
10.5×10
6
VA
=13.26 referred to 34.5 kV
Z
transf2
=0.06×
(34500)
2
V
5.25×10
6
VA
=13.6 referred to 34.5 kV
Z
lineBC
=0.625/km×14.2km=8.875 at 34.5kV
The equivalent circuit is shown in Figure A.5.
Calculation of short-circuit currents
Busbar A:
I
sc(A)
=
34.5×10
3

3(0.463+13.26+8.875+13.6)
=881.43A referred to 34.5 kV
=550.26×(34.5/13.2) =1438.18A referred to 13.2 kV
Busbar B:
I
sc(B)
=
34.5×10
3

3(0.463+13.26+8.875)
=881.43A referred to 34.5 kV
Busbar C:
I
sc(C)
=
34.5×10
3

3(0.463+13.26)
=1451.47A referred to 34.5 kV
Busbar D:
I
sc(D)
=12906.89A referred to 115 kV
Solutions to exercises 313
2.625 MVA 2.625 MVA
1 2
34.5 kV
6 (CT 100/5)
34.5 kV
5.25 MVA
Z%=6.0
3
13.2 kV
A
34.5/13.2 kV
Dy1
4
B
14.2 km
j0.625Ω/km
5
C
115/34.5 kV
Yy0
7
10.5 MVA
Z%=11.7
2570.87 MVASC
8
D
Figure A.4 Single-line diagram for Exercise 5.3
314 Protection of electricity distribution networks
0.463 Ω
Z
source
Z
T1
Z
line
3
34.5kV
13.26Ω
Z
T2
13.6 Ω
D C
8.875 Ω
B A
Figure A.5 Equivalent circuit of the system shown in Exercise 5.3
Calculation of maximum peak values
Circuit breaker 1:
I
peak
=2.55×I
rms.sym.bkr1
=2.55×1438.14A=3667.36 peak amperes
Circuit breaker 5:
I
peak
=2.55×I
rms.sym.bkr5
=2.55×1451.47A=3701.25 peak amperes
Circuit breaker 8:
I
peak
=2.55×I
rms.sym.bkr8
=2.55×12906.89A=32912.57 peak amperes
Calculation of r.m.s. asymmetrical values
I
rms.asym
=I
rms.sym.int

2e
−2(R/L)t
+1
where t =5 cycles =83.33ms, and L/R=0.2, so that R/L=5.
Breaker 1: I
rms.asym.
=1438.18×

2e
−2×0.08333×5

+1
=1438.18×1.3672=1966.27A
Breaker 5: I
rms.asym.
=1451.47×

2e
−2×0.08333×5

+1
=1451.47×1.3672=1984.45A
Breaker 8: I
rms.asym.
=12906.89×

2e
−2×0.08333×5

+1
=12906.89×1.3672=17646.3A
Calculation of CT turns ratios
The CT ratio is obtained using the higher value of either the nominal current, or the
maximum short circuit current for which no saturation is present (0.05×I
sc
).
Relays 1 & 2: I
nom1
=I
nom2
=
P
nom1

3V
1
=
2.625×10
6
VA

3×13.2×10
3
V
=114.81A referred to 13.2 kV
Solutions to exercises 315
Table A.4 Summary of CT ratio calculations for Exercise 5.3
Relay number P
nom
I
s
0.05×I
sc
I
nom
CT ratio
(MVA) (A) (A) (A)
1 and 2 2.625 1438.18 71.91 114.81 150/5
3 5.25 1438.18 71.91 229.63 250/5
4 5.25 881.43 44.07 87.86 100/5
5 5.25 1451.47 72.57 87.86 100/5
6 – 1451.47 72.57 – 100/5
7 10.5 1451.47 72.57 175.72 200/5
8 10.5 12906.89 645.34 52.71 700/5
Relay 3: I
nom3
=
P
nom3

3V
3
=
5.25×10
6
VA

3×13.2×10
3
V
=229.63A referred to 13.2 kV
Relay 4: I
nom4
=
P
nom4

3V
4
=
5.25×10
6
VA

3×34.5×10
3
V
=87.86A referred to 34.5 kV
Relay 5: I
nom5
=I
nom4
=87.86A referred to 34.5 kV
Relay 7: I
nom7
=
P
nom7

3V
7
=
10.5×10
6
VA

3×34.5×10
3
V
=175.72A referred to 34.5 kV
Relay 8: I
nom8
=
P
nom8

3V
8
=
10.5×10
6
VA

3×115×10
3
V
=52.71A referred to 115 kV
The short-circuit and load currents plus the selected CT ratio in line with the above-
mentioned criteria are summarised in Table A.4.
Instantaneous, pick-up and time dial settings
Calculation of instantaneous settings
Relays 1 and 2:
I
inst
=0.5×I
sc
=0.5×1438.18=719.09 primary amps.
Secondary amps =719.09×5/150=23.97A.
Set relay to 24 secondary A, equivalent to 720 primary amps.
Relay 3: The instantaneous element is overridden to avoid the possibility of loss of
discrimination with the transformer secondary protection.
Relay 4: The setting is based on 125 per cent of the short-circuit current which exists
at the busbar on the lower voltage side of the transformer, referred to the higher
voltage side.
I
inst
=1.25×I
sc
, referred to 34.5kV=1.25×1438.18×(13.2/34.5) =687.83 pri-
mary amps, ⇒687.83×5/100=34.39 secondary amps. Set at 35 secondary amps,
equivalent to 700 primary amps.
316 Protection of electricity distribution networks
Relay 5: The setting is calculated on the basis of 125 per cent of the current for the
maximum fault level which exists at the next substation.
I
inst
=1.25×I
sc(B)
=1.25×881.43=1101.79 primary amps, ⇒1101.79×5/100=
55.09 secondary amps. Set at 56 secondary amps, equivalent to 1120 primary amps.
Relay 6: I
inst
=1000 primary amps =50 secondary amps.
Relay 7: The instantaneous element is overridden to avoid the possibility of loss of
discrimination with the transformer secondary protection.
Relay 8: The setting is calculated on the basis of 125 per cent of the current for the
maximum fault level which exists at the next substation.
I
inst
=1.25×I
sc(C)
, referred to the higher voltage side, =1.25×1451.47×
(34.5/115) =554.3 primary amps, ⇒554.3×5/700=3.88 secondary amps.
Set at 6 secondary amps (the minimum setting), equivalent to 840 primary amps.
Calculation of pick-up settings
Pick-up setting=OLF×I
nom
×(1/CTR)
With an overload factor of 1.5:
Relays 1 & 2: PU
1,2
=1.5×114.81×(5/150) =5.74; set at 6
Relay 3: PU
3
=1.5×229.63×(5/250) =6.89; set at 7
Relay 4: PU
4
=1.5×87.86×(5/100) =6.59; set at 7
Relay 5: PU
5
=1.5×87.86×(5/100) =6.59; set at 7
Relay 6: PU
6
=7
Relay 7: PU
7
=1.5×175.72×(5/200) =6.59; set at 7
Relay 8: PU
8
=1.5×52.71×(5/700) =056; set at 1
Calculation of time dial settings
For the relays associated with breakers 1 and 2, choose the smaller dial setting, and
calculate the operating time for the greater value of current that causes operation of
the instantaneous unit.
Relays 1 and 2:
Time dial setting=1/2. PSM=I
inst.sec
×(1/PU
1,2
) =24×(1/6) =4 times
With PSM of 4, and time dial setting of 1/2, from the operating characteristic of the
relay, t
1
=0.16s.
Relay 3:
The relay backs up relays 1 and 2; therefore the required operating time t
3
is
0.16+0.4=0.56s.
PSM
3
is based on the 720 primary amps associated with relays 1 and 2, so that:
PSM
3a
=720×(1/CTR
3
) ×(1/PU
3
) =720×(5/250) ×(1/7) =2.06 times.
With PSM
3a
=2.06, and a back-up operating time of 0.56 s, from the characteristic
curve of a typical relay, time dial setting=1/2.
With this dial setting, relay 3 in fact acts as back-up to relay 1 and 2 in a time of
0.7 s. Since, in this case, the instantaneous unit is overridden the short-circuit current
Solutions to exercises 317
is used and multiplied by the factor 0.86 in order to cover the delta-star transformer
arrangement.
Thus, PSM
3b
=0.86×I
sc(A)
×(1/CTR
3
)×(1/PU
3
)=0.86×1438.18×(5/250) ×
(1/7) =3.53 times.
With a time dial setting of 1/2, and PSM
3b
=3.53, the operating time of the relay,
from the characteristic curve, is t
3
=0.6s.
Relay 4:
The operating time is based on this relay acting as back up to relay 3, i.e. t
3
=
0.2+0.4=0.6s.
PSM
4a
is based on the 1438.18 primary amps of the CT associated with relay 3,
so that:
PSM
4a
=1438.18×(13.2kV/34.5kV) ×(5/100) ×(1/7) =3.93 times.
With PSM
4a
=3.93, and the back-up time of the relay=0.6s, this gives a time dial
setting of 2. With this dial setting, relay 4 actually backs up relay 3 in a time of 0.78 s.
It is now necessary to calculate the operating time of relay 4 just before the
operation of the instantaneous unit:
PSM
4b
=I
inst.prim.4
×(1/CTR
4
) ×(1/PU
4
) =700×(5/100) ×(1/7) =5 times.
With a time dial setting of 2, and PSM=5, t
4
=0.49s.
Relay 5:
The required operating time is calculated on backing up relay 4, i.e. t
5
=0.49+0.4=
0.89s. PSM
5
is based on the 700 primary amps of the CT associated with relay 4, so
that, PSM
5
=700×(1/CTR
5
) ×(1/PU
5
) =700×(5/100) ×(1/7) =5 times. With
PSM
5
=5 and a back-up time of 0.89 s, from the relay operating curve the time dial
setting=4.
With this dial setting, relay 5 will back up relay 4 in a time of one second.
It is now necessary to calculate the operating time of relay 5 just before the
operation of the instantaneous unit:
PSM
5a
=I
inst.prim.5
×(1/CTR
5
) ×(1/PU
5
) =1120×(5/100) ×(1/7) =8 times.
With a time dial setting of 4 and PSM=8, t
5
=0.45s.
Relay 6: time dial setting=5
Relay 7:
Relay 7 backs up the operation of relays 5 and 6 and should be co-ordinated with the
slower of these two relays.
Relay 6 has an instantaneous unit of 1000 primary amps, smaller than that for relay 5
which is set at 1120 amperes. Therefore, the operating time of both relays should be
calculated for this value of current.
Relay 6: PSM
6
=1000×(5/100) ×(1/7) =7.14 times
With PSM
6
=7.14 times and time dial setting=5, t
6
=0.7s.
Relay 5: PSM
7
=1000×(5/100) ×(1/7) =7.14 times
With PSM
5
=7.14 times and a time dial setting=4, t
5
=0.52s.
318 Protection of electricity distribution networks
Table A.5 Summary of settings for Exercise 5.3
Relay
number
CT ratio Pick-up
(A)
Time
dial
Instantaneous
primary (A) secondary (A)
1 and 2 150/5 6 1/2 720 24
3 250/5 7 1/2 overridden overridden
4 100/5 7 2 700 35
5 100/5 7 4 1120 56
6 100/5 7 5 1000 50
7 200/5 7 3 overridden overridden
8 700/5 1 2 840 6
Therefore, in order to be slower than relay 6, the back-up time should be t
7
=0.7+
0.4=1.1s.
It is now necessary to calculate the PSM that represents, in relay 7, the primary
current of 1000 amps in the CT associated with relay 6, the relay with which it needs
to be co-ordinated.
PSM
7
=1000×(1/CTR
7
) ×(1/PU
7
) =1000×(5/200) ×(1/7) =3.57 times.
With t
7
=1.1s and a time dial setting of 3.57, from the relay operating curve,
time dial setting=3.
As the instantaneous unit is overridden, the multiplier is calculated using the
current for a short-circuit on busbar C:
PSM
7
=I
sc7
×(1/CTR
7
) ×(1/PU
7
) =1451.47×(5/200) ×(1/7) =5.18times.
With PSM
7
=5.18 times, and a time dial setting=3, t
7
=0.7s.
Relay 8:
The back-up time of relay 8 over relay 7 is given by t
8
=0.7+0.4=1.1s.
Now calculate the multiplier that represents in relay 8 the 1451.47 primary amps
referred to 115 kV:
MULT
8
=1451.47×(34.5kV/115kV) ×(5/700) ×1=3.11 times.
With MULT
8
=3.11 times, and t
8
=1.1 seconds, from the relay curve DIAL
8
=2.
Coverage of protection
% cover =
K
s
(1−K
i
) +1
K
i
,
where
K
i
=
I
sc.pickup
I
sc.end
=
I
inst.prim.5
I
sc(B)
=
1120 primary A
881.43A
=1.2707
Solutions to exercises 319
and
K
s
=
Z
source
+Z
T1
Z
line
=
0.463+13.26
8.875
=1.5463
% cover =
1.5463(1−1.2707) +1
1.2707
=0.4576
Therefore the cover by the instantaneous unit of the overcurrent relay associated
with breaker 5 is 45.7 per cent of the 34.5 kV line.
Solution to Exercise 7.1
Figure A.6 shows the network for Exercise 7.1.
Calculation of short-circuit levels
Busbar A: I
sc
=6560A
Busbar B:
(a) With the ring closed the equivalent network is as shown in Figure A.7.
I
sc(B)
=
13.2×10
3

3×(1.16+1.556)
=2806.39A
Taking account of the divisionof current inthe network, the fault current that circulates
through breaker 1 is 2806.39×(7/9) =2182.75A, and the current that circulates
through breaker 2 is 2806.39×(2/9) =623.64A.
(b) With the ring open.
With breaker 1 open, I
sc(B)
=
13.2×10
3

3×(1.16+7)
=933.95A=I
sc.max
for relay 3
5 4
6 Ω
1
B
3
3MVA 2MVA
2 Ω
2
C
6
A
1 Ω
150 MVA
SC
Z
source
=1.16Ω
Figure A.6 Network for Exercise 7.1
320 Protection of electricity distribution networks
kV
3
13.2
Z
source
2 Ω
7 Ω
B
1.16 Ω
Figure A.7 Equivalent network for fault on busbar B
kV
3
13.2
Z
source
8 Ω
1 Ω
A
1.16 Ω
Figure A.8 Equivalent network for fault on busbar C
With breaker 2 open, I
sc(B)
=
13.2×10
3

3×(1.16+2)
=2411.72A=I
sc.max
for relay 4
Busbar C:
(a) With the ring closed the equivalent network is as shown in Figure A.8.
I
sc(C)
=
13.2×10
3

3×(1.16+0.889)
=3719.39A
Taking account of the division of current in the network, the fault current that circu-
lates through breaker 1 is 3719.39×(1/9) =413.27A, and the current that circulates
through breaker 2 is 3719.39×(8/9) =3306.12A.
(b) With the ring open.
With breaker 1 open, I
sc(C)
=
13.2×10
3

3×(1.16+1)
=3528.25A=I
sc.max
for relay 5
Solutions to exercises 321
With breaker 2 open, I
sc(C)
=
13.2×10
3

3×(1.16+8)
=831.99A=I
sc.max
for relay 6
Calculation of CT ratios
First, calculate the maximum nominal currents for each breaker.
Breaker 1 open:
I
max,2&6
=
5×10
6
VA

3×(13.2×10
3
V)
=218.69A.
I
max,4&5
=
3×10
6
VA

3×(13.2×10
3
V)
=131.22A.
I
max,1&3
=0
Breaker 2 open:
I
max,1&3
=
5×10
6
VA

3×(13.2×10
3
V)
=218.69A.
I
max,4&5
=
2×10
6
VA

3×(13.2×10
3
V)
=87.48A.
I
max,2&6
=0
The CTs are selected, starting from the maximum nominal currents for each breaker,
and checking the saturation of maximum fault current at each breaker at 0.05×I
sc
:
Breaker 1 : I
max 1
=218.69⇒CT
1
ratio=250/5
Breaker 4 : I
max 4
=131.22⇒CT
4
ratio=200/5
Breaker 6 : I
max 6
=218.69⇒CT
6
ratio=250/5
Setting of instantaneous units
In order to set relays 1, 4 and 6, the ring is opened at breaker 2 and the setting for
each one of the relays is calculated using the maximum short-circuit current that is
seen by the relay which is backing up.
Relay 1: this relay backs up relay 4.
I
inst
=1.25×I
sc.4
=1.25×2411.72A=3014.65 primary amps, i.e. 3014.65×
(5/250) =60.29 secondary amps. Set to 61 secondary amps, equivalent to 3050
primary amps.
Relay 4: this relay backs up relay 6.
I
inst
=1.25×I
sc.6
=1.25×831.99A=1040 primary amps, i.e. 1040×(5/200) =
26 secondary amps. Set to 26 secondary amps, equivalent to 1040 primary amps.
Relay 6:
Set the relay taking 125 per cent of the maximum load current, which comes
from the source: I
inst
=1.25×I
max .6
=1.25×218.69A=273.36 primary amps,
322 Protection of electricity distribution networks
i.e. 273.36×(5/250) =5.47 secondary amps. Set to 6 secondary amps, equivalent to
300 primary amps.
Calculation of pick-up settings
Pick-up setting=OLF×I
nom
×(1/CTR)
Taking an overload factor of 1.5
Breaker 2 open:
Relay 1: PU
1
=1.5×I
nom1
×(1/CTR
1
) =1.5×218.69×(5/250) =6.56;
set PU
1
=7
Relay 4: PU
4
=1.5×I
nom4
×(1/CTR
4
) =1.5×131.22×(5/200) =4.92;
set PU
4
=5
Relay 6: PU
6
=1.5×I
nom6
×(1/CTR
6
) =1.5×218.69×(5/250) =6.56;
set PU
6
=7
Calculation of time dial settings
Here the IEC expression, given in eqn. 5.7, Section 5.3.3 in Chapter 5, is used when
considering very inverse time relays.
Relay 6:
The setting is the lowest time dial setting possible, and for this type of relay this is a
setting of 0.1.
PSM
6
=831.99×(1/CTR
6
) ×(1/PU
6
) =831.99×(5/250) ×(1/7) =2.38.
With a time dial setting of 0.1 and PSM
6
=2.38, using the expression for a very
inverse relay where α =1 and β =13.5:
t
6
=
0.1×13.5
2.38−1
=0.98s
Relay 4:
The back-up time of relay 4 over relay 6, t
4
=0.98+0.4=1.38s.
The multiplier that represents the 831.99 A of relay 6 is calculated for relay 4:
PSM
4
=831.99×(1/CTR
4
) ×(1/PU
4
) =831.99×(5/200) ×(1/5) =4.16.
With t
4
=1.38s and PSM
4
=4.16, using the expression for a very inverse relay where
α =13.5 and β =1:
t
4
⇒1.38=
time dial setting×13.5
4.16−1
⇒time dial setting=0.32.
The time dial setting chosen is 0.4.
Next calculate the multiplier that represents I
sc.max
for relay 4 (2411.72 A):
PSM
4a
=2411.72×(1/CTR
4
) ×(1/PU
4
)=2411.72×(5/200) ×(1/5) =12.06.
With a time dial setting=0.4 and PSM
4a
=12.06, using the expression for a very
inverse relay where α =1 and β =13.5:
t
4
=
time dial setting×13.5
PSM
4a
−1
=
0.4×13.5
12.06−1
=0.49s.
Solutions to exercises 323
Relay 1:
The back-up time of relay 1 over relay 4, t
1
, =0.49+0.4=0.89s. The multiplier that
represents the 2411.72 A of relay 4, at relay 1, is calculated as:
PSM
1
=2411.72×(1/CTR
1
) ×(1/PU
1
) =2411.72×(5/250) ×(1/7) =6.89.
With t
1
=0.89s and PSM
1
=6.89, using the expression for a very inverse relay:
t
1
=
time dial setting×13.5
PSM
1
−1
⇒0.89
=
time dial setting×13.5
6.89−1
⇒time dial setting=0.39s.
The time dial setting chosen is 0.4.
Solution to Exercise 7.2
The network is shown in Figure A.9.
Calculation of short-circuit currents
Busbar B:
With the ring closed, I
sc(B)
=2160A, and with the ring open I
sc(B)
=2880A, the
largest short-circuit current for relay B.
Busbar C:
With the ring closed, I
sc(C)
=1350A, and with the ring open I
sc(C)
=2300A, the
largest short-circuit current for relay C.
F
3
=
2
3
0
0
F
4
=
2
1
6
0
B
F
4
F
5
F
1
=
1
6
9
0
F
2
=
1
3
5
0
F
2
A
F
3
115kV
F
6
=
4
4
4
0
F
5
=
2
8
8
0
F
6
C
34.5kV
F
1
Indicates the location of the
opening of the line
Figure A.9 Network for Exercise 7.2
324 Protection of electricity distribution networks
Calculation of maximum nominal currents
These are calculated with the ring open at the far end of line C-A, and with the ring
open at breaker A, in order to determine the larger value of current.
Ring open at the far end of line C-A:
I
nom.max.(A)
=
25×10
6
VA

3×(13.2×10
3
V)
=1093.47A
I
nom.max.(B)
=
15×10
6
VA

3×(13.2×10
3
V)
=656.1A
I
nom.max(C)
=0
Ring open at breaker A:
I
nom.max.(C)
=
25×10
6
VA

3×(13.2×10
3
V)
=1093.47A
I
nom.max.(B)
=
10×10
6
VA

3×(13.2×10
3
V)
=437.39A
I
nom.max(A)
=0
The CTs are selected, starting from the maximum nominal currents for each breaker,
and checking the saturation of maximum fault current at each breaker at 0.05×I
sc
:
Breaker A : I
nom.max(A)
=1093.47⇒CT ratio=1100/5
Breaker B : I
nom.max(B)
=656.10⇒CT ratio=700/5
Breaker C : I
nom.max(C)
=1093.47⇒CT ratio=1100/5
Setting of instantaneous units
In order to set relays A, B and C, the ring is opened at the far end of line C-A, and
the setting for each of the relays is made with the maximum short-circuit current that
is seen by the back-up relay.
Relay A: This relay backs up relay B
I
inst
=1.25×I
sc(B)
=1.25×2880amps =3600 primary amps ⇒3600×(5/1100) =
16.36 secondary amps.
The relay is set at 17 secondary amps i.e. 3300 primary amps.
Relay B: This relay backs up relay C
I
inst
=1.25×I
sc(C)
=1.25×2300amps =2875 primary amps ⇒2875×(5/700) =
20.54 secondary amps.
The relay is set at 21 secondary amps i.e. 2940 primary amps.
Relay C:
Set the relay taking 125 per cent of the maximum load current, which comes from the
source: I
inst
=1.25×I
nom.max(C)
=1.25×1093.47amps =1366.84 primary amps ⇒
1366.84×(5/1100) =6.21 secondary amps.
Solutions to exercises 325
The relay is set at 7 secondary amps i.e. 1540 primary amps.
The pick-up of each relay is 5.
Time dial setting
The settings of relays A, B and C are determined with the ring open at the far end of
line C-A.
Relay C:
This relay is set using the smallest dial setting, i.e. 1/2.
PSM
C
=I
sc.max(C)
×(1/CTR
C
) ×(1/PU
C
) =2300×(5/1100) ×(1/8) =1.3
With a time dial setting=1/2 and PSM
C
=1.3, from the relay curves, t
C
=2s.
Relay B:
The back-up time of relay B over relay C, t
B
, =2+0.4=2.4s.
The multiplier that represents the 2300 A of relay C is calculated for relay B:
PSM
B1
=2300×(1/CTR
B
) ×(1/PU
B
) =2300×(5/700) ×(1/8) =2.1
With t
B
=2.4s, and PSM
B1
=2.1, from the relay curves, time dial setting=2.
Next calculate the multiplier which represents I
sc.max
for relay B (2880 A):
PSM
B2
=I
sc.max(B)
×(1/CTR
B
) ×(1/PU
B
) =2880×(5/700) ×(1/8) =2.6
With time dial setting=2, and PSM
B2
=2.6, from the relay curves, t
B
=1.8s.
Relay A:
The back-up time of relay A over relay B, t
A
, =1.8+0.4=2.2s.
The multiplier that represents the 2880 A of relay B is calculated for relay A:
PSM
A
=I
sc.max(A)
×(1/CTR
A
) ×(1/PU
A
) =2880×(5/1100) ×(1/8) =1.7
With t
A
=2.2s and PSM
A
=1.7, from the relay curves, time dial setting=1.
Solution to Exercise 8.1
The diagram for this exercise is shown in Figure A.10.
1000/5
12.4/69 kV
16/20 MVA
200/5
Figure A.10 Diagram for Exercise 8.1
326 Protection of electricity distribution networks
The nominal currents for maximum load on the transformer are calculated as
I
nom(69kV)
=
20×10
6
MVA

3×69×10
3
V
=167.35A
and
I
nom(12.4kV)
=
20×10
6
MVA

3×12.4×10
3
V
=931.21A
The currents in the relays are
I
relay(12.4kV)
=931.21×(5/1000) ×

3=8.06A; PU=8.7 is therefore chosen.
I
relay(69kV)
=167.35×(5/200) ×

3=4.18A; PU=4.2 is therefore chosen.
Solution to Exercise 8.2
The diagram for this exercise is shown in Figure A.11.
The nominal currents for maximum load on the transformer are calculated as
I
nom(161kV)
=
40×10
6
MVA

3×161×10
3
V
=143.44A
I
nom(69kV)
=
40×10
6
MVA

3×69×10
3
V
=334.7A
I
nom(12.4kV)
=
10×10
6
MVA

3×12.4×10
3
V
=456.61A
and the currents in the relays are therefore
I
relay(161kV)
=143.44×(5/200) ×

3=6.2A; PU of 6.2 is chosen.
I
relay(69kV)
=334.7×(5/400) =4.18A; PU of 4.2 is chosen.
I
relay(12.4kV)
=456.61×(5/500) =4.65A; PU of 4.6 is chosen.
161 kV
30/40 MVA
69 kV
30/40 MVA
500/5
200/5
12.4 kV
7.5/10 MVA
400/5
Figure A.11 Diagram for Exercise 8.2
Solutions to exercises 327
Solution to Exercise 8.3
The diagram for this exercise is shown on Figure A.12.
The nominal currents under normal conditions are
I
nom.34.5kV
=
10×10
6
VA

3×34.5×10
3
V
=167.35A
I
nom.13.2kV
=
10×10
6
VA

3×13.2×10
3
V
=437.39A
The vector diagram for the Dy1 connection is shown in Figure A.13, taking into
account that the rotation of the phases A, B and C is negative, i.e. clockwise.
The diagram showing the flow of currents in the primary and secondary windings
of the transformer, and in the CT secondaries is given in Figure A.14.
200/5
4500
4500
Dy1
10MVA
34.5/13.2 kV I
b
I
c
4.5 kA
900/5
R R
OP
R R
OP
R R
OP
Figure A.12 Diagram for Exercise 8.3
B
b
a
c
C
C
A
B
A
c
b
a
Figure A.13 Vector group for Dy1 transformers
328 Protection of electricity distribution networks
D
y
1
A B C
1
6
3
.
3
5
0
°
Ω
9
4
.
3
1
3
0
°
9
4
.
3
1
1
5
0
°
9
4
.
3
1
2
7
0
°
2
0
0
/
5
0
.
0
3
0
.
0
3
0
.
0
3
9
0
0
/
5
2
.
4
3
R
R
O
P
R
R
O
P
R
R
O
P
1
6
3
.
3
5
1
2
0
°
Ω
1
6
3
.
3
5
2
4
0
°
Ω
4
.
0
8
0
°
4
.
0
8
1
2
0
°
4
.
0
8
2
4
0
°
4
.
0
8
0
°
Ω
4
.
0
8
1
2
0
°
Ω
4
.
0
8
2
4
0
°
Ω
4
.
2
1
0
°
Ω
4
.
2
1
1
2
0
°
Ω
4
.
2
1
2
4
0
°
Ω
4
3
7
.
3
9
3
0
°
Ω
4
3
7
.
3
9
1
5
0
°
Ω
4
3
7
.
3
9
2
7
0
°
Ω
3
0
°
2
.
4
3
1
5
0
°
2
.
4
3
2
7
0
°
F
i
g
u
r
e
A
.
1
4
D
i
a
g
r
a
m
f
o
r
n
o
r
m
a
l
c
o
n
d
i
t
i
o
n
s
(
E
x
e
r
c
i
s
e
8
.
3
)
Solutions to exercises 329
For a fault between phases B and C, the current that is induced in the delta is
I
Delta
=I
fault
×
N
2
N
1
,
where
N
2
N
1
=
V
2

3×V
1
Therefore
I
Delta
=I
fault
×
V
2

3×V
1
=4500×
13.2

3×34.5
=994.05A
The line currents are equal to those in the delta for two of the phases and double the
current for the phase that has the larger current.
The schematic diagram in Figure A.15 shows the fault between phases B and C
and the different magnitudes and directions of the currents in the conductors and those
that circulate due to the fault. Starting from the figure it is necessary to determine
if the differential relays would operate, taking into account whether any unbalance
exists in the currents through the operating coils.
As will be seen from Figure A.15, only a small amount of current flows through
the operating coils of the differential relays, because there is no large unbalance of
currents due to the fault, and therefore the differential relays do not operate for a fault
between conductors B and C even though the phase rotation is negative.
Solution to Exercise 8.4
Figure A.16 shows the three-phase primary and secondary connections for
Exercise 8.4, while Figure A.17 provides the single-line diagram.
The maximum load currents of the transformer are given by
I
nom(69kV)
=
20×10
6
VA

3×69×10
3
V
=167.35A
I
nom(12.4kV)
=
20×10
6
VA

3×12.4×10
3
V
=931.21A
and the currents in the relays are
I
69kV
=167.35×(5/600) =1.39A
I
12.4kV
=931.21×(5/1200) ×(1/a) ×(1/

3)
Since the currents that enter the relay from each side must balance, the transfor-
mation ratio of the compensation transformer is obtained from 931.21×(5/1200) ×
(1/a) ×(1/

3) =1.39A, from which a =1.61.
330 Protection of electricity distribution networks
D
y
1
A B C
9
9
4
.
0
5
A
9
9
4
.
0
5
A
9
9
4
.
0
5
A
1
9
9
8
.
1
A
9
9
4
.
0
5
A
2
4
.
8
5
A
2
5
A
0
A
4
5
0
0
A
4
5
0
0
4
5
0
0
A
2
5
A
2
5
A
5
0
A
2
5
A
4
9
.
7
A
2
4
.
8
5
A
2
0
0
/
5
0
.
3
0
.
1
5
0
.
1
5
9
0
0
/
5
5
0
A
2
5
A
R
R
O
P
R
R
O
P
R
R
O
P
2
4
.
8
5
A
4
9
.
7
A
2
4
.
8
5
A
F
i
g
u
r
e
A
.
1
5
D
i
a
g
r
a
m
f
o
r
f
a
u
l
t
b
e
t
w
e
e
n
p
h
a
s
e
s
B
a
n
d
C
(
E
x
e
r
c
i
s
e
8
.
3
)
Solutions to exercises 331
a : 1
I
A
I
B
I
C
Y
600/5
Y
1200/5
Yd7
69/12.4kV
Figure A.16 Diagram for Exercise 8.4
87
600/5
Compensation
transformer
a : 1
16/20 MVA
69/12.4 kV
1200/5
Figure A.17 Single-line diagram for Exercise 8.4
The phasor diagram for the Yd1 connection is given in Figure A.18, taking into
account that the rotation of the phases A, B and C is positive.
Fromthe phase diagramfor a Yd7 connection, the three-phase connection diagram
in Figure A.19 is obtained, showing the connection of the CTs, and the magnitude
and direction of the currents in the primary and secondary connections.
332 Protection of electricity distribution networks
a
C
b
c
B
A
B
C
b
a
c
A
Figure A.18 Vector group for Yd7 transformers
1200/5
C
B
I
a
Yd7
a : 1
1.39 0°
I
a
1.39 0°
I
b
1.39 240°
I
b
1.39 240°
I
c
1.39 120°
2.24 0°
2.24 240°
2.24 120°
I
a
3.88 –210°
3.88 –210°
I
b
3.88 30°
3.88 30°
I
c
3.88 –90°
3.88 –90°
I
c
1.39 120°
I
a
1.39 0°
I
b
1.39 240°
I
c
1.39 120°
I
A
167.35 0°
537.63 0°
931.21 –210°
931.21 30°
931.21 –90°
I
B
167.35 240°
537.63 240°
I
C
167.35 120°
537.63 120°
600/5
A
69/12.4 kV
Yd7
Figure A.19 Diagram for normal conditions for Exercise 8.4
Solution to Exercise 9.1
The characteristic of a mho relay in a reactance diagramis a circle that passes through
the origin of the axis of the co-ordinates, and with a centre at the point (a, b), as in
Figure A.20.
Solutions to exercises 333
(a, b)
r
R
X
Figure A.20 Mho characteristics in X-R diagram (Exercise 9.1)
In order to obtain the characteristic of the mho relay in the admittance diagram it
is necessary to create the characteristic of the relay as a function of these values:
Z=R+jX
Y =1/(R+jX)
Multiplying both sides by R−jX gives
Y =(R−jX)/(R
2
+X
2
) =G+jB
Therefore
G=R/(R
2
+X
2
)
and
B =−X/(R
2
+X
2
)
In accordance with the equation for a circle:
(R−a)
2
+(X−b)
2
=r
2
R
2
−2aR+a
2
+X
2
−2bX+b
2
=r
2
where r
2
=a
2
+b
2
Thus:
R
2
−2aR+X
2
−2bX=0
Dividing the above expression by (R
2
+X
2
) gives
R
2
+X
2
R
2
+X
2

2aR
R
2
+X
2

2bX
R
2
+X
2
=0
334 Protection of electricity distribution networks
B
G
1
2b

1
2a
Figure A.21 Mho characteristics in B-G diagram (Exercise 9.1)
and
1−
2aR
R
2
+X
2

2bX
R
2
+X
2
=0
Replacing G and B in the expression gives:
1−2aG+2bB =0
B =
aG
b

1
2b
If B =0, then G=1/2a; and if G=0, then B =−1/2b, so that the characteristic of
the mho relay in the admittance diagram will be as shown in Figure A.21.
Solution to Exercise 9.2
The diagram for exercise 9.2 is shown in Figure A.22.
For a single-phase fault the three sequence networks are connected in series as shown
in Figure A.23.
For a single-phase fault it can be shown that
I
a1
=I
a2
=I
a0
⇒I
a1
=I
a
/3
I
a
=3I
a1
=
3(11.8×10
3
)

3{2×(j0.23+2.3+j5.7) +45+j7.5+(3.5+j24) +3R
F
}
I
a
=
(20.44×10
3
)
(53.1+3R
F
+j43.46)
If R
T
is taken as equal to 53.1+3R
F
, then R
T
+j43.36=(20.44×10
3
)/(I
F

−θ
1
).
If I
F
=200 amps, then R
T
+j43.36=(20.44×10
3
)/(200

−θ
1
), and
R
T
=

102.2
2
+43.36
2
=92.25.
Solutions to exercises 335
B
Z
T1
=Z
T0
=0.23Ω
Earthing transformer
Z
0
=j2.5Ω
15 Ω
800/1
R
Z
L1
=2.3 +j5.7 Ω
Z
L0
=3.5+j24 Ω
75 MVA
145/11.8 kV
A
Phase to
earth fault
Fault current =200A
Figure A.22 Diagram for Exercise 9.2
j43.36
3R+53.1
102.2
Zero sequence network
Negative sequence network
3 ×15 =45 Ω
A
3 ×2.5 =j7.5 3.5Ω j24
Positive sequence network
3
j0.23
A
j0.23
11.8kV
A
2.3 Ω j5.7
2.3 Ω j5.7
B
F
3R
F
F
B
F
B
Figure A.23 Sequence networks for Exercise 9.2
336 Protection of electricity distribution networks
Therefore the fault resistance, R
F
, =(92.55−53.1)/3=13.15.
The residual compensation constant, K, is given by K=
Z
L0
−Z
L1
3×Z
L1
K=(3.5+j24−2.3−j5.7)/3(2.3+j5.7) =(1.2+j18.3)/(6.9+j17.1) =0.99
Calculation of the secondary impedance that the relay sees if it is used with a
residual compensation equal to 1.0 (100 per cent):
Z
R
=
V
Ra
I
Ra
=
I
0
×(Z
L1
+Z
L2
+3R
F
+Z
L0
)
3I
0
+KI
R
=
2Z
L1
+3R
F
+Z
L0
3(1+K)
,
and K=1.
Since I
R
=3I
0
for this fault, then
Z
R
=
2(2.3+j5.7) +(3×13.15) +(3.5+j24)
6
=7.93+j5.9
The secondary ohms are
primary ohms ×
CT ratio
VT ratio
=(7.93+j5.9) ×
800/1
11800/110
=59.14+j44
Solution to Exercise 9.3
The system arrangements are shown in Figure A.24 for reference.
The operating zones for the distance relay located in the Juanchito substation on the
Pance line are
Zone 1: protects 85 per cent of the Juanchito-Pance line.
Zone 2: protects 100 per cent of the Juanchito-Pance plus 50 per cent of the shortest
adjacent line, which in this case is the Pance-Yumbo line.
Zone 3: protects 100 per cent of the Juanchito-Pance line plus 100 per cent of the
longest adjacent line, which is the Pance-Alto Anchicayá line, plus 25 per cent of the
shortest remote line i.e. the Alto Anchicayá-Yumbo line.
As it is required to calculate the reach, in secondary ohms, of zone 3, then the
infeed constants for both the longest adjacent line and the shortest remote line need
to be calculated, using the impedances given in Figure A.24.
Z
3
=Z
protected line
+(1+K
2
)(Z
longest adjacent line
)
+0.25(1+K
3
)(Z
shortest remote line
),
where
Z
protected line
=11.40

83.48

(Juanchito-Pance line)
Z
longest adjacent line
=27.64

82.45

(Pance-Alto Anchicayá line)
and
Z
shortest remote line
=27.84

82.45

(Alto Anchicayá-Yumbo line)
Solutions to exercises 337
13.72 82.42° Ω
Pance 115
21
11.40 83.48° Ω
Juanchito
220
9
J11.61
*
10
25.05 83.51° Ω
27.64 82.45° Ω
Pance 220
7
27.84 82.45° Ω
A. Anchicayá
Salvajina
Yumbo 220
7
7
* Equivalent of two
autotransformers
7
7
220
Yumbo
Figure A.24 Power system for Exercise 9.3
Thus:
K
2
=
I
not seen by relay
I
relay
=
I
2−7
+I
10−7
+I
8−7
+0.5(I
9−7
)
0.5(I
9−7
)
From the printout in Figure 9.53:
K
2
=
1278

−87.96

+1210.67

−87.6

+2446.77

−86.18

+0.5(1814.2

−86.59

)
0.5(1814.2

−86.59

)
=6.44

−0.34

The calculation of the second infeed constant for the coverage of the shortest
remote line, K
3
, is similar to that for the first infeed constant for the longest adja-
cent line K
2
, (calculated above), but it is now necessary to take into account the
contribution of the generation at Alto Anchicayá. This contribution does not affect
the coverage of the Zone 3 relay on the Pance-Alto Anchicayá line, but does on the
remote Alto Anchicayá-Yumbo circuit. Therefore
K
3
=
I
not seen by relay
I
relay
=
I
2−7
+I
10−7
+I
8−7
+0.5(I
9−7
) +I
contribution from gen at AA
0.5(I
9−7
)
The current contributions fromthe Alto Anchicayá substation that should be included
in the above formula are not available for this exercise. However, K
3
can be assumed
to be equal to K
2
, so that the reach of Z
3
will be reduced by a small amount, which
is not critical since this zone acts as a remote back-up as mentioned in Section 9.3.
Z
3
=11.4

83.5

+
¸
(1+6.44

−0.34

) ×27.64

82.45

¸
+
¸
0.25(1+6.44

−0.34

) ×27.84

82.45

¸
=268.82

82.21

primary ohms
The value of Z
3
in secondary ohms is
Z
3
=(primary ohms) ×(CT ratio)/(VT ratio)
=268.82

82.21

×(800/5) ×(1/2000) =21.5 secondary ohms
338 Protection of electricity distribution networks
It is necessary to make a check for the proximity of maximum load in order to
verify that the maximum load impedance is never inside the characteristic of zone 3:
Z
x
=
0.55×Z
3
×sin α
sin(φ −30

)
where
β =sin
−1
{0.818×sin(φ −30

)}
α =210

−β −φ
The angle of the setting of the relay is φ =75

; therefore β =35.34

, and α =
210

−35.34

−75

=99.66

.
Using these values in the equation for Z
x
:
Z
x
=
0.55×268.82×sin 99.66

sin(75

−30

)
=206.13 primary ohms.
The maximum load impedance is
Z
c
=
V
2
P
=
(220×10
3
)
2
40×10
6
=1210 primary ohms
The check consists of verifying that
Z
c
−Z
x
Z
c
≥0.5
For the double circuit line:
Z
c
−Z
x
Z
c
=
1210−206.13
1210
=0.829
which is greater than 0.5. Therefore the setting of the relay can be considered
appropriate, and it is not necessary to reduce its reach.
Index
adaptive protection 39, 102
alarms processing 269, 274
expert systems 271–2
fingerprinting approach 273–4
hypothesis approach 275
processing methods 270
rules 273
amplitude comparator 174–5
distance protection 173
asymmetric current 13
asymmetrical fault calculations 23
back-up protection 3, 5, 7, 8–9, 117, 186,
252, 301
blocking characteristic
impedance protection 198–9
busbar differential protection 168, 266
cabling lists 268
compensation transformer 154, 156–8,
172, 329
computerised protection
adaptive 39
architectures 40
co-ordination 19, 68, 71, 73, 85–6, 88–9,
94, 96–9, 102, 109, 111, 113–14,
117, 120–3, 125, 128, 137, 142, 144,
186–7, 219, 225, 227, 230, 236–7,
242, 252, 286, 292, 305,
fuse-fuse 117
overcurrent relays 86, 88–9, 109
Dy transformers 75, 86
fuses 96
recloser-fuse 117
recloser-recloser 120
electronic 111, 120, 122
hydraulic 111, 120, 122
recloser-relay 122
recloser-sectionaliser 123
time/current devices 117
coverage 190, 194, 212, 318, 337
instantaneous overcurrent units 68–9
transformer differential protection 152
current transformers 5, 26, 51, 60, 72, 92,
149, 170, 226–7, 265, 285, 289
a.c. saturation 52
accuracy classes 58
burden 53
d.c. saturation 59
equivalent circuit 51–2, 60
errors 52, 72, 150
excitation curve 52–4, 56
flux density 56
installation tests 283
knee point 53–4
magnetisation curve 51, 53–4, 56
saturation density 56
saturation point 53
secondary injection test 53, 285–6
selection 53, 92, 154, 304
tests 58
diagrams
a.c. connections 262, 264–5
d.c. connections 262, 265
logic 262, 268
single line 28, 60, 89, 104, 154, 243,
262–3
substation design 261–8
substation layout 263–4
wiring 262, 266, 285
340 Index
differential protection
busbar 152, 168
high impedance 169–70
multiple restraint 168
compensation transformer 154, 156–7,
172
line 164
operating characteristics 164
percentage biased 150
rotating machines 152, 164
transformer 149, 152
coverage 152
magnetisation inrush 153
sensitivity 153, 155, 162
variable percentage characteristic 150,
152, 168
variable tappings 150
directional earth fault relays 131–2
directional overcurrent relays 127, 137,
141, 261
connections 138
instantaneous units 137
co-ordination 137
time delay units 141
pick-up setting 141
time dial setting 146
discrimination margin 71–2, 88, 98–9,
104, 235
distance protection 173–224
amplitude comparator 174–5
arc resistance, effect 179–80, 182, 193
blocking characteristic 198
directional relay 179
earth fault units 196
effect of infeeds 188
effective cover 199–200, 208
generator excitation loss detection 219
impedance relay 179
infeed constant 188–9, 199
intertripping 212
permissive over reach 214
permissive under reach 213
under reach 213
lens characteristics 183
maximum load impedance 200
mho relays 200
polygonal relays 202–3
mho relay 181
polarised 182
reach 185
overlap 186, 188–9, 192, 207, 210
over reach 189, 214–15, 217–18
phase angle comparator 176–7
phase units 195
polygonal relay 184, 203
power swing blocking unit 199
power system oscillations 196–7
reactance relay 180
fault resistance 180, 182–3, 193
relay characteristic angle 193, 201
residual compensation 194, 207, 222,
336
series compensated lines 214
setting 185
tee circuits 215
under reach 213, 217–18
equivalent impedances 26, 301, 312
excitation curves 52, 56
current transformer 51, 53, 55, 59
fault calculations 11, 16, 23, 28
computer 28
interactive program 28
fault currents 7, 11, 23, 59, 139
asymmetrical values 23
mathematical derivation 24
frequency relays 239
settings 240, 242–3, 247, 249
frequency variation 240–1, 243, 250
fuse-fuse co-ordination 117
fuse-links 115, 118
fuses 109, 114–17
classification 115
maximum clearance time 114
minimum melting time 114, 117–18,
123
nominal current 116, 123
nominal voltage 114, 116
notation 117
speed ratio 115
generator protection schemes 251
impedance protection
blocking characteristic 198–9
impedance relays 168, 179, 252
impedances, equivalent 26
industrial plant 225–6
load shedding 239–50
automatic system 240
industrial system protection 225–37
low voltage breakers 227, 236
motors 226
moulded case circuit breakers 225
Index 341
overcurrent relays 225–6
thermal relays 226
thermal relay-contactor and fuse 225
inertia constant 239–41
infeed constant
distance protection 188–9
intertripping
distance protection 212, 214
load shedding 239
automatic system 240–2
frequency relays 239
setting 240, 242–3, 247, 249
frequency variation 240–1, 243, 250
modelling parameters 241
voltage variation 242
low voltage breakers 227, 236
maintenance criteria 291
mho relay 181–2, 201
motor
locked rotor current 226, 230
service factor 230
motor protection 252
low voltage breakers 226, 253
thermal relays 226
moulded case circuit breakers 225
negative sequence relays
co-ordination 96
numerical protection 39–42
digital logic 39, 41, 99
logic schemes 100
principles 99
performance tests 293
setting parameters 291
operator ‘a’ 263
overcurrent relays 63
ANSI/IEEE constants 72, 73
co-ordination requirements 73
minimum short-circuit level 73
pick-up values 75
thermal limits 73
definite current 63
definite time 66
definite time/current 66
IEC constants 72, 73
instantaneous units 67
coverage 68
distribution circuits 68
selectivity 65–68, 72, 88
transformers 63, 67–8, 71, 73
inverse time 66–7, 79
negative-sequence units 96
setting 66
software techniques 98
time/current curves 66
IEC/UK 73–4
IEEE/US 75
typical 74
time delay units 97, 141
co-ordination criteria 117
pick-up setting 141
time dial setting 142
voltage control units 97
periodic maintenance 290
pilot differential protection 164, 166
polygonal relay 183, 202
power system oscillations 196, 198
distance protection 193
pre-commissioning tests 185
primary injection test 189
protection
computerised 39
adaptive 39
architectures 42
differential 149–70
busbar 168
distance 173–222
generator 251
industrial systems 225–7
line 261
motor 252
overcurrent 63
selectivity 63
transformer 258
protection equipment 283
installation 283
testing 283
factory 285
pre-commissioning 285
primary injection 289
secondary injection 286
protection zones 5
reactances
synchronous machines 22
recloser-fuse co-ordination 117
recloser-recloser co-ordination 120
recloser-relay co-ordination 122
recloser-sectionaliser co-ordination 123
342 Index
reclosers
co-ordination 117–23
installing 113
operating sequence 111
relay re-set time 122
relay circuit supplies 43
relays
attracted armature 33
co-ordination requirements 73
computerised 33
cup 35
directional overcurrent 127
identification code 252
impedance 179
induction 35
mho 181
moveable coil 34
overcurrent 63, 66, 70, 97–8, 127
polygonal 184, 202–3
reactance 180
shaded pole 36
solenoid 33
thermal 32, 226–7, 252
wattmetric 36–7
residual compensation
distance protection 194, 207, 222, 336
secondary injection test 285–6
sectionalisers 109, 113
selectivity 5, 67, 88, 96, 237
sequence
components 19, 26, 28
currents 22
impedances 22–3, 26, 28
networks 22–3, 28
voltages 27, 32
series compensated lines
distance protection 214
software techniques 98
symmetrical components 19, 22–3, 26
synchronous machines 14, 22
reactance values 16, 22
tests
factory 285
pre-commissioning 285
primary injection 289
secondary injection 285–6
thermal limits 73
thermal relays 226
thermal relay-contactors 225
transformers
differential protection 152–64
inrush current 68
overcurrent protection 186
protection schemes 154, 258
thermal capacity 74
voltage transformers
capacitor type 47
equivalent circuit 45
inductive 47
burden 47
equivalent circuit 45
error limits 47, 49
errors 46
flux density 45
magnetic impedance 51

Contents

Preface and acknowledgements Preface to 2nd edition 1 Introduction 1.1 General 1.2 Basic principles of electrical systems 1.3 Protection requirements 1.4 Protection zones 1.5 Primary and back-up protection 1.5.1 Primary protection 1.5.2 Back-up protection 1.6 Directional protection 1.7 Exercise 1.1 Calculation of short-circuit currents 2.1 Modelling for short-circuit current calculations 2.1.1 Effect of the system impedance 2.1.2 Effect of rotating machinery 2.1.3 Types of fault duty 2.1.4 Calculation of fault duty values 2.2 Methods for calculating short-circuit currents 2.2.1 Importance and construction of sequence networks 2.2.2 Calculation of asymmetrical faults using symmetrical components 2.2.3 Equivalent impedances for a power system 2.3 Supplying the current and voltage signals to protection systems 2.4 Calculation of faults by computer Classification and function of relays 3.1 Classification 3.1.1 Construction

xi xiii 1 1 3 3 5 5 5 5 7 9 11 11 11 13 14 16 19 22 23 26 26 28 31 31 31

2

3

vi

Contents 3.1.2 Incoming signal 3.1.3 Function 3.1.4 International identification of electrical devices Electromechanical relays 3.2.1 Attraction relays 3.2.2 Relays with moveable coils 3.2.3 Induction relays Evolution of protection relays Numerical protection 3.4.1 General 3.4.2 Characteristics of numerical relays 3.4.3 Typical architectures of numerical relays 3.4.4 Standard functions of numerical relays Supplies to the relay circuits 31 32 32 33 33 34 35 38 39 39 39 40 41 43 45 45 45 46 47 47 47 51 51 52 52 53 53 59 59 63 63 63 63 66 66 66 67 68 70

3.2

3.3 3.4

3.5 4

Current and voltage transformers 4.1 Voltage transformers 4.1.1 Equivalent circuit 4.1.2 Errors 4.1.3 Burden 4.1.4 Selection of VTs 4.1.5 Capacitor voltage transformers 4.2 Current transformers 4.2.1 Equivalent circuit 4.2.2 Errors 4.2.3 AC saturation 4.2.4 Burden 4.2.5 Selection of CTs 4.2.6 DC saturation 4.2.7 Precautions when working with CTs Overcurrent protection 5.1 General 5.2 Types of overcurrent relay 5.2.1 Definite-current relays 5.2.2 Definite-time/current or definite-time relays 5.2.3 Inverse-time relays 5.3 Setting overcurrent relays 5.3.1 Setting instantaneous units 5.3.2 Coverage of instantaneous units protecting lines between substations 5.3.3 Setting the parameters of time delay overcurrent relays

5

3 90◦ connection (30◦ AMT) 7.2.2 Sectionalisers 6.1 Construction 7.3 Logic schemes Adaptive protection with group settings change Exercises vii 73 73 73 75 86 96 96 97 98 99 99 99 100 102 104 109 109 109 113 114 117 117 117 120 122 123 123 127 127 128 128 129 129 130 131 131 137 141 141 142 146 5.5 Co-ordination of instantaneous units 7.1.1 Equipment 6.4 90◦ connection (45◦ AMT) 7.9 5.8 5.3.1 Reclosers 6.2.2 Principles of digital logic 5.5 Recloser-sectionaliser co-ordination 6.2 Time dial setting 7.11 5.2 Recloser-fuse co-ordination 6.6.3.2.4.3.7 5.4 Directional earth-fault relays 7.2. reclosers and sectionalisers 6.5 5.10.6.1 General 5.6 5.2 Thermal limits 5.3 Recloser-recloser co-ordination 6.7 Exercises 7 .3 Relay connections 7.3 Pick-up values Co-ordination across Dy transformers Co-ordination with fuses Co-ordination of negative-sequence units Overcurrent relays with voltage control Setting overcurrent relays using software techniques Use of digital logic in numerical relaying 5.2.1.3.1.2 Criteria for co-ordination of time/current devices in distribution systems 6.Contents 5.3 Fuses 6.1 Fuse-fuse co-ordination 6.2 60◦ connection (0◦ AMT) 7.2.1 Minimum short-circuit levels 5.4 Recloser-relay co-ordination 6.6 Setting of time-delay directional overcurrent units 7.6 Recloser-sectionaliser-fuse co-ordination Directional overcurrent relays 7.1 Pick-up setting 7.4.12 6 Fuses.10 5.10.1 30◦ connection (0◦ AMT) 7.2 Principle of operation 7.4 Constraints of relay co-ordination 5.10.4.

5 Completely polarised mho relay 9.6 Busbar differential protection 8.8 Relays with combined characteristics 9.3.2.1 Impedance relay 9.2.4 Mho relay 9.5 The effect of arc resistance on distance protection 9.2 Directional relay 9.2 Earth-fault units 9.2.7 Exercises Distance protection 9.7.13 Distance relays on series-compensated lines 149 149 152 152 153 154 159 161 162 164 164 168 168 169 170 173 173 174 176 179 180 181 182 183 183 185 185 188 193 194 195 195 196 196 199 200 203 212 213 213 214 214 9 .10 Maximum load check 9.1 Basic considerations 8.1 General 8.2.2.7 Impedances seen by distance relays 9.3.1 Phase units 9.viii 8 Contents Differential protection 8.3 Permissive over reach intertripping 9.1 General 9.6 Residual compensation 9.7 Relays with polygonal characteristics 9.8 Power system oscillations 9.3 Setting the reach and operating time of distance relays 9.3.4 The effect of infeeds on distance relays 9.6.2 Types of distance relays 9.3.2 Classification of differential protection 8.1 Differential system with multiple restraint 8.4 Determination of the slope 8.3 Reactance relay 9.5 Line differential protection 8.3 Transformer differential protection 8.4 Differential protection for generators and rotating machines 8.7.12.12 Intertripping schemes 9.3.3 Percentage of winding protected by the differential relay during an earth fault 8.1 Under reach with direct tripping 9.12.2 Selection and connection of CTs 8.6.12.9 The effective cover of distance relays 9.2 High impedance differential system 8.2.6 Relays with lens characteristics 9.5 Distribution of fault current in power transformers 8.2.2.11 Drawing relay settings 9.2 Permissive under reach intertripping 9.

2 Substation design diagrams 12.1 Generator protection 12.1.4.1 Single-line diagrams 12.16 10 Protection of industrial systems 10.2 Design of an automatic load shedding system 11.14.2 Motor protection 12.1 Thermal relays 10.2.1 Protection devices 10.5 Determination of the frequency relay settings 11.14 Technical considerations of distance protection in tee circuits 9.4.1.1.4 Line protection 12.4.4 Load shedding stages 11.2.4 Example of calculating and setting frequency relays in an industrial plant 11.1 Power system operation after loss of generation 11.3.1.2 Determination of the frequency variation 11.2 Parameters for implementing a load shedding system 11.2 Load to be shed 11.1 Protection schemes 12.2 Low voltage breakers Industrial plant load shedding 11.2 Substation layout diagrams 11 12 .3 Criteria for setting frequency relays 11.3 Combined thermal relay contactor and fuse 10.1.3 Frequency levels 11.15 9.6 Verification of operation Protection schemes and substation design diagrams 12.1.1 Tee connection with infeeds at two terminals 9.1 Calculation of overload 11.2.2.4.4.2 Tee connection with infeeds at all three terminals Use of distance relays for the detection of the loss of excitation in generators Exercises ix 215 215 218 219 222 225 225 225 225 226 226 226 227 239 239 240 240 241 242 242 243 243 243 243 243 243 244 247 251 251 251 252 258 261 261 262 263 9.3.2.2.3 Transformer protection 12.Contents 9.1 Simple machine model 11.14.1 Operating times 11.2 Criteria for setting overcurrent protection devices associated with motors 10.4.1 Overcurrent relays 10.1.2 Direct acting devices in power and moulded-case circuit breakers 10.

6 Finger printing approach 13.6 12.1 Factory tests 14.7 Hypothesis approach Installation.2 Alarm processing methods 13.3 12.3.5 12.2 Precommissioning tests 14.3 Periodic maintenance 14.2. testing and maintenance of protection systems 14.2.2.2.2.3.4 Equivalent alarms 13.1 Installation of protection equipment 14.x Contents 12.2 Performance tests 14 Bibliography Appendix: Solutions to exercises Index .2 Testing protection schemes 14.2.2.3 Commissioning numerical protection 14.1 General 13.7 Diagrams of AC connections Diagrams of DC connections Wiring diagrams Logic diagrams Cabling lists 264 265 266 268 268 269 269 270 271 272 273 273 275 283 283 285 285 285 290 292 292 293 297 301 339 13 Processing alarms 13.4 12.5 Rules 13.2.3 Expert systems 13.1 Setting the parameters 14.

Lo at Strathclyde University for his guidance. It is hoped that the many detailed examples and exercises throughout the book. F. C.Preface and acknowledgements The quality of electricity supplies is an important factor in the socio-economic development of any area. Subsequently the authors have expanded the text considerably and added much up to date material.J. by various organisations to include copyright material which is acknowledged in the text. 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. Professor K. Gers E. Colombia Stourbridge. for their ideas and assistance. The British Council for continuing support during many years of institutional exchange. and Stephen and Philip Holmes for sharing so generously their computer expertise. and customers rightly expect a high level of security for their supply. Delgado for his valuable comments on the initial manuscripts and our colleagues at GERS Ltd. 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.L. which the authors have taken from actual case studies in the field. England . J. Holmes Cali. it is also essential to provide suitable protection schemes and relay settings to ensure that faults are quickly disconnected to minimise outage times and improve the continuity of supplies to customers. Thanks are due to the University of Valle and the Colombian Institute of Sciences (COLCIENCIAS) for financial help. the authors have been most grateful for the help received from many sources and the permission. readily given. In addition. particularly those engaged in the co-ordination and setting of protection on distribution systems. Finally.M. will provide worthwhile material for planning and design engineers and maintenance staff. With this in mind this book has been produced as a reference guide for professional engineers and students. Although this can be achieved by good distribution network design using proven equipment. Pacheco in producing the diagrams. The book is based on original material by Mr Gers and translated from the Spanish by Mr Holmes. Approximately 75 per cent of all customer hours lost are owing to faults on the distribution networks.

.

They also have very good communication facilities which allow them to work in virtually any automated scheme. Chapters 3 and 5 have been considerably extended to include more detail on numerical relays. metering. Gers E. Our thanks are due to our colleague Professor K. Lo of the University of Strathclyde. Chapter 12. while the testing procedures covered in the last chapter now include ample reference to numerical protection.L. J. Most of the latest types of relays are now multifunctional devices with control. We have also taken the opportunity to update sections of the original text and have added a new chapter on the processing of alarms since the fast and efficient 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. Therefore.com . The last chapter was also thoroughly updated. modern relays now offer better protection coverage and can be programmed to automatically adjust for changes in power system topologies and different operating conditions owing to the multiple setting groups feature incorporated in most of them. Scotland.com ejholmes@compuserve. reporting and alarm functions in addition to their protection capabilities that normally include several types within the same device. for his help with this material.Preface to 2nd edition In the six years since this book was first published there have been considerable advances in relay protection design. particularly in respect of testing procedures applicable to numerical relays.M. Finally. Glasgow. once again we acknowledge the support of our wives Pilar and Maggie during the work on this edition. dealing with protection schemes. Holmes Weston. The development of powerful numerical algorithms and further improvements in digital technology have greatly extended the scope of protection systems. has been updated to take account of the new technology available.J. Florida Stourbridge USA England jmgers@gersusa.

.

minimising the costs of nondistributed energy is receiving increasing attention. A further requirement is the safety of the electricity supply. Associated with the distribution networks themselves are a number of ancillary systems to assist in meeting the requirements for safety. in both developing and developed countries. whereas permanent faults can result in longer outages.1 General With the increasing dependence on electricity supplies. staff and public. reliability and quality of supply. 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. and the network overall. In addition. In order to avoid damage. 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. the need to achieve an acceptable level of reliability. A properly co-ordinated protection system is vital to ensure that an electricity distribution network can operate within preset requirements for safety for individual items of equipment. The most important of these are the protection systems which are installed to clear faults and limit any damage to distribution equipment. vandalism. This then leads to an improved electricity service with better continuity and quality of supply. or just the shortest of interruptions.Chapter 1 Introduction 1. quality and safety at an economic price becomes even more important to customers. Automatic operation is necessary to isolate faults on the networks as quickly as possible in order to minimise damage. the . the deterioration of insulation. suitable and reliable protection should be installed on all circuits and electrical equipment. Amongst the principal causes of faults are lightning discharges. The majority of faults are of a transient nature and can often be cleared with no loss of supply. and tree branches and animals contacting the electricity circuits. When providing protective devices on any supply network the following basic principles must apply. On the occurrence of a fault or abnormal condition. Protective relays initiate the isolation of faulted sections of the network in order to maintain supplies elsewhere on the system.

2 Protection of electricity distribution networks EHV EHV/HV Major HV substation HV/MV HV network HV/MV MV MV overhead network to adjacent MV network MV cable network urban distribution rural distribution MV/LV LV board LV network Generator Transformer Circuit breaker Disconnector Feeder Fuse Customer Figure 1.1 EHV/HV/MV/LV network arrangements (reproduced from Electricity Distribution Network Design) .

the tendency is to replace these by microprocessor and numerical relays. but the vast majority of customers are connected at low voltage and supplied via MV/LV distribution substations and their associated networks.2. an extra high voltage (EHV) system. 500 and 765 kV. It has two elements – dependability. is required.Introduction 3 protection system must be capable of detecting it immediately in order to isolate the affected section. Figure 1. 1. The protection must be sensitive enough to operate when a fault occurs under minimum fault conditions. the medium voltage (MV) networks and low voltage (LV) networks are operated as radial systems. using the most appropriate voltage level. Below the transmission system there can be two or three distribution voltage levels to cater for the variety of customers and their demands. thus permitting the rest of the power system to remain in service and limiting the possibility of damage to other equipment. yet be stable enough not to operate when its associated equipment is carrying the maximum rated current. typical values being 400. Reliability: the ability of the protection to operate correctly. sometimes termed major or primary transmission.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. Disconnection of equipment must be restricted to the minimum amount necessary to isolate the fault from the system. In general. High voltage (HV) networks transport large amounts of power within a particular region and are operated either as interconnected systems or discrete groups. Such systems operate in the 300 kV plus range. 1. 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. particularly in the more complex protection arrangements. HV/MV transforming substations situated around each HV network supply individual MV networks. The HV and MV networks provide supplies direct to large customers. as shown in Figure 1. While electromechanical relays can still be found in some utilities. 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 in order to improve the reliability of the protection system. The HV networks are supplied from EHV/HV substations which themselves are supplied by inter-regional EHV lines. which may be a short-time value.3 Protection requirements The protection arrangements for any power system must take into account the following basic principles: 1.1 illustrates the interrelation of the various networks. which is the certainty of a correct operation on the occurrence of . Where the transport of very large amounts of power over large distances is involved.

and security. 2. which is the ability to avoid incorrect operation during faults. Speed: minimum operating time to clear a fault in order to avoid damage to equipment.2 Block schematic of transmission and distribution systems (reproduced from Electricity Distribution Network Design) a fault.4 Protection of electricity distribution networks Generation EHV Generation EHV/HV HV HV/MV Consumption MV MV/LV LV Consumption Figure 1. .

However. To achieve this. The protection element covers one or more components of the power system. designated as unit type protection. In general a power system can be divided into protection zones – generators. and can be used to back up the primary protection as a second line of defence. Figure 1. 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. lines and busbars. It is essential that any fault is isolated. more than one set of protection relays should operate. inevitably a compromise is required to obtain the optimum protection system. transformers. 1. such as electrical machines. The overlap is obtained by connecting the protection relays to the appropriate current transformers as illustrated in Figure 1. motors. Some relays.5. 1. operate only for faults within their protection zone.1 Primary protection Primary protection should operate every time an element detects a fault on the power system.4 Protection zones The general philosophy for the use of relays is to divide the system into separate zones. if a fault occurs in these overlap areas. usually in adjacent zones. It should be noted that the zones overlap at some points indicating that. busbars and lines. 4. 1. the primary protection does not work. this does not imply that they all have to operate for the same fault. for whatever reason. wherever possible. which can be individually protected and disconnected on the occurrence of a fault. Cost: maximum protection at the lowest cost possible.5. in order to permit the rest of the system to continue in service wherever possible. It is possible for a power system component to have various primary protection devices. Since it is practically impossible to satisfy all the above-mentioned points simultaneously.2 Back-up protection Back-up protection is installed to operate when. the back-up protection relay has a sensing . every element in the power system should be protected by both primary and back-up relays. 1. groups of generator transformers. Therefore.3 shows a system with different protection zones.4.5 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 a fault. in some cases it can be sited in an adjacent substation. Selectivity: maintaining continuity of supply by disconnecting the minimum section of the network necessary to isolate the fault. are able to detect faults both within a particular zone and also outside it.Introduction 5 3. even if the associated main protection does not operate. Other relays designated as non-unit protection.

6 Protection of electricity distribution networks M ~ M ~ Figure 1.4 Overlap of protection zones .3 Protection zones Busbar protection Line protection Figure 1.

Example 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. In these cases. examples are given where there has been incorrect operation of protection and the associated breakers. On protection schematic diagrams the directional protection is usually represented by an arrow underneath the appropriate symbol. their ability to inhibit opening of the associated switch when the fault current flows in the opposite direction to the setting of the relay. when fault currents can circulate in both directions around the mesh.5.1 .1 Check on correct operation of protection Using the power system shown in Figure 1. leading to the operation of back-up protection to isolate the fault from the system. indicating the direction of current flow for relay operation. Relays provided with this characteristic are important in protecting mesh networks.Introduction 7 element which may or may not be similar to the primary protection. or where there are various generation sources. 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. by this means. 1. One relay can provide back-up protection simultaneously to different pieces of system equipment.5 Power system for Example 1. followed by an example A 2 G2 B 5 G3 9 G4 G1 1 3 4 6 10 F1 7 F2 8 11 F3 F4 Figure 1. but which 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. directional protection prevents the unnecessary opening of switchgear and thus improves the security of the electricity supply.

8 10 8.5 kV 27 San Luis 34.1 Breakers that mal-operated 3 6 – 8 Tripped by primary protection 4 – 10 11 Tripped by back-up protection 1. The directional protection is indicated by the arrows below the corresponding breakers. 8 – – Breakers that operated 1.6 Schematic diagram for Exercise 1.5 kV F3 17 Tabor 115 kV 16 18 21 23 22 24 26 Diesel 34.1 shows the breakers that failed to open and those that were tripped by the primary protection and by the back-up protection.1 Case Relay/breaker operations for Example 1. 2 3.5 kV F2 19 25 G3 Figure 1. 4 3. 5. 5.8 Protection of electricity distribution networks of correct relay operation. 2. Table 1. 11 F1 F2 F3 F4 G1 1 Termoyumbo 115 kV G2 2 3 4 F1 10 Bajo Anchicayá 115 kV 5 6 Chipichape 115 kV 9 29 San Luis 115 kV 28 11 12 15 Pailon 115 kV 13 14 7 8 Chipichape 20 34. Table 1.1 . with a final example of unnecessary relay operation.

the directional protection on breakers 3 and 8 operated to open the incoming feeders from the adjacent busbars. as in Example 1.1 For the power system arrangement shown in Figure 1. 23 24.6. Please note that. With fault F2 . 4. 22. Fault F4 was correctly cleared by the operation of breaker 11. With breaker 3 failing to open. complete Table 1. 1.7 Exercise 1. taking into account the operation of the circuit breakers as shown for each fault case. when breaker 6 failed to operate.Introduction 9 For fault F1 . with the last one illustrating maloperation from a security standpoint.2 Case Relay/breaker operations for Exercise 1. 3. Thus the first two cases illustrate mal-operation from a dependability point of view. the protection correctly tripped breaker 4 to open one end of the faulted feeder. 19 F1 F2 F3 .1. 5 21. and the back-up protection on breaker 5 tripped to stop G3 feeding into the fault. so that the tripping of breaker 8 was incorrect. Table 1.1 Breakers that mal-operated Tripped by primary protection 2. 17. 27 10. 5 Tripped by back-up protection Breakers that operated 2. Fault F3 was correctly cleared by the tripping of feeder breaker 10.2. some of the circuit breakers that operated may have done so unnecessarily. Relay 8 is directional and operation should not have been initiated for flows from 7 to 8. breakers 1 and 2 were tripped by back-up protection to stop fault current flowing into the fault from generators G1 and G2. 11. Any fault current flowing along inter-busbar feeder 7-8 before breaker 11 opened would have been from 7 to 8.

.

While the use of these short-circuit calculations in relation to protection settings will be considered in detail. 2. from the start of the event at time t = 0+ until stable conditions are reached. which results in a decaying AC component. which produces a decaying DC component. which result in a decaying DC component. consider an RL circuit as a simplified equivalent of the circuits in electricity distribution networks. given the large increase in current flow when a short-circuit occurs. 2. To perform the corresponding calculations.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. In order to illustrate the transient nature of the current. For this reason a review of the concepts and procedures for calculating fault currents will be made in this chapter. and it is therefore necessary to use differential equations when calculating these currents. 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. This simplification is important because all the system equipment must be modelled . the treatment of electrical faults should be carried out as a function of time. and the performance of the rotating machinery. together with some calculations illustrating the methods used. the selection of conductor sizes and for the specifications of equipment such as power circuit breakers. for example calculating the substation earthing grid. it is important to bear in mind that these calculations are also required for other applications.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.Chapter 2 Calculation of short-circuit currents The current that flows through an element of a power system is a parameter which can be used to detect faults.1.

2) . the first term varies sinusoidally and the second term decreases exponentially with a time constant of L/R. of which the solution is in two parts: ia (t) : ih (t) + ip (t) where ih (t) is the solution of the homogeneous equation corresponding to the transient period and ip (t) is the solution to the particular equation corresponding to the steadystate period. By the use of differential equation theory.1. in eqn. For the circuit shown in Figure 2. which will not be discussed in detail here.2.1) dt This is a differential equation with constant coefficients. see Figure 2.2. the complete solution can be determined and expressed in the following form: i(t) = where Z = (R 2 + ω2 L2 ) α = the closing angle. which defines the point on the source sinusoidal voltage when the fault occurs.12 Protection of electricity distribution networks L Vmax sin ( t + ) R Figure 2. and φ = tan−1 (ωL/R) It can be seen that. 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 Vmax sin(ωt + α − φ) − sin(α − φ)e−(R/L)t Z (2. while the second term can be recognised as the DC component of the current having an initial maximum value when α − φ = ±π/2. The first term corresponds to the AC component. and zero value when α = φ.1 RL circuit for transient analysis study in some way in order to quantify the transient values that can occur during the fault condition. the mathematical expression that defines the behaviour of the current is di e(t) = L + Ri(t) (2. 2.

2 Effect of rotating machinery When a fault occurs close to the terminals of rotating machinery. including the AC and DC components.4 it will be noted that the variation of current with time. comes close to the three discrete levels of current. 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.asym = Irms + IDC (2. Notwithstanding this. transient and steady state currents respectively. An approximate formula for calculating the effective value of the total asymmetric current. I and I . can be interpreted as a reactance that varies with time. The corresponding values of direct axis reactance . Here. the DC component reaches its theoretical maximum value half a cycle later. I . In Figure 2. takes place when the sinusoidal component is at its negative peak. a decaying AC current is produced. The reduction in current from its value at the onset.2 Variation of fault current with time: (a) α − φ = 0. due to the gradual decrease in the magnetic flux caused by the reduction of the m. This effect is known as armature reaction. with acceptable accuracy can be used by assuming that these components are in quadrature with the following expression: 2 2 Irms.1. due to a fault. the sub-transient. If the tripping of the circuit. (b) α − φ = −π/2 reach. and which makes the calculations quite difficult.3) 2. of the induction current. 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.m.Calculation of short-circuit currents (a) 13 (b) Figure 2. can be seen in Figure 2. The physical situation that is presented to a generator. I (t). similar in pattern to that flowing when an AC voltage is applied to an RL circuit as discussed in the previous section.3.f.

14

Protection of electricity distribution networks
(a)

ia

(b)

ib

(c)

ic

Figure 2.3

Transient short-circuit currents in a synchronous generator: (a) typical A phase short circuit current; (b) typical B phase short circuit current; (c) typical C phase short circuit current

are denoted by Xd , Xd and Xd , and the typical variation with time for each of these is illustrated in Figure 2.5.

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.

Calculation of short-circuit currents

15

I

I I

I (t)

t1

t2

Figure 2.4

Variation of current with time during a fault

Xd

X (t)

Xd Xd t1 t2

Figure 2.5

Variation of generator reactance with time during a fault

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 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 indicated

16

Protection of electricity distribution networks

by Ik . These are the currents that are sensed by circuit breaker protection equipment when a fault occurs and are therefore also called close and latch currents. They are calculated with DC offset but no AC decrement in the sources, and using the machine sub-transient reactances. Peak currents correspond to the maximum currents during the first cycle after the fault occurs and differ from the first cycle currents that are totally asymmetrical r.m.s. 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 3 to 5 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 r.m.s. values can be defined depending on whether the DC component √ included or not. is The peak values are obtained by multiplying the r.m.s. values by 2. The asymmetrical values are calculated as the square root of the sum of the squares of the DC component and the r.m.s. value of the AC current, i.e.
2 2 Irms = IDC + IAC

(2.4)

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 per cent of their initial values after the first half cycle. From this, the value of the r.m.s. current would then be
2 2 Irms.asym.closing = IDC + IAC.rms.sym.

√ = (0.9 2V /Xd )2 + (0.9V /Xd )2 = 1.56V /Xd = 1.56Irms.sym. (2.5)

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. The peak value is obtained by arithmetically adding together the AC and DC components. It should be noted that, in this case, the AC component is multiplied by √ a factor of 2. Thus Ipeak = IDC + IAC √ √ = (0.9 2V /Xd ) + (0.9 2V /Xd ) = 2.55Irms.sym. (2.6)

Calculation of short-circuit currents
Duty calculation (see Note 1) System component Reactance value for medium- and high-voltage calculations per IEEE Std C37.010-1979 and IEEE Std C37.5-1979 Xs 1.0 Xd 0.75 Xd 1.0 Xd 1.0 Xd 1.0 Xd 1.2 Xd 1.67 Xd (see Note 6) Xs 1.0 Xd 0.75 Xd 1.5 Xd 1.5 Xd 1.5 Xd 3.0 Xd Neglect Reactance value for low-voltage calculations (see Note 2)

17

First cycle (momentary calculations)

Power company supply All turbine generators; all hydrogenerators with amortisseur windings; all condensers Hydrogenerators without amortisseur windings All synchronous motors Induction motors Above 1000 hp Above 250 hp at 3600 rev/min All others, 50 hp and above All smaller than 50 hp Power company supply All turbine generators; all hydrogenerators with amortisseur windings; all condensers Hydrogenerators without amortisseur windings All synchronous motors Induction motors Above 1000 hp Above 250 hp at 3600 rev/min All others, 50 hp and above All smaller than 50 hp

Xs 1.0 Xd 0.75 Xd 1.0 Xd 1.0 Xd 1.0 Xd 1.2 Xd 1.67 Xd N/A N/A

Interrupting calculations

N/A N/A N/A N/A N/A N/A

Notes: 1: 2: 3: 4: 5: 6: 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. Reactance (X) values to be used for low-voltage breaker duty calculations (see IEEE Std C37.131990 and IEEE Std 242-1986). Xd of synchronous-rotating machines is the rated-voltage (saturated) direct-axis subtransient reactance. Xd of synchronous-rotating machines is the rated-voltage (saturated) direct-axis transient reactance. Xd of induction motors equals 1 divided by per-unit locked-rotor current at rated voltage. 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)

18

Protection of electricity distribution networks

When considering the specification for the switchgear opening current, the socalled r.m.s. value of interrupting current is used in which, again, the AC and DC components are taken into account, and therefore
2 2 Irms.asym.int. = IDC + IAC.rms.sym.int.

Replacing the DC component by its exponential expression gives Irms.asym.int. = √ 2Irms.sym.int. e−(R/L)t
2 2 + Irms.sym.int.

= Irms.sym.int. 2e−2(R/L)t + 1

(2.7)

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.
150 140 130 120 110 100 90 Ratio X/R 80 70 60 50 40 30 20 10
Co
rt

4
pa ctn ta

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)

in

g ti

me

, cy

cle s

3 2 1

Calculation of short-circuit currents

19

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 , is 60 Hz and the ratio X/R is given as 50, with t = 2 cycles = 0.033 s, then (X/R) = (ωL/R) = 50. Thus (L/R) = (50/ω) = (50/2π f ) = 0.132. Therefore Iasym = 2 e−(0.033×2)/0.132 + 1 = 1.49 Isym as can be seen from Figure 2.7. In protection co-ordination studies, r.m.s. 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 therefore can be treated by using a single-phase representation. This symmetry is lost during asymmetric 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 which were named the symmetrical components of the original phases. While 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 current, is replaced by three components so that a total of nine vectors uniquely represents the values of the three phases. The three system balanced phasors are designated as: 1. Positive-sequence components, which consist of three phasors of equal magnitude, 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 positivesequence phasors.

20

Protection of electricity distribution networks

With this arrangement, voltage values of any three-phase system, Va , Vb , and Vc , can be represented thus: Va = Va0 + Va1 + Va2 Vb = Vb0 + Vb1 + Vb2 Vc = Vc0 + Vc1 + Vc2 It can be demonstrated that Vb = Va0 + a 2 Va1 + aVa2 Vc = Va0 + aVa1 + a 2 Va2 where a is a so-called operator which gives a phase shift of 120◦ clockwise and a multiplication of unit magnitude i.e. a = 1 120◦ , and a 2 similarly gives a phase shift of 240◦ , i.e. a 2 = 1 240◦ . Therefore, the following matrix relationship can be established: ⎡ ⎤ ⎡ ⎤ ⎡ ⎤ 1 1 1 Va Va0 ⎣Vb ⎦ = ⎣1 a 2 a ⎦ × ⎣Va1 ⎦ Va2 Vc 1 a a2 Inverting the matrix of coefficients ⎡ ⎤ ⎡ ⎤ ⎡ ⎤ 1 1 1 Va0 Va 1⎣ ⎣Va1 ⎦ = 1 a a 2 ⎦ × ⎣Vb ⎦ 3 Va2 Vc 1 a2 a From the above matrix it can be deduced that Va0 = 1/3(Va + Vb + Vc ) Va1 = 1/3(Va + aVb + a 2 Vc ) Va2 = 1/3(Va + a 2 Vb + aVc ) The foregoing procedure can also be applied directly to currents, and gives Ia = Ia0 + Ia1 + Ia2 Ib = Ia0 + a 2 Ia1 + aIa2 Ic = Ia0 + aIa1 + a 2 Ia2 Therefore Ia0 = 1/3(Ia + Ib + Ic ) Ia1 = 1/3(Ia + aIb + a 2 Ic ) Ia2 = 1/3(Ia + a 2 Ib + aIc ) In three-phase systems the neutral current is equal to In = (Ia + Ib + Ic ) and, therefore, In = 3Ia0 . By way of illustration, a three-phase unbalanced system is shown in Figure 2.8 together with the associated symmetrical components.

4° Vc0 Vc2 Vc Vc1 Va1 Va Va0 Vb1 Vb Vb0 Vb2 Va2 Va2 = 4.0 143.8 138.1° Vb0 = 2.3 –86.0 –90° Vc = 16.8 18.2° Vb2 = 4.3 33.8 Symmetrical components of an unbalanced three-phase system .8 –101.0 143.4° Vb1 = 9.2° Va0 = 2.6° Vc1 = 9.0 143.1° Figure 2.8° Vc2 = 4.1° Vc0 = 2.1° Vb0 Vc0 Vb = 6.3 –206.0 0° Vc1 Va1 Vb2 Vc2 Va0 Va2 Vb1 Va1 = 9.0 143.Calculation of short-circuit currents Vc 21 Va Vb Va = 8.

5 with an earth wire. and 3 to 5 for three-core cables.and negative-sequence impedances are equal because in static circuits these impedances are independent of the phase order.m. Xd and Xd . the positive. transient and synchronous reactances. depending on the transformer connections. is at earth potential so only zero-sequence currents flow through the impedances between neutral and earth.5. provided that the applied voltages are balanced. denoted by Xd . The zero-sequence network carries only zero-sequence current in one phase. When modelling small generators and motors it may be necessary to take resistance into account. Typical per-unit reactances for three-phase synchronous machines are given in Table 2. as are those of cables. depending on the machine characteristics and fault clearance time. The resistance of the windings is much smaller and can generally be neglected in short-circuit calculations.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.and zero-sequence impedances. the generated voltages are of positive sequence only. The subtransient reactance is the reactance applicable at the onset of the fault occurrence. the reference busbar for the positiveand negative-sequence networks is the generator neutral which. for most studies only the reactances of synchronous machines are used.s but only include impedances to the flow of negativeand zero-sequence currents respectively. Three values of positive-reactance are normally quoted – subtransient.and zero-sequence networks do not contain e.m. The positive. The zero-sequence impedances of lines differ from the positive. Since generators are designed to supply balanced voltages. The current that flows in the impedance Zn between the neutral and earth is three times the zero-sequence current.2. those in which only negative. For transformers.f. The reference busbar for zero-sequence networks is the earth point of the generator. The zero-sequence impedance is either the same as the other two impedances. For a single circuit line. similarly.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. Therefore the positive-sequence network is composed of an e. These sequence impedances are designated Z1 .1. In connecting sequence networks together.f.9 illustrates the sequence networks for a generator. being independent of the phase if the applied voltages are balanced. For a double circuit line Z0 /Z1 = 5. .and negative-sequence impedances of overhead line circuits are identical.and negative-sequence currents is different from that for the zero-sequence currents.25 for single core. Z2 and Z0 respectively and are used in calculations involving symmetrical components. In fault studies the subtransient and transient reactances of generators and motors must be included as appropriate. Within 0.and negative-sequence impedances since the magnetic field creating the positive.22 Protection of electricity distribution networks 2. The negative. The following ratios may be used in the absence of detailed information. Figure 2. For underground cables Z0 /Z1 can be taken as 1 to 1. which has an impedance of Z0 = 3Zn + Ze0 .and zero-sequence currents flow are called the negative. source in series with the positive-sequence impedance. However. or infinite. in these networks. Z0 /Z1 = 2 when no earth wire is present and 3.

18 0. and can be deduced from Figure 2.09 0.28 Xd 0. Ic = 0 and Va = 0. carrying currents I1 . and thus the treatment and connection of the sequence networks will be similar.and zero-sequence networks. Z0 = zero-sequence impedance of the generator (Zg0 ) plus three times the impedance to earth. . 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. for each type of fault. Z2 = negative-sequence impedance of the generator. the appropriate combination of sequence networks is formed in order to obtain the relationships between fault currents and voltages.12 The voltage and current components for each phase are obtained from the equations given for the sequence networks. Then.2. it can easily be deduced that Ia1 = Ia2 = Ia0 = Ea /(Z1 + Z2 + Z0 ).15 0. Line-to-earth fault The conditions for a solid fault from line a to earth are represented by the equations Ib = 0. 2. As in the previous equations.20 X2 0. I2 and I0 respectively. negative.09 0.20 1.14 0.07 0. the sequence networks will be connected in series.03 0. Ea then being the voltage behind the reactance before the fault occurs. The equations for the components of voltage. as indicated in Figure 2.30 0.20 0.25 1.70 1. The current and voltage conditions are the same when considering an open-circuit fault in phases b and c.1 Typical per-unit reactances for three-phase synchronous machines Xd 2 pole 4 pole with dampers without dampers 0. corresponding to the a phase of the system. are connected together in a particular arrangement to represent a given unbalanced fault condition. it is essential to determine the individual sequence impedances and combine these to make up the correct sequence networks.2 Calculation of asymmetrical faults using symmetrical components The positive-.14 0.35 X0 23 Type of machine Turbine Generator Salient pole Generator 0.30 Xd 1.Calculation of short-circuit currents Table 2. are obtained from the point a on phase a relative to the reference busbar. The same approach can be used with equivalent power systems or to loaded generators. Therefore. in order to calculate fault levels using the method of symmetrical components. Z1 = positivesequence impedance of the generator.22 0.20 0.10a. Consequently.9 as follows: Va1 = Ea − Ia1 Z1 Va2 = −Ia2 Z2 Va0 = −Ia0 Z0 where: Ea = no load voltage to earth of the positive-sequence network.

9 Equivalent sequence networks and current flows for a synchronous generator: (a) positive-sequence current. (f ) zero-sequence network . (c) negative-sequence current. (d) negative-sequence network.24 (a) Protection of electricity distribution networks Ia1 a Z1 + Ec – – – + Ea Eb + Z1 b Ib1 Ic1 (c) a Reference bus Z2 Ia2 (d) Z1 a Ia1 Ea + Va1 – (b) Reference bus Z1 c Z2 c Z2 Z2 b Ib2 Ic2 Va2 a Ia2 (f ) Reference bus (e) a Ia0 Ib0 = Ia0 Ic0 = Ia0 Zn Ze0 c Ze0 b Ia0 Ze0 3Zn Z0 Ib0 Ic0 Ze0 a Ia0 Va0 Figure 2. (e) zerosequence current. (b) positive-sequence network.

(c) double phase-to-earth fault Line-to-line fault The conditions for a solid fault between lines b and c are represented by the equations Ia = 0. For this case.Calculation of short-circuit currents (a) Ea Va1 Z1 Z2 Va2 Va0 Ia1 25 Z0 (b) Ea Va1 Va2 Z1 Z2 Ia1 Ia2 (c) Ea Z1 Va1 Z2 Va2 Z0 Va0 Ia1 Ia2 Ia0 Figure 2. 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.10 Connection of sequence networks for asymmetrical faults: (a) phaseto-earth fault. with no zero-sequence current. Equally it can be shown that Ia0 = 0 and Ia1 = Ea /(Z1 + Z2 ) = −Ia2 . Ib = −Ic and Vb = Vc . (b) phase-to-phase fault. .

it may be assumed that overall Z2 = Z1 .and negative-sequence impedances. From these equations it can be proved that: Ia1 = Ea Z1 + (Z0 Z2 )/(Z0 + Z2 ) The three sequence networks are connected in parallel as shown in Figure 2.and negative-sequence impedances can be calculated directly from: Z = V 2 /P where: Z = equivalent positive. V = nominal phase-to-phase voltage. where VLN = line-to-neutral voltage and Z0 = (3VLN /Ia ) − 2Z1 . 2. Ia1 = Ia2 = Ia0 = VLN /(Z1 + Z2 + Z0 ). the advantage of using symmetrical 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. given this. In Figure 2. The equivalent zero-sequence of a system can be derived from the expressions of sequence components referred to for a single-phase fault. on the basis that the generator impedances are not significant in most distribution network fault studies. It is very important to emphasise that. The equivalent positive. i.2.. which simplifies the calculations. For lines and cables the positive.11 the positive.3 Equivalent impedances for a power system When it is necessary to study the effect of any change on the power system. Thus. Thus. the system must first of all be represented by its corresponding sequence impedances. the above formula reduces to Ia = 3Ia0 = 3VLN /(2Z1 + Z0 ).e.and negative-sequence values of current and voltage for different faults are shown together with the summated values of current and .and negative-sequence impedances are equal. the relays operate on the basis of the corresponding values of fault current and/or voltages.3 Supplying the current and voltage signals to protection systems In the presence of a fault the current transformers (CTs) circulate current proportional to the fault current to the protection equipment without distinguishing between the vectorial magnitudes of the sequence components. in the majority of cases. where VLN = the line-to-neutral voltage. Therefore. regardless of the values of the sequence components.26 Protection of electricity distribution networks Line-to-line-to-earth fault The conditions for a fault between lines b and c and earth are represented by the equations Ia = 0 and Vb = Vc = 0.10c. 2. P = three-phase short-circuit power.

b. e c1 b1 a2 a1 Calculation of short-circuit currents c. c 0 0 0 c b. b a2 c1 b1 a1 b. c c1 a1 a. e b=c=0 b a=c=0 c a=b=0 b.11 27 Currents and voltages for various types of faults: (a) sequence currents for different types of fault. e c1 b1 a1 c. e a2 c1 b1 a1 b. c. c c1 a b a=b a b1 a1 a. b0. b . b0. b0. c0 b=0 a c2 c2 b2 a . b. c 0 0 0 b2 a2 a . b. a. a b2 c1 b1 a c a=c b a1 a. c0 b1 a1 b c1 a a0. b. b0. c c2 c1 b1 a1 c. e c1 a1 c1 a1 c. c0 b2 b2 a2 a2 c2 a2 a0. e c1 b1 b2 a1 b. c 0 0 0 b2 a0. e c1 b1 c2 Figure 2. c0 c=0 c b1 b2 c2 a2 a2 a2 a0. (b) sequence voltages for different types of fault . b . a. b0. e c1 a1 a.(a) Zerosequence network current a c b c1 b1 a1 b b2 c b=c a2 c2 a2 c2 b2 c2 b1 a1 c. c0 c2 a . b c1 b1 a2 c2 c2 b2 b2 c a=0 b=0 a a2 a2 c2 c=0 a a. b0. e c1 b1 c2 b1 a2 b1 b2 c2 b2 a . c0 a0. c. a c c2 a0. e c2 c a=b=0 a b=c=0 b a=c=0 a=0 c a b=0 a b b c1 b1 a1 c. b . b0. e b2 c1 b1 a1 a. c0 a0. b0. c 0 0 0 b2 a2 a0. c0 c a=0 c=0 a c1 b b. e c2 b2 a2 b2 b1 a1 b a. b . c Zero a1 fault Fault currents Zerosequence network voltage Fault Positivesequence network voltage a1 Negativesequence network voltage Fault voltages Fault Positivesequence network current a1 Negativesequence network current (b) a.

line-to-line. an updated single-line diagram can be produced. The interactive program permits examination of the results as they are printed on the screen or via the printer. (i. line-to-earth. however.e. Amongst the relays that require this type of filter in order to operate are those used in negative-sequence and earth-fault protection.4 Calculation of faults by computer The procedure for calculating fault levels starts by taking the single-line diagram of the system under analysis. line-to-line-to-earth.e. and this is achieved by using filters that produce the mathematical operations of the resultant equations to resolve the matrix for voltages and for currents. and enables the user to select those results that are important in the study. it is possible to alter variables such as the type of fault being analysed. negative. conductor diameter. When investigating electrical faults. this method speeds up the calculations considerably bearing in mind that. the values of connections to earth. relays are available that can operate with specific values of some of the sequence components. The corresponding positive-. indicating the positive-sequence impedance values referred to the respective base quantities. three-phase. A large proportion of the existing programs have been developed with interactive algorithms whose principal characteristic is the man-machine dialogue. print out. but also for faults such as those between systems of different voltages. as well as being able to keep direct control on the performance of the program.and zero-sequence networks can then be built up to form the basis of the calculations of the voltage and current under fault conditions.).28 Protection of electricity distribution networks voltage. In these cases there must be methods for obtaining these components. Modern software packages enable the following features to be carried out: • Calculations not only for the standard fault types. the calculation of fault levels for large power systems is now invariably carried out using computers. given the vast facilities that are now available. interrogate files. 2. collecting the sequence impedances for all the components. i. etc. when to stop/check. Having obtained these values. . However. etc. collecting background data including machine impedances. Relays usually only operate using the summated values in the right-hand columns. For small systems. hand calculations can still be used since short-circuit calculations do not require an iterative process. the growth of electronics has led to their being used increasingly in logic circuits. the length. the faulted busbar and values of impedance. both in hardware and software. calculating the Thevenin equivalent of neighbouring systems. etc. and the configuration of the feeders. Having collected and processed the basic information. This is much superior to the batch process used earlier. Although these filters can be constructed for electromagnetic elements.

. Include pre-fault values. Calculate simultaneous faults. Calculate the different duties associated with a fault and handle IEEE and IEC Standards. (sequence and phase values). which can be slightly different especially in the pre-fault voltage level. Calculate faults along different line lengths. from the different elements whether or not they are associated with the faulted node.Calculation of short-circuit currents • • • • • 29 Indicate the fault contributions.

.

pressure. 3. usually from a current and/or voltage source.2 • • • • • • • • Incoming signal current voltage power frequency temperature pressure speed others .1 • • • • • Construction electromechanical solid state microprocessor numerical non-electric (thermal.1.1 Classification Protection relays can be classified in accordance with their construction.1. If the magnitude of the incoming signal is outside a pre-set value the relay will carry out a specific operation.) 3. etc. the incoming signal and function. generally to close or open electrical contacts to initiate some further operation.Chapter 3 Classification and function of relays A protection relay is a device that senses any change in the signal it is receiving. for example the tripping of a circuit breaker. 3.

3 • • • • • • • Function overcurrent directional overcurrent distance overvoltage differential reverse power others 3.1.1. which is used in the following chapters. for flow or level of liquid or gases earth protection relay directional overcurrent relay blocking relay alarm relay out-of-step relay reclosing relay frequency relay carrier or pilot-wire receiver relay lockout relay differential relay auxiliary tripping relay . is given below: 21 24 25 26 27 32 37 40 41 43 46 47 49 50 51 52 55 59 60 62 63 64 67 68 74 78 79 81 85 86 87 94 distance relay volts/hertz synchronising or synchronism-check device thermal device undervoltage relay reverse-power relay under-current or under-power relay relay for field excitation field circuit breaker manual transfer or selector device negative-sequence current relay negative-sequence voltage relay thermal relay instantaneous overcurrent relay time-delay overcurrent relay circuit breaker power factor relay overvoltage relay voltage or current balance relay time-delay relay pressure relay.4 International identification of electrical devices The international classification for the more common relays.32 Protection of electricity distribution networks 3.

according to the design. T for transformer. There are two main types of relay in this class. Their construction characteristics can be classified in three groups.1. They are also referred to as electromagnetic relays due to their magnetic components. In this case.1 Armature-type relay . in which a bar or piston is attracted axially within the field of the solenoid. 3.Classification and function of relays 33 In some cases a letter is added to the number associated with the protection in order to specify its place of location.2 Electromechanical relays These relays are constructed with electrical. for example G for generator. Non-electric relays are outside the scope of this book and therefore are not referred to. which is shown in Figure 3. etc.2. illustrated in Figure 3. when the armature is attracted to the coil. and are very robust and reliable. The other type is the piston or solenoid type relay. magnetic and mechanical components and have an operating coil and various contacts.2.1 Attraction relays Attraction relays can be supplied by AC or DC. 3. the piston also carries the operating contacts. Fixed contact Moving contact Restraining spring Armature Coil Pivot Figure 3. The armature carries the moving part of the contact which is closed or opened. and operate by the movement of a piece of metal when it is attracted by the magnetic field produced by a coil. consists of a bar or plate of metal that pivots when it is attracted towards the coil. as detailed below. The attracted armature type.

Attraction relays effectively have no time delay and. From the above equation it will be noted that the torque developed is proportional to the current. 3. The torque produced in the coil is given by T = BlaNi where: T = torque. K2 is the restraining force. the air gap. In order to control the value at which the relay starts to operate. thus modifying the restricting force.2 Relays with moveable coils This type of relay consists of a rotating movement with a small coil suspended or pivoted with the freedom to rotate between the poles of a permanent magnet.2 Solenoid-type relay It can be shown that the force of attraction is equal to K1 I 2 − K2 . When the √ relay is balanced. amongst other factors. The speed of movement is controlled by the damping action which is . B = flux density. The coil is restrained by two springs which also serve as connections to carry the current to the coil. usually produced by a spring. are widely used when instantaneous operation is required. N = number of turns on the coil. l = length of the coil. the restraining tension of the spring or the resistance of the solenoid circuit can be varied. so that I = (K1 /K2 ) = constant. for that reason. where K1 depends upon the number of turns on the operating solenoid.2. i = current flowing through the coil. the effective area and the reluctance of the magnetic circuit. the resultant force is zero and therefore K1 I 2 = K2 .34 Protection of electricity distribution networks Coil Fixed contact Piston Moving contact Restraining spring Figure 3. a = diameter of the coil.

and functions through the interaction of electromagnetic fluxes with the parasitic Foucalt currents that are induced in the rotor by these fluxes. produce a torque that can be expressed by T = K1 1 2 sin .2. Then. generally in the form of a disc or cup. i and i 2 1 ∝ d 1 ∝ dt 1 cos ωt ∝ d 2 ∝ dt 2 cos(ωt + ) . where 1 and 2 are the interacting fluxes and is the phase angle between 1 and 2 . These two fluxes.3 Inverse time characteristic proportional to the torque. It can be shown that 1 = 1 sin ωt.3 Induction relays An induction relay works only with alternating current. and 2 = 2 sin(ωt + ). which are mutually displaced both in angle and in position. It thus follows that the relay has an inverse time characteristic similar to that illustrated in Figure 3. where is the angle by which 2 leads 1 .3. It consists of an electromagnetic system which operates on a moving conductor. The relay can be designed so that the coil makes a large angular movement. 3.Classification and function of relays 35 Time in seconds Multiples of tap value current Figure 3. It should be noted that the torque is a maximum when the fluxes are out of phase by 90◦ and zero when they are in phase. for example 80◦ .

5).4 shows the interrelationship between the currents and the opposing forces. Figure 3. Thus: F = (F2 − F1 ) ∝ ( F∝ F∝ F∝ F∝ 2 sin(ωt 1 1 1 2 [sin(ωt 2i 1 − 1i 2 ) 1 sin ωt 2 cos(ωt + ) 1 cos ωt − + ) + ) cos ωt − sin ωt cos(ωt + )] + ) − ωt}] 2 [sin{(ωt 2 sin ∝T Induction relays can be grouped into three classes as set out below. with just one supply for the top coil. which induces an out-of-phase flux in the lower coil because of the air gap. This creates a flux in the area influenced by the short-circuited section (the so-called shaded section) which lags the flux in the non-shaded section (see Figure 3.7).4 Electromagnetic forces in induction relays Figure 3. in some cases. . this type of relay uses an arrangement of coils above and below the disc with the upper and lower coils fed by different values or. (ii) Wattmetric type relay In its more common form. (iii) Cup type relay This type of relay has a cylinder similar to a cup which can rotate in the annular air gap between the poles of the coils. and has a fixed central core (see Figure 3.6 illustrates a typical arrangement. (i) Shaded pole relay In this case a portion of the electromagnetic section is short-circuited by means of a copper ring or coil.36 Protection of electricity distribution networks Φ2 iΦ2 iΦ2 Φ1 iΦ1 F2 F1 iΦ1 iΦ2 iΦ1 Figure 3.

5 Shaded-pole relay Core Upper coil Disc Lower coils Figure 3. A special spring provides the restraining torque. Configurations with four or eight poles spaced symmetrically around the circumference of the cup are often used. The movement of the cylinder is limited to a small amount by the contact and the stops.6 Wattmetric-type relay The operation of this relay is very similar to that of an induction motor with salient poles for the windings of the stator. The torque is a function of the product of .Classification and function of relays Moving contact Core 37 Fixed contact Spindle Coil Disc Shading ring Figure 3.

As electronic technology progressed. which are provided with a disc. 3. I1 and I2 are the currents through the two coils. electromechanical relays were superseded in the 1960s by electronic or static designs using transistors and similar types of electronic elements. Ks . the design of protection relays has changed significantly over the past years with the advancement in microprocessor and signal processing technology. However. the inertia of the disc provides the time delay characteristic. low prices and reliability. and 12 is the angle between I1 and I2 . The time delay can be increased by the addition of a permanent magnet.3 Evolution of protection relays The evolution of protection relays started with the attraction type of relay referred to earlier. . The integrated circuit (IC) enabled static designs to be further extended and improved in the 1970s. Subsequently in the 1990s microprocessor technology. In the first two types of relay mentioned above. along with the improvements in mathematical algorithms.38 Protection of electricity distribution networks Corner coil Air gap Lateral coil Central core Cup 8 pole stator Figure 3. where K. for example in instantaneous units. basic programmable microprocessor or micro-controlled multifunction protection relays first started to appear in the early 1980s. The cup type relay has a small inertia and is therefore principally used when high-speed operation is required. and are design constants.7 Cup-type relay the two currents through the coils and the cosine of the angle between them. Following the development of the microprocessor. The torque equation is T = [KI1 I2 cos( 12 − ) − Ks ]. spurred the development of the so-called numerical relays which are extremely popular for their multifunctional capabilities.

or some other action. such as communications. the energising of a status input. extensive communication features and an advanced human-machine interface (HMI). These functions can be integrated into one digital system so that a substation can be operated in a more rapid and reliable manner.2 Characteristics of numerical relays Numerical relays are technically superior to the conventional types of relays described earlier in this chapter. for example. System currents and/or voltages. 3.Classification and function of relays 39 3. or reference value.4 3. like all other quantities. Most numerical relays are multifunctional and can be regarded as intelligent electronic devices (IEDs). which is normally built into most relays. disturbance records can be generated in a number of analogue channels. Their general characteristics are: • • Reliability: incorrect operations are less likely with numerical relays. provide easy access to the available features. the numerical relay can provide adaptive protection. are not monitored on a continuous basis but. are sampled one at a time. Depending upon the algorithm used. Adaptive protection: with the programming and communication capacity of digital systems. Event and disturbance records: these relays can produce records of events whenever there is a protection function operation. After acquiring samples of the input waveforms. In addition. or any hardware failure. Self-diagnosis: numerical relays have the ability to conduct continuous selfdiagnosis in the form of watchdog circuitry. or only once over many cycles. together with all the status input and relay output information. Integration of digital systems: the present technology now includes many other tasks at one substation. measurement and control. Fibre optics are now being used to provide communication links between various system elements to avoid the interference problems that can occur when using metallic conductors.1 Numerical protection General Numerical protection relays operate on the basis of sampling inputs and controlling outputs to protect or control the monitored system. can be taken as necessary by the protection relay. normally the relays either lock out or attempt a recovery. the final value may be calculated many times within a single sampling cycle. the appropriate comparison to a setting. calculations are performed to convert the incremental sampled values into a final value that represents the associated input quantity based on a defined algorithm. This feature • • • . The proper handling of all the features requires a flexible programmable logic platform for the user to apply the available functions with complete flexibility and be able to customise the protection to meet the requirements of the protected power system. Programmable I/O. depending on the disturbance detected. which includes memory checks and analogue input module tests. Once the final value of an input quantity can be established. In case of failure.4.4. and other system design or protection requirements.

– ROM (read only memory) or PROM (programmable ROM). The main modules are as follows: • Microprocessor: responsible for processing the protection algorithms. 3. The algorithms for relay settings are usually in low-level languages because of the need for a short time response which is not obtained with high-level languages such as Pascal or Fortran. Figure 3.40 Protection of electricity distribution networks enables the relay setting to be changed depending on the operating conditions of the network. including retaining the incoming data that is input to the processor and is necessary for storing information during the compilation of the protection algorithm. Input module: the analogue signals from the substation are captured and sent to the microprocessor and the module typically contains the following • Phase A Filter and analog to digital converter Display Keypad Isolation Inputs Phase B Phase C Micro-processor Phase N Outputs Power supply Com port Figure 3.4. thus guaranteeing suitable relay settings for the real-time situation by not using a setting based on the most critical system arrangement.3 Typical architectures of numerical relays Numerical relays are made up from modules with well-defined functions. which are used for storing programs permanently. It includes the memory module which is made up from two memory components: – RAM (random access memory).8 General arrangement of numerical relays (reproduced by permission of Basler Electric) . which sometimes does not provide the most appropriate solution. which has various functions.8 shows a block diagram for numerical relays using typical modules.

It integrates the multifunction protection. Communication module: contains series and parallel ports to permit the interconnection of the protection relays with the control and communications systems of the substation. which converts the normalised DC signal into a binary number that is then sent directly to the microprocessor or to a communications buffer. breaker monitoring and automatic reclosing. . breaker failure. control. directional/nondirectional earth-fault overcurrent. 3. and input/output elements into a unique protection and control system. distance. and indicator lights. switches.Classification and function of relays 41 • • elements: – analogical filters. In the digital multifunction environment the process of wiring individual protection or control elements is replaced by entering logic settings. reactive power and true power.and under-voltage. diagrams provide information on wiring protection elements. automatic reclosing. – analogue digital convertor. The process of creating a logic scheme is the digital equivalent of wiring a panel. differential. over. frequency. current. negative-sequence overcurrent. over. In the single-function and electromagnetic environment. It is made up of a digital output which generates a pulse as a response signal.4. meters.4 Standard functions of numerical relays Numerical multifunction relays are similar in nature to a panel of single-function protection relays. Metering functions include voltage. Threephase currents and voltages are digitally sampled and the fundamental is extracted using a discrete Fourier transform (DFT) algorithm. apparent power. Both must be wired together with ancillary devices to operate as a complete protection and control system. – signal conditioner. and a signal conditioner which amplifies and isolates the pulse. which converts the signal from the CTs into a normalised DC signal. power factor.and under-frequency. directional power. Metered values are viewed through any communication port using serial commands or at the front panel HMI if available. Output module: conditions the microprocessor response signals and sends them to the external elements that they control. (ii) Measurement Numerical relays normally incorporate outstanding measurement functions. (i) Protection Protection functions of numerical relays may typically include one or more of the following: directional/non-directional three-phase overcurrent. field loss. which are active low-bandpass filters that eliminate any background noise that has been induced in the line. over-excitation.

and perform control operations. DNP. Panel communication ports can be connected to computers. Most numerical relay communication protocols support ASCII and binary data transmissions. (iv) Communication In numerical relaying. relay and power system information can be retrieved from a remote location using the ASCII command interface which can also be used to enter settings. In most numerical relays. IEC 608750-5 and MMS/UCA2. A pick-up expression is also used by the setting group selection function to prevent a setting group change during a fault. time the length of the fault from pick-up to drop-out (fault clearance time) and to control the recording of oscillograph data. the settings are introduced as logic equations and trip expressions are used by the fault reporting function to start logging targets for an event and to record the fault current magnitudes at the time of trip. The virtual switches can be used to trip and close additional switches and breakers. The HMI also uses the pick-up expression to control the flashing of a trip LED. The most basic fault reporting functions provided by a protection relay are the signalling or visual flags which indicate the type of fault. or enable and disable certain functions. Binary data is used for computer communication and transmission of raw oscillographic fault data if available. The breaker monitoring function also uses the trip expression to start counting the breaker operating time. sequence of events recorder reports. at least one of the following protocols is available – Modbus. which can be opened by numerous programs. Pick-up expressions are used by the fault reporting function to time-stamp the fault summary record. (v) Reporting and alarms The fault reporting functions provide means of recording and reporting information about faults that have been detected by the relay. .42 Protection of electricity distribution networks (iii) Control Most numerical relays incorporate at least one virtual breaker control switch and several virtual switches which can be accessed locally from the HMI or remotely from the communications port. and oscillographic records. usually referred to as targets. Courier. These include fault summary reports. ASCII data is used to send and receive human readable data and commands. A communication port on the relay front panel provides a temporary local interface for communication. The HMI uses the trip expression to display the respective target by means of the associated LED. retrieve reports and metering information. terminals. the numerical relay can provide many advanced fault reporting features. modems. Above all. and logic intermediate communication/control interfaces such as RS-232 serial multiplexors. The virtual breaker control switch permits the tripping and closing of a selected breaker. serial printers. Communication ports on the rear panel provide a permanent communication interface. it is essential to be able to download information to COMTRADE files. In addition. In numerical relays.

Various alarm and control signals (for example. In the case of the so-called primary relays the connection is made directly to the supply system.5 Supplies to the relay circuits Protection relays are usually designed for either alternating current or direct current circuits. those used to open switches) are carried along DC circuits. the two types generally being totally independent of each other. Voltage and/or current signals are taken from measurement transformers to feed the AC circuits. The AC signals feed the control circuits of the relays which then determine whether or not fault conditions exist. These circuits normally obtain their supply from banks of batteries so that faults on the AC system do not affect the operation of the switchgear mechanisms. .Classification and function of relays 43 3.

.

In order to achieve this.1a. depending on whether their connection is line-to-line or line-to-neutral. control and measurement equipment from the high voltages of a power system. 4. When only voltage or current magnitudes are required to operate a relay then the relative direction of the current flow in the transformer windings is not important. factors such as the transient period and saturation must be taken into account when selecting the appropriate transformer.1. The majority of protection relays have nominal voltages of 120 or 69 V. the polarity must be kept in mind when the relays compare the sum or difference of the currents.1 Voltage transformers With voltage transformers (VTs) it is essential that the voltage from the secondary winding should be as near as possible proportional to the primary voltage. The behaviour of current and voltage transformers during and after the occurrence of a fault is critical in electrical protection since errors in the signal from a transformer can cause mal-operation of the relays. and 120 V for the voltage coils.Chapter 4 Current and voltage transformers Current or voltage instrument transformers are necessary to isolate the protection. VTs are designed in such a way that the voltage drops in the windings are small and the flux density in the core is well below the saturation value so that the magnetisation current is small. in this way a magnetisation impedance is obtained that is practically constant over the required voltage range. However. The secondary voltage of a VT is usually 115 or 120 V with corresponding line-to-neutral values.1 Equivalent circuit VTs can be considered as small power transformers so that their equivalent circuit is the same as for power transformers. In addition. as shown in Figure 4. The magnetisation . 4. and for supplying the equipment with the appropriate values of current and voltage – generally these are 1 A or 5 A for the current coils.

1 Voltage transformer equivalent circuits: (b) simplified circuit (a) equivalent circuit. especially for those values close to the nominal system voltage.2. due to the problems of having to cope with a variety of different relays. 4. although the precision requirements of a VT for protection applications are not so high at nominal voltages. The vector diagram for a VT is given in Figure 4. VTs have an excellent transient behaviour and accurately reproduce abrupt changes in the primary voltage. Vp Vcd = n Vs = Vef IL jIL(XH /n2 + XL) IL(RH /n2 + RL) Figure 4. The secondary voltage Vs lags the voltage Vp /n and is smaller in magnitude.1b.2 Vector diagram for a voltage transformer branch can be ignored and the equivalent circuit then reduces to that shown in Figure 4. for example for billing and control purposes.46 Protection of electricity distribution networks (a) a Vp b (b) a Vp b d d n:1 c Vp n Rm Xm f ZH/n2 c Vp n IL Vs f ZB ZL e ZH/n2 ZL e Vs ZB n:1 Figure 4. the accuracy of a VT is important. the nominal maximum errors are relatively small. secondary wiring burdens . In spite of this. with the length of the voltage drops increased for clarity. Notwithstanding this.2 Errors When used for measurement instruments.1.

and the nearest service voltage in mind. errors in a VT are due to differences in magnitude and phase between Vp /n and Vs . If the error is positive. or between phase and earth. and the value of volt-amperes (VA). then the secondary voltage exceeds the nominal value. This range should be between 5 and 173 per cent of the nominal primary voltage for VTs connected between line and earth. or when it is necessary to measure the voltage and power factor of each phase separately. errors should be contained within narrow limits over a wide range of possible voltages under fault conditions. and the primary voltage V1 . it is usual to add together all the nominal VA loadings of the apparatus connected to the VT secondary winding.1 gives standard burdens based on ANSI Standard C57.1. is to use a capacitor voltage . The nominal primary voltage of a VT is generally chosen with the higher nominal insulation voltage (kV). The voltage error is the percentage difference between the voltage at the secondary terminals. and a more economic solution. V2 . caused by the drop in voltage from the circulation of the magnetisation current through the primary winding. the cost increases in a similar manner to that of a high voltage transformer.1a. 4. 4. One alternative.1. Table 4. and errors due to voltage drops as a result of the load current IL flowing through both windings. In order to select the nominal power of a VT. These consist of the errors under opencircuit conditions when the load impedance ZB is infinite.4 Selection of VTs Voltage transformers are connected between phases. The phase error is considered positive when the secondary voltage leads the primary voltage.2. for this reason. the size of an inductive VT is proportional to its nominal voltage and. especially if the distance between the transformers and the relays is large. multiplied by the nominal transformation ratio.1. The nominal secondary voltages are generally standardised at 115 and 120 V.5 Capacitor voltage transformers In general.13. Errors in magnitude can be calculated from ErrorVT = {(nVs − Vp )/ Vp } × 100%. Referring to the circuit in Figure 4. Voltage transformers are specified in IEC publication 186A by the precision class.Current and voltage transformers 47 and the uncertainty of system parameters. The connection between phase and earth is normally used with groups of three singlephase units connected in star at substations operating with voltages at about 34.5 kV or higher. it is important to take account of the voltage drops in the secondary wiring.3 Burden The standard burden for voltage transformers is usually expressed in volt-amperes (VA) at a specified power factor. The allowable error limits corresponding to different class values are shown in Table 4. 4. In addition. where Vn is the nominal voltage.

85 0.0403 1.09 0.034 0.101 0.10 0.2 163.356 inductance (H) impedance ( ) resistance ( ) inductance (H) Characteristics for 69.4 54.1 Characteristics for 120 V and 60 Hz resistance ( ) 115.268 0.364 0.01 0.00 0.070 1152 575 192 72 36 411 38.70 0.089 0.20 .00 200.4 1.00 400.00 35.0168 0.2 403.85 0.3 V and 60 Hz Standard burdens for voltage transformers Standard burden Protection of electricity distribution networks design voltamperes power factor impedance ( ) 384 192 64 24 12 137 W X Y Z ZZ M 12.04 1.00 75.2 31.3 3.4 10.2 27.4 134.2 61.48 Table 4.4 20.85 0.50 25.2 82.

For improved accuracy a high voltage capacitor is used in order to obtain a bigger voltage at the point of connection.0 40. under stable system conditions the capacitor VT acts like a conventional transformer. in addition. Ie is small compared to Is so that the vector difference between . while Vs and Is represent the secondary voltage and current. The capacitor divider differs from the inductive divider in that the equivalent impedance of the source is capacitive and the fact that this impedance can be compensated for by connecting a reactance in series at the point of connection. Ri and Rs are not large and.2 0.0 1.4 is the same as the equivalent circuit of a power transformer. 1. It can be seen that.1 0.0 40. in an actual situation on a network some resistance is always present.0 40.2 Class 49 Voltage transformer error limits Voltage error (±%) 0.4 in which Vi is equal to the nominal primary voltage. A simplified equivalent circuit of a capacitor VT is shown in Figure 4.0 80. respectively.0 0.0 2.0 80.1 0.0 0. the circuit in Figure 4.5 1.0 3.05Vn Vn transformer.3. Ri represents the resistance of the primary winding of transformer T plus the losses in C and L.0 40.2 0.5 1.5 1.0 80. This device is effectively a capacitance voltage divider and is similar to a resistive divider in that the output voltage at the point of connection is affected by the load – in fact the two parts of the divider taken together can be considered as the source impedance which produces a drop in voltage when the load is connected.0 Primary voltage 0. and Ze is the magnetisation impedance of transformer T .0 1. The divider can reduce the voltage to a value that enables errors to be kept within normally acceptable limits. Therefore. L is the resonance inductance.1 0. with the exception of C.1 0.Current and voltage transformers Table 4. which can be reduced to a standard voltage using a relatively inexpensive transformer as shown in Figure 4. the resistance of the secondary circuit and the load impedance are represented by Rs and ZB .0 1.5 10.2 2.2 0.2Vn 0.0 2.0 80.5 1.0 0.0 120.0 0. Referred to the intermediate voltage. C is the numerically equivalent impedance equal to (C1 + C2 ).2 0.0 20.0 Phase error (± min) 0.8Vn . With an ideal reactance there are no regulation problems – however.0Vn and 1. at the system frequency when C and L are resonating and cancelling out each other.

which is drawn for a power factor close to unity.3 Ii Capacitor VT basic circuit C EC L EL VC2 Ri Ie RS IS VS Vi Ze ZB Figure 4.5. This is illustrated in the vector diagram shown in Figure 4.4 Capacitor VT equivalent circuit IiRi VC2 EL IS RS EC VS I Ie S Vi Ii Figure 4. The voltage error is the difference in magnitude between Vi and Vs .50 Protection of electricity distribution networks C1 L Vp C2 VC2 VS T VS ZB Figure 4. whereas the phase error is indicated by the angle . is very small. From the diagram it can be . which constitutes the error in the capacitor VT.5 Capacitor VT vector diagram Vi and Vs .

in general. In all these cases the CT should be able to supply sufficient current so that the relay operates satisfactorily. C and the transformer T . provide better protection. Capacitor VTs have a better transient behaviour than electromagnetic VTs since the inductive and capacitive reactances in series are large in relation to the load impedance referred to the secondary voltage and thus. for frequencies different from the resonant frequency.1 Equivalent circuit An approximate equivalent circuit for a CT is given in Figure 4. 4.6a. where n2 ZH represents the primary impedance ZH referred to the secondary side. (a) a 1:n c ZB b d Rm Xm f (b) IH a 1:n c Es b d IE IH/n Xm f RL e IL Vs ZB n2ZH ZL e Figure 4. the values of EL and EC predominate.Current and voltage transformers 51 seen that.2. causing serious errors in magnitude and phase. care should be taken to ensure that under the most critical faults the CT operates on the linear portion of the magnetisation curve. CTs can become saturated at high current values caused by nearby faults. when the primary voltage collapses. to avoid this.2 Current transformers Even though the performance required from a current transformer (CT) varies with the type of protection.6 Current transformer equivalent circuits . the secondary voltage is maintained for some milliseconds because of the combination of the series and parallel resonant circuits represented by L. Good quality CTs are more reliable and result in fewer application problems and. high grade CTs must always be used. 4. The quality of CTs is very important for differential protection schemes where the operation of the relays is directly related to the accuracy of the CTs under fault conditions as well as under normal load conditions. and the secondary impedance is ZL . Rm and Xm represent the losses and the excitation of the core.

The magnitude error is the difference in magnitude between IH /n and IL and is equal to Ir .7 Vector diagram for the CT equivalent circuit The circuit in Figure 4. It should be noted that a moderate inductive load. the primary impedance of a CT does not have the same influence on the accuracy of the equipment – it only adds an impedance in series with the line.2 Errors The causes of errors in a CT are quite different to those associated with VTs. when the . the primary current referred to the secondary side. The values of the magnitude and phase errors depend on the relative displacement between Ie and IL . The vector diagram. the component of Ie that is in quadrature with IL . is directly proportional to the secondary current. Thus. it is essential to measure or calculate the excitation curve. which can be ignored.7. Es . In effect. so that Ie is the principal source of error. The phase error. so much so that. is related to Iq . with the voltage drops exaggerated for clarity. 4. The current flowing through Xm is the excitation current Ie . represented by . with Ie and IL approximately in phase.7). 4. but neither of them can exceed the vectorial error Ie . the component of Ie in line with IL (see Figure 4.52 Protection of electricity distribution networks Vs ILRL Es IL Ie IH/n Ie Ir Iq Figure 4. for a given CT.3 AC saturation CT errors result from excitation current. From this it can be concluded that. The magnetisation current of a CT depends on the cross section and length of the magnetic circuit.2.6b. In general ZL is resistive and Ie lags Vs by 90◦ . Note that the net effect of Ie is to make IL lag and be much smaller than IH /n. since it does not influence either the current IH /n or the voltage across Xm . it can be seen that the voltage across the magnetisation impedance.6b where ZH can be ignored. in order to check if a CT is functioning correctly.6a can be reduced to the arrangement shown in Figure 4. the number of turns in the windings. The errors are principally due to the current that circulates through the magnetising branch.2. and the magnetic characteristics of the material. has a small phase error and the excitation component results almost entirely in an error in the magnitude. and referring to the equivalent circuit of Figure 4. is shown in Figure 4.

the excitation current should be measured at various values of voltage – the so-called secondary injection test. from the primary current. and CTR = current transformer transformation ratio. Figure 4.3. If the impedance of the magnetic circuit.8a shows the typical relationship between the secondary voltage and the excitation current. These factors can be assessed from: • • • formulae. IEC Standard Publication 185(1987) specifies CTs by the class of accuracy followed by the letter M or P.2. leaving the primary winding open-circuited. Those CT classes marked with ‘ext’ denote wide range (extended) current transformers with a rated continuous current of 1. The phase error is considered positive when the secondary current leads the primary current. By way of example. CT magnetisation curves.4a and 4. The third only provides a qualitative estimation. determined in this way. 4. {(CTR × I2 ) − I1 } ÷ I1 (%).2 or two times the nameplate current rating. When investigating the behaviour of a CT. The current error is the percentage deviation of the secondary current multiplied by the nominal transformation ratio.4 Burden The burden of a CT is the value in ohms of the impedance on the secondary side of the CT due to the relays and the connections between the CT and the relays. This point is referred to in the ANSI/IEEE standards as the intersection of the excitation curves with a 45◦ tangent line as indicated in Figure 4. Usually it is more convenient to apply a variable voltage to the secondary winding. The current and phase error limits for measurement and protection CTs are given in Tables 4. I2 = secondary current (A). Xm . where I1 = primary current (A)..13. based on ANSI Standard C57.8b. The European knee point is at a higher voltage than the ANSI/IEEE knee point. these currents reach a point when the core commences to saturate and the magnetisation current becomes sufficiently high enough to produce an excessive error. i.e. is . the standard burdens for CTs with a nominal secondary current of 5 A are shown in Table 4. In European standards the point Kp on the curve is called the saturation or knee point and is defined as the point at which an increase in the excitation voltage of ten per cent produces an increase of 50 per cent in the excitation current. The first two methods provide precise facts for the selection of the CT. which denotes if the transformer is suitable for measurement or protection purposes respectively. 4.4b. it is important to ensure that the fault level and normal load conditions do not result in saturation of the core and that the errors do not exceed acceptable limits. The secondary voltage Es in Figure 4. CT classes of accuracy.Current and voltage transformers 53 primary current and therefore the secondary current is increased.5 Selection of CTs When selecting a CT.2.6b has to be determined for all three methods.

Below this line the exciting current for a given voltage for any unit will not exceed the curve value by more than 25% 45° 1200:5 1000:5 900:5 800:5 600:5 500:5 400:5 300:5 200:5 100:5 B A 10 0.31 600:5 120:1 0.10 200:5 60:1 0. (b) typical excitation curves for a multi-ratio class C CT ( from IEEE Standard C576. Figure 4. reproduced by permission of the IEEE) .0 10 100 Secondary excitation current (Ie).15 300:5 80:1 0.41 800:5 160:1 0. V r.s.51 1000:5 200:1 0.1 1.61 1200:5 240:1 * OHMS at 75°C.s. A r.01 0.20 400:5 0.46 900:5 180:1 0.m.m.05 100:5 40:1 0.25 500:5 100:1 0.54 (a) Protection of electricity distribution networks Volts Secondary excitation voltage Vk 10% Vk Kp 50% Iek Iek mA Secondary excitation current (b) 1000 Secondary excitation voltage (Vef).8 CT magnetisation curves: (a) defining the knee point in a CT excitation curve according to European standards.13-1978. 100 Current Turns Sec* ratio ratio res 20:1 0.0001 0.

5 B-1 B-2 B-4 B-8 Table 4.5 0.5 1.2 1.5 1.0 0.5 0.5 – 0.0 3.5 1.0 – 1.25 – 0.0 8.20 0.4 Impedance ( ) 1.0 – Phase error (minutes) at proportion of rated primary current shown 1.50 2.75 0. .0 3.0 60 120 0.35 0.0 4.2 ext 0.5 0.5 – 120 0.5 – 3.5 – 0.0 0.1 0.0 0.1 8 15 45 90 – 8 15 45 90 – 0.3 4.5 – 0.0 – – – 0.0 Volt-amps (at 5 A) 25 50 100 200 Power factor 0.20 0.5 5 10 30 60 – 5 10 30 60 – – – – – 120 – – – – 120 0.1 ext 0.00 – – 3.0 2.00 – 1.4b Class Error limits for protection current CTs Current error (%) at proportion of rated primary current shown 1.Current and voltage transformers Table 4.75 1.2 0.00 – – 3.40 0.1 0.0 Inductance (mH) 2.4a Class Error limits for measurement current transformers % phase error at the given proportion of the rated current shown below 2.0∗ 1.2 1.0 0.2 0.5 ext 1.1 0.3 Designation 55 Standard burdens for protection CTs with 5 A secondary current Resistance ( ) 0.05 % current error at the given proportion of rated current shown below 2.50 – 0.2 – 0.0 ext 3.0 ext – – – – – 0.5 1.2 90 – 0.0 Total error for nominal error limit current and nominal load is 5% for 5P and 5P ext CTs and 10% for 10P and 10P ext CTs.0 3.50 0.1 – 0.1 2.0 – – – 10 – 20 – 60 – 120 – – – 10 15 20 30 60 90 120 – – – ∗ ext = 200% Table 4.1 0.2 18.05 0.0 4.00 1.0 – – – – – 0.0 0.50 2.0 3.0 2.6 9.2 0.75 1.2 0.2 – 0.35 0.0 – 1.50 1.2 0.0 0.1 120 – 5P and 5P ext 10P and 10P ext 1.2 – – – – – 5 10 30 60 120 5 10 30 60 120 – – – – – 1.1 – 0.5 0.0∗ 1.25 0.

1 × 108 = 70 030 lines/in2 4. The cross-sectional area of metal and the saturation flux density are sometimes difficult to obtain. giving Es = Vs .2. A = cross-sectional area of core (in2 ). The method consists .s. this can be removed from the equivalent circuit with little error.5 × (0.m. there could be appreciable errors in the secondary current and the CT selected would not be appropriate. and a secondary winding with a resistance of 0.31 + 2) = 202.1 V. 4. The curves give the magnitude of the excitation current required in order to obtain a specific secondary voltage.25 × 400 Since the transformer in this example has a steel core of high permeability this relatively low value of flux density should not result in saturation. 4.1.m.2. including connections.2) where: f = frequency in Hz. ZB = external impedance connected. Bmax = flux density (lines/in2 ).44 × 50 × 3. N = number of turns. state the r. is 35 000 × 5/2000 = 87.s. and thus: Vs = IL (ZL + ZC + ZB ) (4. which are supplied by manufacturers. Using eqn. If Bmax exceeds the saturation density. and Vs = 87. Bmax can now be calculated: Bmax = 202. N = 2000/5 = 400 turns. Using the magnetisation curve Typical CT excitation curves. ZL = impedance of the secondary winding.m. Use of the formula This method utilises the fundamental transformer equation: Vs = 4. with the primary winding open-circuited. and Bmax is then calculated using eqn.31 . Vs is determined from eqn.25 in2 . is 2 . having a steel core of high permeability.1 Assume that a CT with a ratio of 2000/5 is available. which is a typical value for modern transformers. voltage to the secondary winding. IL .56 Protection of electricity distribution networks high. Solution If the CT is not saturated. IL = maximum secondary current in amperes (this can be determined by dividing the maximum fault current on the system by the transformer turns ratio selected). The latter can be taken as equal to 100 000 lines/in2 . ZC = impedance of the connecting wiring. The impedance of the relays. then the secondary current.1) where: Vs = r.s voltage induced in the secondary winding. 4. current obtained on applying an r. Determine whether the CT would be saturated by a fault of 35 000 A at 50 Hz. Example 4.44f ANBmax 10−8 volts (4. a cross-sectional area of 3.5 A. To use the formula.

9. Ie . This method incurs an error in calculating IH /n by adding Ie and IL together arithmetically and not vectorially. By joining the points together the curve of IL against IH is obtained. which implies not taking account of the load angle and the magnetisation branch of the equivalent circuit. (d) Calculate IH /n (= IL + Ie ) and multiply this value by n to refer it to the primary side of the CT. e: draw the point on the curve of producing a curve that shows the relationship between the primary and secondary currents for one tap and specified load conditions. then a value of IH can be obtained that is closer to the fault current. The process is summarised in the following steps: (a) Assume a value for IL . After constructing the curve it should be checked to confirm that the maximum primary fault current is within the transformer saturation zone. the value of the corresponding primary current can be determined.1. and find the associated value of the magnetisation current. In practice it is not necessary to draw the complete curve because it is sufficient to take the known fault current and refer to the secondary winding. c: find Ie from the curve. then it will be necessary to repeat the process. this error is not great and the simplification makes it easier to carry out the calculations. If not. If the tap is found to be suitable after finishing the calculations. b: Vs = IL (ZL + Zc + ZB ).Current and voltage transformers Vs Tap 4 Tap 3 Tap 2 Tap 1 IL 57 b a e c Ie d IH Figure 4. assuming that there is no saturation for the tap selected. . This converted value can be taken as IL initially for the process described earlier. The process is then repeated to obtain other values of IL and the resultant values of IH . 4. and with the help of the magnetisation curves. (e) This provides one point on the curve of IL versus IH .9 Using the magnetisation curve: a: assume a value for IL . (c) Locate the value of Vs on the curve for the tap selected. However. d: IH = n(IL + Ie ). (b) Calculate Vs in accordance with eqn. Starting with any value of secondary current. changing the CT tap until the fault current is within the linear part of the characteristic. such as shown in Figure 4.

For the example given.2 . on the occurrence of a fault. C indicates that the transformation ratio can be calculated.4 and the power factor is 0. the voltage drops in the secondary can be ignored only if the current does not exceed 100 A. Example 4. Notwithstanding this. Determine if.33 ). and in consequence it can only feed a portion of the load without exceeding the ten per cent error limit. each tap will have a voltage capacity proportionally smaller. For example. The nominal CT ratio is 1200/5 and the CT is to be used with a tap of 800/5.13) is described by two symbols – a letter and a nominal voltage. while T indicates that the transformation ratio can be determined by means of tests. is more than the maximum permissible (1. the resistance of the secondary is 0. then the error could exceed ten per cent during a fault of 12 000 A. The classification C includes those CTs with uniformly distributed windings and other CTs with a dispersion flux that has a negligible effect on the ratio. the total secondary load is 2. The CT class is C-200. the class C-200 indicates that the CT could withstand 200 V plus the drop produced by the resistance of the secondary with a current range equal to 20 times the nominal value and with a load power factor as low as 0.4 . NP is the fraction of the total number of turns being used and VC is the ANSI voltage capacity for the complete CT.2 The maximum fault current in a given circuit is 12 000 A. by definition. IL = 12 000 × (5/800) = 75 A.58 Protection of electricity distribution networks Accuracy classes established by the ANSI standards The ANSI accuracy class of a CT (Standard C57. it is necessary to . The classification T includes those CTs whose dispersion flux considerably affects the transformation ratio. these define the capability of the CT. with a CT of class C-100. Consequently. The permissible load is given by ZB = (NP VC ) ÷ 100 NP = 800/1200 = 0. 2. within defined limits.5. These accuracy classes are only applicable for complete windings.6. When considering a winding provided with taps. the ratio can be calculated and the error should not exceed ten per cent if the secondary current does not go outside the range of 1 to 20 times the nominal current and if the load does not exceed 1 (1 × 5 A × 20 = 100 V) at a minimum power factor of 0. The permissible load is defined as ZB = (NP VC )/100. Solution The resistance of the secondary winding of the CT can be ignored since. the error will exceed ten per cent.334 Since the loading of the circuit.667 so that ZB = (0.5. where ZB is the permissible load for a given tap of the CT.667 × 200 V) ÷ 100 A = 1. which results in a maximum secondary current of 75 A.

I = secondary symmetrical current (amperes − r.s. high fault currents circulate through the CTs. the behaviour of a CT has been discussed in terms of a steady state.7 Precautions when working with CTs Working with CTs associated with energised network circuits can be extremely hazardous. where: VK = voltage at the knee point of the magnetisation curve.28 IRT . If saturation occurs in different CTs associated with a particular relay arrangement.).10 The effect of DC saturation on the secondary current . However. the DC component has more influence in producing severe saturation than the AC component.10 shows an example of the distortion and reduction in the secondary current that can be caused by DC saturation. Rp = resistance of the primary circuit. DC saturation is particularly significant in complex protection schemes since. determined by the extension of the straight part of the curve. this could result in the circulation of unbalanced secondary currents that would cause the system to malfunction. 4. 4. increase the current transformer tap or use another CT of a higher class. f = frequency.Current and voltage transformers 59 reduce the load.m.6 DC saturation Up to now. Figure 4.2. opening the secondary circuit of a CT could result in dangerous Secondary current without saturation Secondary current with saturation Figure 4. i. T = DC time constant of the primary current in cycles. R = total resistance of the secondary. However. the DC component of the fault current does not produce saturation of the CT if VK ≥ 6. In particular. in the case of external faults.2.e. without considering the DC transient component of the fault current. Lp T= f Rp where: Lp = inductance of the primary circuit.

Associated with this circuit is a 500/5 CT feeding a measurement system whose total load is 10 VA.2 kV feeder.11. The equivalent circuit of the CT referred to the secondary side is shown in Figure 4.2 Ω j50 Ω 150 Ω Figure 4.12 Single-line diagram for Example 4. Example 4. referred to the secondary side. the equivalent primary-circuit impedance is almost unaffected but a high voltage will be developed by the primary current passing through the magnetising impedance.3 .3 Zload V = 13200/ 3 500/5 Figure 4. Solution The single line diagram is given in Figure 4.3 Consider a 13. which might harm operational staff or lead to equipment being damaged because the current transformers are designed to be used in power circuits that have an impedance much greater than their own.2 kV feeder that is carrying a load of 10 MVA at 1.12 and the equivalent circuit in Figure 4. an example is given next using typical data for a CT and a 13.13. 0.0 power factor. As a consequence.60 Protection of electricity distribution networks overvoltages. when secondary circuits are left open. Thus.11 CT equivalent circuit. Calculate the voltage that would occur in the secondary circuit of the CT if the measurement system was accidentally opened. To illustrate this. for Example 4. secondary circuits associated with CTs must always be kept in a closed condition or short-circuited in order to prevent these adverse situations occurring.

12 Referring the values to the secondary side of the CT gives 13200 500 V= √ × = 762102.Current and voltage transformers 174240 Ω 0.4 Ω Figure 4. the current is only able to circulate across the shunt branch.4 174240 + 0.47 71. . In these conditions the voltage that appears at the terminals of the CT is VCT = 762102.36 A × 0.55◦ V 174240 + (150 j50) Therefore the voltage increases by almost 120 times. ignoring the shunt branch.22 500 × 10 5 10 = 0.13 Equivalent circuit for Figure 4.2 Ω 61 V = 762102.36 V 5 3 Zload = 13. the voltage across the measurement system can be calculated approximately. as Vmeter = 762102.2 + 0.3V j50 Ω 150 Ω 0.75 V If the secondary circuit is opened.36 × (150 j50) = 207.37 A × 0.4 = 1.4 = 4.4 52 2 = 174240 Zmeter = When the secondary circuit is closed.

.

especially for protecting lines and distribution transformers. which also illustrates the combination of an instantaneous relay with one having an inverse time characteristic. . which form the basis of this chapter. The first two types have simple operating arrangements and are principally used in the protection of low voltage equipment. The setting is chosen so that. overcurrent relays can be classified into three groups: definite current or instantaneous. at the substation furthest away from the source. and overcurrent relays. Fuses are also often used at low voltages. are the most common form of protection used to deal with excessive currents on power systems.2 Types of overcurrent relay Based on the relay operating characteristics.Chapter 5 Overcurrent protection 5. Among the more common types of protection are thermomagnetic switches.1 Definite-current relays This type of relay operates instantaneously when the current reaches a predetermined value. 5.2. They should not be installed purely as a means of protecting systems against overloads – which are associated with the thermal capacity of machines or lines – since overcurrent protection is primarily intended to operate only under fault conditions. Overcurrent relays. The characteristic curves of these three types are shown in Figure 5. and inverse time. moulded-case circuit breakers (MCCBs). However. which can vary in design depending on the complexity and accuracy required.1. the relay settings that are selected are often a compromise in order to cope with both overload and overcurrent conditions. fuses.1 General Very high current levels in electrical power systems are usually caused by faults on the system. These currents can be used to determine the presence of faults and operate protection devices. definite time. 5.

1 Time/current operating characteristics of overcurrent relays .64 Protection of electricity distribution networks t Definite current A t t1 Definite time A t Inverse time t A Inverse time with instantaneous unit A Figure 5.

Overcurrent protection 65 the relay will operate for a low current value and the relay operating currents are progressively increased at each substation. Isc(B) = VS / 3(ZS + ZR ) . Thus. leading to the possibility of poor discrimination. (a) ZR = impedance of √ protected element. Isc(A) = VS / 3 × √ 1/ZS .2a illustrates the effect of the source impedance on the short-circuit level at a substation. moving towards the source. From Figure 5.2 Illustration of different levels of fault current. This type of protection has the drawback of having little selectivity at high values of short-circuit current. Figure 5.2b it can be appreciated that the fault currents at F1 and F2 are almost the same. and it is this that makes it difficult to obtain correct settings for the relays. and for a fault at point B down the line. Another disadvantage is the difficulty of distinguishing between the fault current at one point or another when the impedance between these points is small in comparison to the impedance back to the source. the relay with the lower setting operates first and disconnects load at the point nearest to the fault. When there is some (a) ZS A ZR B VS/ 3 (b) F3 F2 F1 Figure 5. (b) ZS = source impedance.

The settings can be adjusted in such a way that the breaker nearest to the fault is tripped in the shortest time. may be cleared in a relatively long time. Inversetime relays are generally classified in accordance with their characteristic curve that indicates the speed of operation. and then the remaining breakers are tripped in succession using longer time delays.2. 5. 5. but their use as an instantaneous unit is common where other types of protection are in use.66 Protection of electricity distribution networks considerable impedance between F1 and F2 . much shorter tripping times can be obtained without risk to the protection selectivity. based on this they are commonly defined as being inverse. The difference between the tripping times for the same current is called the discrimination margin. Their advantage over definite-time relays is that. this could result in some breakers operating unnecessarily if the fault level increases. If the protection settings are based on maximum fault level conditions. Since the operating time for definite-time relays can be adjusted in fixed steps. the protection is more selective. Inverse-time relays are also referred to as inverse definite minimum time or IDMT overcurrent relays. As a consequence.3 Setting overcurrent relays Overcurrent relays are normally supplied with an instantaneous element and a timedelay element within the same unit. very inverse. even though these two points are physically close. as illustrated by the characteristic curves shown later. there will be a considerable difference between the currents for faults at F2 and F3 . definite current relays are not used as the only overcurrent protection. if a lower value of fault level is used when calculating the relay settings. which result in bigger currents. then the fault current at F1 will be less than at F2 . However. plus a time dial setting to obtain the exact timing of the relay operation. then these settings may not be appropriate for the situation when the fault level is lower. 5. due to the impedance of the transformer. or extremely inverse. for very high currents. When electromechanical relays were more . Similarly.3 Inverse-time relays The fundamental property of these relays is that they operate in a time that is inversely proportional to the fault current.2.2 Definite-time/current or definite-time relays This type of relay enables the setting to be varied to cope with different levels of current by using different operating times. The big disadvantage with this method of discrimination is that faults near to the source. It should be noted that the time-delay setting is independent of the value of the overcurrent required to operate the relay. moving back towards the source. for example when the fault F1 is located down a long line. These relays are used a great deal when the source impedance is large compared to that of the power system element being protected when fault levels at the relay position are similar to those at the end of the protected element. This type of relay has a current or pick-up setting – also known as the plug or tap setting – to select the value at which the relay will start.

The more modern microprocessor protection has a three-phase overcurrent unit and an earth-fault unit within the same case. The criteria for setting instantaneous units vary depending on the location. Although the two processes are similar.m. the three-phase short-circuit current should be used for setting the phase relays while the phase-to-earth fault current should be used for the earth-fault relays. They offer two fundamental advantages: • • they reduce the operating time of the relays for severe system faults. The t A Figure 5.3 Preservation of selectivity using instantaneous units . current for the maximum fault level at the next substation. 5. When calculating the fault currents. (i) Lines between substations The setting of instantaneous units is carried out by taking at least 125 per cent of the symmetrical r. this is obtained by setting the instantaneous units so that they operate before the relay characteristics cross. once for the phase relays and then repeated for the earth-fault relays. as shown in Figure 5. The same procedure can be applied to multiple circuit arrangements. and the type of system element being protected. Setting overcurrent relays involves selecting the parameters that define the required time/current characteristic of both the timedelay and instantaneous units. which is possible with numerical relays for example – then better discrimination is obtained if the calculations are carried out on the basis of each one of the transformers being out of service in turn. they avoid the loss of selectivity in a protection system consisting of relays with different characteristics. distribution lines.3. and transformers. as indicated earlier. This process has to be carried out twice. the overcurrent protection was made up from separate single-phase units.Overcurrent protection 67 popular.3. However. the power system is assumed to be in its normal operating state.s. at a busbar that has two or more transformers connected in parallel and protected with relays that do not have the facility of multiple setting groups – the ability to be adjusted to accommodate the prevailing system conditions. Three groups of elements can be defined: lines between substations.1 Setting instantaneous units Instantaneous units are more effective when the impedances of the power system elements being protected are large in comparison to the source impedance.

referred to the high voltage side. thus making it difficult to obtain correct co-ordination. (ii) Distribution lines The setting of the instantaneous element of relays on distribution lines that supply only pole-mounted MV/LV transformers is dealt with differently to the previous case. They therefore do not have to fulfil the co-ordination conditions that have to be met by lines between substations and so one of the following two values can be used to set these units: 1.4. X.68 Protection of electricity distribution networks procedure must be started from the furthest substation. 2. When the characteristics of two relays cross at a particular system fault level. If the instantaneous units of the transformer secondary winding overcurrent protection and the feeder relays are subjected to the same short-circuit level. The following parameters are defined: Ki = and Ks = Zsource Zelement Ipickup Iend . since these lines are at the end of the MV system. This value is higher that those mentioned previously to avoid lack of co-ordination with the higher currents encountered due to the magnetic inrush current when energising the transformer. a higher value should be used since the X/R ratio becomes larger. it is necessary to set the instantaneous unit of the relay at the substation that is furthest away from the source to such a value that the relay operates for a slightly lower level of current. then continued by moving back towards the source. can be illustrated by considering the system shown in Figure 5. The 25 per cent margin avoids overlapping the downstream instantaneous unit if a considerable DC component is present. 5. then the transformer instantaneous units need to be overridden to avoid loss of selectivity unless there are communication links between these units that can permit the disabling of the transformer instantaneous overcurrent protection for faults detected by the feeder instantaneous overcurrent protection. In HV systems operating at 220 kV or above.2 Coverage of instantaneous units protecting lines between substations The percentage of coverage of an instantaneous unit which protects a line. as does the DC component. (iii) Transformer units The instantaneous units of the overcurrent relays installed on the primary side of the transformers should be set at a value between 125 and 150 per cent of the short-circuit current existing at the busbar on the low voltage side.3. Between six and ten times the maximum circuit rating. thus avoiding loss of co-ordination. 50 per cent of the maximum short-circuit current at the point of connection of the CT supplying the relay.

2) (5. Iend = V Zs + Zab Zs + Zab Zs + Zab − Zs Ki ⇒X= Ki = Zs + XZab Zab Ki (5. and Ipick up = minimum current value for relay pick up.25. 5. and using a value of 1.3) This gives: Ks = Zs Ks (1 − Ki ) + 1 ⇒X= Zab Ki (5.25 for Ki in eqn. X = percentage of line protected.4: Ipickup = V Zs + XZab (5.5. Iend = current at the end of the line. on the coverage provided by the instantaneous protection can be appreciated by considering the system in Figure 5. if Ki = 1.1 The effect of reducing the source impedance. Zs = source impedance. Example 5.4) For example.Overcurrent protection A ZS ZAB B 69 50 V Figure 5. From this: ZS ( ) 10 2 ZAB ( ) 10 10 IA (A) 100 500 IB (A) 50 83 % coverage 60 76 .4 Coverage of instantaneous units From Figure 5.e.1) where: V = voltage at the relay CT point.6. Zab = impedance of the element being protected = Zelement . the protection covers 60 per cent of the line. then X = 0. i.4. Zs . and Ks = 1.

or plug setting. the relay does not trip before any other protection situated closer to the fault.6 Overcurrent inverse-time relay curves associated with two breakers on the same feeder 5. Definite-time relays and inverse-time relays can be adjusted by selecting two parameters – the time dial or time multiplier setting.70 Protection of electricity distribution networks A ZS ZAB B 1000 V Figure 5. and fault currents seen by the relay are expressed as multiples of this. illustrating the difference in the operating time of these relays at the same fault levels in order to satisfy the discrimination margin requirements. in the presence of a fault.5 Equivalent circuit for Example 5.1 t A B Discrimination margin B A current Figure 5. which is defined as the ratio . and the pick-up or plug setting (tap setting). The curves of inverse-time overcurrent relays associated with two breakers on the same feeder in a typical system are shown in Figure 5.3. This value is usually referred to as the plug setting multiplier (PSM).3 Setting the parameters of time delay overcurrent relays The operating time of an overcurrent relay has to be delayed to ensure that. The pick-up setting The pick-up setting. is used to define the pick-up current of the relay.6.

Overcurrent protection

71

of the fault current in secondary amps to the relay pick-up or plug setting. For phase relays the pick-up setting is determined by allowing a margin for overload above the nominal current, as in the following expression: Pick-up setting = (OLF × Inom ) ÷ CTR (5.5)

where: OLF = overload factor that depends on the element being protected; Inom = nominal circuit current rating; CTR = CT ratio. The overload factor recommended for motors is 1.05. For lines, transformers and generators it is normally in the range of 1.25 to 1.5. In distribution systems where it is possible to increase the loading on feeders under emergency conditions, the overload factor can be of the order of 2. In any case Inom has to be smaller than those of the CT and the thermal capacity of the conductor; otherwise the smallest value has to be taken to calculate the pick-up setting. For earth-fault relays, the pick-up setting is determined taking account of the maximum unbalance that would exist in the system under normal operating conditions. A typical unbalance allowance is 20 per cent so that the expression in eqn. 5.5 becomes Pick-up setting = (0.2 × Inom ) ÷ CTR (5.6)

In HV transmission lines the unbalance allowance could go down to 10 per cent, while in rural distribution feeders the value could be as high as 30 per cent. Time dial setting The time dial setting adjusts the time delay before the relay operates whenever the fault current reaches a value equal to, or greater than, the relay current setting. In electromechanical relays the time delay is usually achieved by adjusting the physical distance between the moving and fixed contacts; a smaller time dial value results in shorter operating times. The time dial setting is also referred to as the time multiplier setting. The criteria and procedures for calculating the time dial setting, to obtain the appropriate protection and co-ordination for the system, are considered next. These criteria are mainly applicable to inverse-time relays, although the same methodology is valid for definite-time relays. 1. Determine the required operating time t1 of the relay furthest away from the source by using the lowest time dial setting and considering the fault level for which the instantaneous unit of this relay picks up. This time dial setting may have to be higher if the load that flows when the circuit is re-energised after a loss of supply is high (the cold load pick-up), or if it is necessary to co-ordinate with devices installed downstream, e.g. fuses or reclosers. 2. Determine the operating time of the relay associated with the breaker in the next substation towards the source, t2a = t1 + tmargin , where t2a is the operating time of the back-up relay associated with breaker 2 and tmargin is the discrimination margin. The fault level used for this calculation is the same as that used to determine the timing t1 of the relay associated with the previous breaker.

72

Protection of electricity distribution networks

3. With the same fault current as in 1 and 2 above, and knowing t2a and the pickup value for relay 2, calculate the time dial setting for relay 2. Use the closest available relay time dial setting whose characteristic is above the calculated value. 4. Determine the operating time (t2b ) of relay 2, but now using the fault level just before the operation of its instantaneous unit. 5. Continue with the sequence, starting from the second stage. The procedure referred to above is appropriate if it can be assumed that the relays have their characteristic curves scaled in seconds. For those relays where the time adjustment is given as a percentage of the operating curve for one second, the time dial setting can be determined starting from the fastest multiplier applied to the curve for time dial 1. In most modern relays the time settings can start from values as low as 0.1 s, in steps of 0.1 s. Time discrimination margin A time discrimination margin between two successive time/current characteristics of the order of 0.25 to 0.4 s should be typically used. This value avoids losing selectivity due to one or more of the following: • • • breaker opening time; relay overrun time after the fault has been cleared; variations in fault levels, deviations from the characteristic curves of the relays (for example, due to manufacturing tolerances), and errors in the current transformers.

In numerical relays there is no overrun, and therefore the margin could be chosen as low as 0.2 s. Single-phase faults on the star side of a Dy transformer are not seen on the delta side. Therefore, when setting earth-fault relays, the lowest available time dial setting can be applied to the relays on the delta side, which makes it possible to considerably reduce the settings and thus the operating times of earth-fault relays nearer the source infeed. Use of mathematical expressions for the relay characteristics The procedure indicated above for phase and earth units can easily be used when the operating characteristics of the relays are defined by mathematical formulae instead of by curves on log-log paper. IEC and ANSI/IEEE Standards define the operating time mathematically by the following expression: t= kβ +L (I /Is )α − 1 (5.7)

where: t = relay operating time in seconds; k = time dial, or time multiplier, setting; I = fault current level in secondary amps; Is = pick-up current selected; L = constant. The constants α and β determine the slope of the relay characteristics. The values of α, β and L for various standard overcurrent relay types manufactured under ANSI/IEEE and IEC Standards are given in Table 5.1. Typical characteristics for both types are shown in Figures 5.7 and 5.8.

Overcurrent protection Table 5.1 ANSI/IEEE and IEC constants for standard overcurrent relays
Standard IEEE IEEE IEEE CO8 CO2 IEC IEC IEC UK α 0.02 2.0 2.0 2.0 0.02 0.02 1.0 2.0 1.0 β 0.0515 19.61 28.2 5.95 0.0239 0.14 13.5 80.0 120 L 0.114 0.491 0.1217 0.18 0.0169 0 0 0 0

73

Curve description Moderately inverse Very inverse Extremely inverse Inverse Short-time inverse Standard inverse Very inverse Extremely inverse Long-time inverse

Given the relay characteristic, it is a straightforward task to calculate the time response for a given time dial setting k, pick-up setting, and the other values of the expression in eqn. 5.7. Likewise, if a particular time response and pick-up setting have been determined, the time dial setting is found by solving k from the same equation.

5.4 5.4.1

Constraints of relay co-ordination Minimum short-circuit levels

When the time delay unit has been set, using maximum fault levels, it is necessary to check that the relays will operate at the minimum fault levels, and in the correct sequence. For this it is sufficient to verify that the plug setting multiplier – (I /Is ) in eqn. 5.7 – under these conditions is greater than 1.5.

5.4.2

Thermal limits

Once the curves for the overcurrent relays have been defined, a check should be made to ensure that they lie below the curves for the designated thermal capacity of machines and cables. In the case of conductors, manufacturers’ graphs, which indicate the length of time that different sizes can withstand various short-circuit values, should be used. A typical graph for copper conductors with thermoplastic insulation is given in Figure 5.9. For motors, the manufacturers’ information should also be consulted. In the case of transformers, the magnitude of the fault current that they can withstand during a given time is limited by their impedance. ANSI/IEEE Standard 242-1986 defines curves of short-circuit capacity for four categories of liquid-immersed transformers, based on the nominal kVA rating of the transformer and the short-circuit impedance.

74

Protection of electricity distribution networks
1000

100

Operating time (s)

10

UK LTI IEC SI 1 IEC VI

IEC EI 0.1

1

10 Current (multiples of Is)

100

Figure 5.7

IEC overcurrent relay curves

Figures 5.10 to 5.13 show the curves of thermal capacity of transformers with the following characteristics: Category I – power rating between 5 and 500 kVA single phase; 15 to 500 kVA three phase. (ii) Category II – power rating between 501 and 1667 kVA single phase; 501 to 5000 kVA three phase. (iii) Category III – power rating between 1668 and 10 000 kVA single phase; 5001 kVA to 30 000 kVA three phase. (iv) Category IV – power rating above 10 000 kVA single phase; above 30 000 kVA three phase. (i)

Overcurrent protection
100

75

10 Operating time (s) 1 IEEE MI IEEE VI US C02 US C08 IEEE EI 0.1 1 10 Current (multiples of Is) 100

Figure 5.8

ANSI/IEEE overcurrent relay curves

The thermal limit curves for Dy transformers have to be shifted to the left by a √ ratio of 1/ 3 to make them more sensitive. This compensates for the lower value of current seen by the relays installed on the primary side, relative to the currents seen by the relays on the secondary side, during single-phase fault conditions, as discussed in Section 5.5.

5.4.3

Pick-up values

It is also important to check that the relay settings are not going to present problems when other system elements are energised. This is particularly critical for motors,

76

Protection of electricity distribution networks
500 400 300 Copper conductor – thermal plastic insulation 75°C

200

50 Short circuit current (kA) 40 30

20

2 1.5

1 kcmil AWG 10 8 20 6 30 40 50 60 70 80 90 100 200 4 2 1/0 2/0 3/0 4/0 300 400 600 800 1000

Figure 5.9

Thermal limits of copper conductors with thermoplastic insulation

and the appropriate code letter, which indicates the number of times nominal current taken when the motor is starting, should always be borne in mind. In the case of transformers, the initial magnetisation inrush current that a transformer takes can be expressed as IInrush = K × Inom , where Inom is the nominal transformer current, and the constant K depends on the transformer capacity; from

18

3 2.5

0

cy

90

cy

cl

4

60

cy

cl

5

30

10 9 8 7 6

4 2 1 cy cy cy cy cl cl cl cl es es es e0 cl es es es es es 0. 0. 0. .0 1. 1. 0. 0. 13 06 03 17 3. 5 0 5 26 0 3 7 3 se s se se 7 se se se se co eco co co se co co co co nd nd nd co nd nd nd nd nd s s s nd s s s s s s 8 16 cy cl cy cl

100 90 80 70 60

Overcurrent protection
10000 9000 8000 7000 6000 5000 4000 3000 2000 Through-fault protection curve for faults that will occur frequently or infrequently

77

1000 900 800 700 600 500 400 300 200

t= 100 90 80 70 60 50 40 30 20

1250
2

I where I = Symmetrical fault current in times normal base current (ANSI/IEEE C57.12.00-1980) Note = The maximum short-circuit withstand capability of category I transformers is defined in (ANSI/IEEE C57.12.00-1980).

Time in seconds

10 9 8 7 6 5 4 3 2

1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2

0.1 2 3 4 5 6 7 8 9 10 20 30 40 50 2 3 4 5 6 7 8 9 10 20 30 40 50

Times normal base current

Figure 5.10

Thermal capacity of transformers between 5 and 500 kVA single phase; 15 to 500 kVA three phase ( from ANSI/IEEE Standard 242-1986; reproduced by permission of the IEEE)

500 to 2500 kVA, K = 8, and above 2500 kVA, K = 10. The inrush point then remains defined by the appropriate inrush current during 0.1 s. Example 5.2 For the system shown in Figure 5.14, and starting from the data that are given there, carry out the following: 1. Calculate the nominal currents and three-phase short-circuit levels at each breaker.

1 Times normal base current Figure 5. Determine the values of the pick-up setting.9 0. 5. Draw the time/current characteristics of the relays on the system. 501 to 5000 kVA three phase ( from ANSI/IEEE Standard 242-1968. 2 3 4 5 6 7 8 9 10 20 30 40 50 2 3 4 5 6 7 8 9 10 20 30 40 50 This curve may also be used for back-up protection where the transformer is exposed to frequent faults normally cleared by high-speed relaying 1 0.3 0. reproduced by permission of the IEEE) 2.11 Thermal capacity of transformers between 501 and 1667 single phase.00-1980) K = Constant determined at maximum I with t = 2 seconds Note = Sample I2t = K curves have been plotted for selected transformer impedances as noted. Find the percentage of the line BC protected by the instantaneous unit of the overcurrent relay associated with breaker 2.5 0. Select the transformation ratios of the CTs. time dial and instantaneous settings of all phase relays to ensure a co-ordinated protection arrangement.12.8 0. 4.78 10000 9000 8000 7000 6000 5000 4000 3000 2000 Protection of electricity distribution networks Through-fault protection curve for faults that will occur frequently (typically not more than 10 in transformer lifetime) Through-fault protection curve for faults that will occur frequently (typically not more than 10 in transformer lifetime) 1000 900 800 700 600 500 400 300 200 Time in seconds 100 90 80 70 60 50 40 30 20 10 9 8 7 6 5 4 3 2 12 10 8 7 6 5 4 K transformer impedance For fault currents from 70% to 100% of maximum possible : I2t = K where I = Symmetrical fault current in times normal base current (ANSI/IEEE C57.6 0.4 0.2 0.7 0. 3. .

All relays have inverse time characteristics.2 0.12.1 2 3 4 5 6 7 8 9 10 20 30 40 50 2 3 4 5 6 7 8 910 20 30 40 50 Times normal base current Figure 5. 3.8 0. as shown in Figure 5. .00-1980) K = Constant determined at maximum I with t = 2 seconds Note = Sample I2t = K curves have been plotted for selected transformer impedances as noted. The discrimination margin to be 0.4 0. reproduced by permission of the IEEE) Take into account the following considerations: 1.9 0. 5001 to 30000 kVA three phase ( from ANSI/IEEE Standard 2421968.15.3 0.7 0.6 0.Overcurrent protection 10000 9000 8000 7000 6000 5000 4000 3000 2000 79 Through-fault protection curve for faults that will occur frequently (typically not more than 5 in transformer lifetime) Through-fault protection curve for faults that will occur frequently (typically not more than 5 in transformer lifetime) 1000 900 800 700 600 500 400 300 200 Time in seconds 100 90 80 70 60 50 40 30 20 10 9 8 7 6 5 4 3 2 12 10 8 7 6 5 4 K transformer impedance For fault currents from 50% to 100% of maximum possible : I2t = K where I = Symmetrical fault current in times normal base current (ANSI/IEEE C57.12 Thermal capacity of transformers between 1668 and 10000 kVA single phase. Relay data: Pick-up setting: 1 to 12 A in steps of 1 A Time dial setting: as in Figure 5.5 0.15 Instantaneous: 6 to 144 A in steps of 1 A.4 s. 2. This curve may also be used for back-up protection where the transformer is exposed to frequent faults normally cleared by high-speed relaying 1 0.

can be obtained: 115 + 103 V2 = = 13.2 0.00-1980) K = Constant determined at maximum I with t = 2 seconds Note = Sample I2t = K curves have been plotted for selected transformer impedances as noted.80 10000 9000 8000 7000 6000 5000 4000 3000 2000 Protection of electricity distribution networks 1000 900 800 700 600 500 400 300 200 Through-fault protection curve for faults that will occur frequently or infrequently Time in seconds 100 90 80 70 60 50 40 30 20 For fault currents from 50% to 100% of maximum possible : I2t = K where I = Symmetrical fault current in times normal base current (ANSI/IEEE C57. reproduced by permission of the IEEE) Solution Calculation of nominal currents and three-phase short-circuit levels From Figure 5. 10 9 8 7 6 5 4 3 2 12 10 8 7 6 5 4 K transformer impedance 1 0.92 Zsource = Psc 950 × 106 2 referred to 115 kV .7 0.12.14 the short-circuit level at busbar A.4 0.8 0.3 0.5 0.9 0.1 2 3 4 5 6 7 8 9 10 20 30 40 50 2 3 4 5 6 7 8 9 10 20 30 40 50 Times normal base current Figure 5.13 Thermal capacity of transformers above 10000 kVA single phase.6 0. above 30000 kVA three phase ( from ANSI/IEEE Standard 242-1968. and the impedance of the line BC.

2 × 103 .125 Ω (13.14 Schematic diagram for Example 5.16.8 A 81 85.2 kV 25 MVA Z % = 4.Overcurrent protection 950 MVASC 4 Y Y 3 B 2 115/13.35 Ω (115 kV) 1. Nominal currents 3 × 106 P =√ = 131.048 × ZlineBC = 85.2 A Inom1 = √ 3×V 3 13.2 kV) C 1 3 MVA 3 MVA 3 MVA Figure 5.2 Ztransf = Zpu × ZBase = 0.35 115 × 103 25 × 106 2 = 25.39 referred to 115 kV referred to 115 kV The equivalent circuit of the system referred to 115 kV is shown in Figure 5.

7 0.07 0.03 0.08 0.04 0.8 0.82 Protection of electricity distribution networks 100 90 80 70 60 50 40 30 20 1 2 3 4 5 6 7 8 9 10 20 30 40 10 9 8 7 6 5 4 3 2 Time (seconds) 1.0 0.09 0.4 0.06 0.02 ½ 3 2 Time dial setting 1 0.6 0.9 0.15 Typical operating curves for an inverse-time relay .10 0.2 11 10 9 8 7 6 5 4 0.5 0.01 1 2 3 4 5 6 7 8 9 10 Multiples of tap value current 20 30 40 Figure 5.05 0.3 0.

92 Ω 25.92 Choice of CT transformer ratio The transformation ratio of the CTs is determined by the larger of the two following values: (i) Inom (ii) maximum short-circuit current without saturation being present.92 + 25.39) = 1689(115/13.2/115) = 125.39 + 83.2 kV 115 × 103 = 1689.2 = 393.16 Equivalent circuit of the system shown in Figure 5.35) = 532.5 A 3 115 × 103 Short-circuit levels The equivalent circuit gives: 115 × 103 = 532.2) = 14714.2 summarises the calculations.5 A Inom3 = √ 3 13.39 Ω B 85.35 Ω C 83 115 kV 3 Figure 5.0 A referred to 115 kV IfaultB = √ 3 (13.8 A referred to 115 kV IfaultA = √ 3 × 13. To fulfil this condition and assuming that a C100 core is used and that the total burden is 1 .2 kV 115 × 103 = 4769.14 Inom2 = 3 × Inom1 = 3 × 131.2 A referred to 13.6(115/13.Overcurrent protection A 13.2 × 103 25 × 106 Inom4 = √ = Inom3 × (13.6 A referred to 115 kV IfaultC = √ 3 (13. then Isc (5/X) ≤ 100 A where Isc is the short-circuit current. Table 5.6 A 25 × 106 = 1093.8 A referred to 13.2) = 4640.92 + 25. .

84 Protection of electricity distribution networks Table 5.2 kV.1 + 0.18 s Relay 3 To discriminate with relay 2.2 Pnom (MVA) 3 9 25 25 Inom (A) 131.5 × 4640) 5/300 =38.8 14714.5 s.5(1093.69 A.75 times From Figure 5.5)5/300 = 3.8 (5/100)Isc (A) 232.75 and a time dial setting at 1.7 735. Iinst. and top at least 0. set at 37 A Iinst. with a plug setting multiplier of 9.6)5/800 = 3. Plug setting multiplier.15.prim = (37) 800/5 = 5920 A at 13.5 Isc )(1/CTR) = (0.58 s.28 A. PSMb = (2340 A × 5/300) × 1/4 = 9.25 times With 9. ⇒ time dial setting = 2 .5(393.18 + 0.58 s PSMa = 5920 A × 5/1100 × 1/8 = 3. PSMa = 2340 A × 5/800 × 1/4 = 3.25 PSM and time dial setting = 2 ⇒ t2b = 0.8 4769. ⇒ pick-up set at 4 A Relay 2: 1. ⇒ pick-up set at 4 A Determination of time dial setting and calibration of the instantaneous setting Relay 1 Ipick up = 4 × 300/5 = 240 A Time dial setting selected is 1.66 times.4 = 0. time dial = 2 is chosen Instantaneous setting = (1.5(131.2)5/300 = 3.5)5/1100 = 7.6 1093.14A.36 PSM and top = 0.2 393.5(125.7 238.25 (4640)(5/800) = 36.0 Setting of instantaneous element = (0. time dial and instantaneous setting values: calculation of the pick-up settings Relay 1: 1.1 s Relay 2 2340 A should produce operation of t2a in at least 0.25 A. t1b = 0.5 CT ratio 300/5 800/5 1100/5 300/5 Breaker number 1 2 3 4 Determination of the pick-up setting.4 = 0.2 Nominal currents.67 A. take Iinst.5 125.prim2 = 5920 A Require operation of t3a in at least 0. set at 39 A.5 s.0 1735.0 14714.46 A.5 Isc (A) 4640.36 times With 3. trip = 39 × 300/5 = 2340 A primary at 13.66 times With PSM at 3.25 IfaultC )(1/CTR) = 1.2 kV PSMb = 5920 A × 5/800 × 1/4 = 9. short-circuit currents and CT ratios for Example 5. ⇒ pick-up set at 8 A Relay 4: 1. ⇒ pick-up set at 4 A Relay 3: 1.

25 × IfaultB )(1/CTR) = 1.prim = 36 (300/5) = 2160 A referred to 115 kV Iinst.46 (1 − 1.8 at 13.36 times PSM and time dial setting = 2.8 A (13. ⇒ t3b = 0.21 + 0.0 Time dial – 2 2 5 Instantaneous Isec (A) 40 37 – 36 Instantaneous Iprim (A) 2400 5920 – 18818 Relay associated with breaker number 1 2 3 4 .46 Ks = Zelement 85.3 Summary of settings for Example 5.25 (1689) 5/300 = 35.61 s ⇒ time dial setting = 5.68 X% = 1.04 times. PSM = 14714. Table 5. as referred to in Section 5.0 8.2 A referred to 13.92 + 25.19 A Setting = 36 A.2/115) 5/300 × 1/4 = 7.5 4.3 summarises the four relay settings. Iinst.28 Isc end 4640 Zsource 13. the instantaneous element of the relay associated with breaker 3 is overridden and the discrimination time is applied for a fault on busbar B to avoid lack of co-ordination with the instantaneous units of the relays associated with the feeders from the busbar. Setting of instantaneous element = (1. the instantaneous element covers 68 per cent of the line BC.Overcurrent protection 85 However.61 s With 7.prim = 2160 (115/13.2) = 18818.36 times With 8.2 kV.8 A × 5/1100 × 1/8 = 8.04 PSM and top = 0. PSMb = 14714.2 Pick-up (A) 2.3.4 = 0.0 4.8 A. Based on Isc = 14714.2 kV Table 5. Require t4 = 0.21 s Relay 4 For 14714.28) + 1 = 0. Percentage of line A-B protected by the instantaneous element of the relay associated with breaker 2 Ks (1 − Ki ) + 1 X% = Ki Isc pick up 5920 Ki = = = 1.1.35 and 0.39 = = 0.28 Therefore.

100 kA 1.0 Amp CT = 800/5 Amp R3: Tap = 8. it can be assumed that the voltages between the phases of the transformer are the same.0 Inst = 39. shown in Figure 5.0 Amp CT = 300/5 Amp R2: Tap = 4. .0 Amp Time dial = 2.0 Amp Time dial = 2. To simplify the operations.17.5 Co-ordination across Dy transformers In the case of overcurrent relay co-ordination for Dy transformers the distribution of currents in these transformers should be checked for three-phase. 5.0 Inst = 36.0 Amp Time dial = 5.0 s L1 0.18. for both the primary and the secondary windings.0 Amp CT = 300/5 Amp Thermal limits T1: Power transformer L1: Feeder cable 100 s T1 10 s R1 R2 R4 R3 1.17 Relay co-ordination curves for Example 5.10 s 0.0 Inst = Disable CT = 1100/5 Amp R4: Tap = 4.86 Protection of electricity distribution networks 1000 s Relay settings R1: Tap = 4.2 The co-ordination curves of the relays associated with this system are shown in Figure 5.0 kA 10 kA – l(13 kV) 100 kA 1000 kA Figure 5. It should be noted that these are all drawn for currents at the same voltage – in this case 13.0 Amp Time dial = 1. phase-to-phase.2 kV. and single-phase faults on the secondary winding.100 s 0.0 Inst = 37.

13) .Overcurrent protection 87 Three-phase fault Phase-to-phase fault Phase-to-earth fault Figure 5. the number of turns √ the primary is equal to on of the secondary.12) (5.10) From the above it can be seen that the currents that flow through the relays associated with the secondary winding are equal to the currents that flow through those relays associated with the primary winding.18 Distribution of current for a fault on a Dy transformer √ 3 times the number of turns Thus. Phase-to-phase fault √ √ 3 × Eφ−n Eφ−φ 3 If = = = I 2X 2X 2 √ N2 I 3 ×I × = Idelta = 2 N1 2 Iprimary = 2Idelta = I (5. Three-phase fault Eφ−n =I X N2 I =√ Idelta = I N1 3 √ Iprimary = 3Idelta = I If = (5. i. as expected.11) (5.e. since the primary and secondary voltages are equal and the fault involves all three phases. N1 = 3N2 .8) (5.9) (5.

which could lead to a situation where the selectivity between the two relays is at risk. For this reason. From Figure 5. calculate the following: 1.88 Protection of electricity distribution networks For this case. Assume that the total burden connected to each CT is 1 and that C100 cores are used.16) Thus. which has a ratio of 250/5. given that the primary turns are multiples of 100 except for the CT for breaker number 9. The results of the three cases are summarised in Table 5. the current distribution at the primary is 1-1-2. Table 5. it can be seen that the critical case for the co-ordination of overcurrent relays is the phase-to-phase fault. Analysing the results.3 For the system shown in Figure 5. In this case the relays installed in the secondary carry a current less than the equivalent current flowing through the primary relays. as shown in Figure 5. for this fault. the current through √ relays installed in the secthe ondary winding circuit on the faulted phase is equal to 3 times the current that flows through the relays associated with the primary winding on the same phase.4 Fault Three-phase Phase-to-phase Phase-to-earth Summary of fault conditions Iprimary I I I Isecondary I √ √3I /2 3I . and 0-1-1 at the secondary. for a phase-to-earth fault.18 it is clear that. 2.4.20. Phase-to-earth fault Eφ−n =I X N2 I =√ Idelta = I × N1 3 If = I Iprimary = √ 3 (5. The three-phase short-circuit levels on busbars 1 and 2. Example 5. the current that goes through the relays installed in the secondary √ winding circuit is equal to 3/2 times the current that flows through the relays associated with the primary on the phase that has the largest current value.15) (5. and on the operating time for the primary relays for the full fault current value If . The transformation ratios of the CTs associated with breakers 1 to 8.19.14) (5. the discrimination margin between the relays is based on the operating time of the √ secondary relays at a current equal to 3If /2.

5 kV 15 MVA Z%=10 YYo TR1 2.086 pu 0.3 Dy1 TR2 4 Busbar 1 1 2 13.453 kA Busbar 6 0.5 kV ZL1 = 1.3 showing fault currents .453 kA Busbar 4 L3 7 0.5 kV 5 34.170 kA 182.87 MVA 3.5 kV 1 MVA Busbar 2 34.5 kV 0.20 Single line diagram for Example 5.000 kA L2 6 Busbar 5 L4 34.Overcurrent protection t 89 0.651 kA 115 kV 9 250/5 115/34.19 Co-ordination of overcurrent relays for a Dy transformer Busbar 7 0.5/13.2 kV 3 1 MVA 1 MVA 1 MVA Figure 5.5 kV 34.060 kA 8 34.2 kV 3 MVA Z%=7.4 s 3 I 2 f If A Figure 5.000 kA Busbar 3 34.

and the pick-up and time dial settings of the relays to guarantee a co-ordinated protection arrangement.11 MVA).5 kV busbar at substation A can be obtained from the values in Figure 5. allowing a discrimination margin of 0.93 . Take into account the following additional information: 1. Relay data: Pick-up setting = 1 to 12 A in steps of 1 A Time dial setting = 0.5 kV line protected by the instantaneous element of the overcurrent relay associated with breaker 6.3 Instantaneous = 1100 A primary current 3.5 kV P = 100 MVA 2. The setting of the instantaneous elements of the relays associated with the feeders is to be carried out on the basis of ten times the maximum nominal current. referred to 34.086 (34500)2 3 × 106 (34500)2 100 × 106 = 28. impedances are calculated on the following bases: V = 34.073 Zline = 1. referred to 34.20. for a fault on the 34.11 × 106 (34500)2 15 × 106 = 7. the equivalent impedance of the system behind the busbar is calculated as follows: V2 (34500)2 = = 6. The settings of the instantaneous elements. are given in Figure 5.u.5 kV Zbase = Ztransf1 = 0.1 = 88.1 4.20 (183.5 kV Psc 183. referred to 34. referred to 115 kV Ztransf 2 = 0. The short-circuit MVA and fault currents.5 kV busbar at substation A. The percentage of the 34.5 kV = 12. Solution Calculation of equivalent impedance The short-circuit level on the 34.4 s.90 Protection of electricity distribution networks 3. The settings of relay 7 are: Pick-up = 4 A Time dial = 0. The p.17 . Using this. 6.05 Instantaneous: 6 to 14 A in steps of 1 A 5.5 kV . All the relays have an IEC very inverse time characteristic with the values given in Table 5.05 to 10 in steps of 0.93 . referred to 34.96 .5 . 4.

50 Ω 12. Nominal currents Inom1.31 A at 115 kV Short-circuit levels √ The short-circuit MVA (Psc ) of the transformer at 34.5 kV Inom7 = √ = 16.5 kV BUSBAR 2 34.18 .73 A Inom8 = 251.34 × 34.2.20 A at 34. referred to 34.5)2 = 9.2 × 103 3 × 106 3 × 34.21.69 = 101. referred to 34.5 kV 28.5 kV 6.5 × 103 15 × 106 3 × 115 × 103 = 43. is shown in Figure 5.5 kV 3 Figure 5.5 kV Inom6 = 50.02 A at 34.74 A at 13.69 MVA.2 × 103 3 × 106 3 × 13.5 kV Inom9 = √ = 75.3 The equivalent positive sequence network.2 kV = 50.17 = 13. Ztransf1 + Zbase = (34.5 kV = 3 × 2170.5 kV 129.80 .22 A at 13.5 kV.21 Positive sequence network for Example 5.93 Ω BUSBAR 1 34.Overcurrent protection 6-SUB-A 34.97 .20 A at 34.3 = √ Inom4 = √ Inom5 = √ 1 × 106 3 × 13.2 kV = 131.5 × 103 = 129.97 . referred to 115 kV − 88. referred to 115 kV Zsystem = 101.96 Ω 91 34.5 × 103 1 × 106 3 × 34.

74 Isc (A) 4797.2. 3 (6.20 131.5 + 12.15 A at 34.96) = 1075.25 A at 115 kV Ifault9 = √ 3 × 13.80 Selection of current transformers Table 5.2.4 = √ = 411.3 = 43.84 A at 13.01 153.34 3060.06 (5/100)Isc (A) 239.20)(5/100) = 3.2.28 53.92 A.28 A.34 A at 34.2.5 × 10 = 1025.5 kV Ifault6.5 gives the main values for determining the transformation ratio of the CTs.80 53.5 kV Ifault8 = √ 3 × 34.74)(5/100) = 3.31 251. 3 Determination of CT ratios for Example 5.22 A.02 16. PU4 = (1. PU4 = 5 A Iload5 = 50.20 A.5)(43.92 Protection of electricity distribution networks 34.63 A at 34.5) 129.20 50.74 A.7 = √ 3 (6. PU5 = (1.5)(50. (Isc × 5/X) ≤ 100 ⇒ X ≥ (Isc × 5/100) Determining the pick-up (PU) values Iload1.67 1076. PU5 = 4 A Table 5.80 CT ratio 250/5 300/5 200/5 200/5 100/5 200/5 100/5 .34 1025.5)(131.2 kV 34.40 3060.51 153.93 + 28.3 Pnom (MVA) 15 15 1 3 3 3 1 Inom (A) 75.40 A at 34.76 A.22 43.5 × 103 Ifault1. maximum short-circuit current for which no saturation is present.3 = (1.5 kV.3 = 4 A Iload4 = 131.22)(5/200) = 4.69 × 106 = 2170. PU1.35 2170.01 51. 2. which is taken as the larger of the two following values: • • nominal current.5 × 103 = 3064.5 kV Ifault5 = √ 3 (6.5 + 12.3.87 108.73 50.5 × 103 115 × 103 = 4811. PU1. Therefore.93) 3 34.5 Breaker number 9 8 7 6 5 4 1.06 1076.

31 A. PSM = 22/4 = 5.28 A.59 s. Relay 4 To discriminate with relay 3 at 440 A requires operation in t4a − 0. PSM5a = 1075.1.2/34.15 times. so that Iprim.5 = 0. and with a time dial setting of 0.55 s.4 = 0.9 s.84) × (13.7. PU6 = 2 A PU7 = 4 A (as given in the example data) Iload8 = 251.5 times.20. PSM4b = (0.25(1025.Overcurrent protection Iload6 = 50.15 s. PU6 = (1. the relay operating time is (0. with the time dial setting of 0. From the given information.19 s.9 s.5)/(5.26 A.5]/(PSM − 1).20.15 times.63 times.5 times.5)(75. 2 and 3. By similar calculations to those for relays 1. trip = 32 A× 200/5 = 1280 A.5) × (5/100) × (1/4) = 5. from the relay characteristics and eqn. PSM4a = (440 × 5/200)(1/5) = 2. this gives a required time dial setting of 0.55 × (2. 5. the minimum time dial setting of 0.20 A. 2 and 3 When calculating the settings for the relays situated at the end of the circuit. this relay has to operate in t6a = 0.55 s. The setting of the instantaneous element is 1. Instantaneous setting = 1. Given the constants for the IEC very inverse overcurrent relay are α = 1.04 ⇒ 32 A. PU9 = (1.5) × (5/100) = 25.02)(5/300) = 6.02 A.84 × (13.20)(5/200) = 1. β = 13. Iinst.87 A ⇒ set at 22 A.4 = 0. At 6. Relay 5 The back-up to relay 4 is obtained by considering the operating time for a line-to-line fault of t5a = 0.73 A ⇒ 26 A.15 + 0. PSM6a = 520 × (5/200) × (1/2) = 6.05 is selected. the relay operating time t is [(time dial setting) × 13. as referred to in Section 5. Iprim.2 times.049 ⇒ 0. and t4a = 0. At 6. Thus. from eqn.31)(5/250) = 2.05. and t5a = 0. the time dial setting = 0. the setting of the instantaneous element is based on ten times the maximum load current seen by the relay.88 A. The operating time of the time delay element is calculated from PSM5b = (1/4) × 26 = 6. t4b = 0.5)(50. At 2.05. the time dial setting = 0.05 × 13. The operating time for a line-to-line fault is determined by taking 86 per cent of the three-phase fault current.5 s.5 times and.5 times and t6a = 0. 5.2 − 1)/13.4 = 0.5 + 0. This relay has no setting for the instantaneous element. PU8 = (1.5 times.37 ⇒ 0.86)(1075.5.59 s.2 times.19 + 0.5 − 1) = 0.2/34. trip = 10 × Inom × (1/CTR) = 10 × 43. PU9 = 3 A 93 Determining the instantaneous and time dial settings Relays 1.25(1075. and L = 0 then.84 × 5/200)(1/5) = 4.74 × (5/100) = 21. t5b = 0. Relay 6 At 520 A. trip = 26(100/5) = 520 A. At 5.40.0. .15)(5/200) = 32. where PSM is the ratio of the fault current in secondary amps to the relay pick-up current. trip = 22(100/5) = 440 A. Iprim. PU8 = 7 A Iload9 = 75.7.5)(251.3.

Therefore the operating time to give correct discrimination with relay 7 is t8a = 0.07 ⇒ 0.2.end 1025. For back-up relay 8.4. The maximum short-circuit current to be used for this relay is that which flows from the 115 kV busbar to the 34.5/115) × (5/250) = 16. so that PSM8a = 1100 × (2170. Relay 7 has an instantaneous primary current setting of 1100 A.75 times.86 times.1. Only the infeed from the transformer has to be taken into account. At 13. t8b = 0. The instantaneous setting = 1. and so the operating time of both relays is determined by this value.69 s. equivalent to 27.1. At 1.32 s.20.1. The co-ordination curves of the relays associated with this system are shown in Figure 5.75 times and with a time dial setting of 0. top = 0.72 s. At 5.09 s.25 = Isc.25 × 2170.86 times and t8a = 1. Percentage of 34. the time dial setting = 0. Relay 9 This relay backs up relay 8 in a time of t9a = 0.34 times.40) × (5/300) × (1/7) = 1.3.42 s.34 × (5/300) × (1/7) = 5. For relay 7 PSM = 1100 × 5/200 × 1/4 = 6.28 A ⇒ 17 A.5 kV busbar for a fault on the latter. and summarised in Table 5.4 + 0. At 4. Relay 8 Relay backs up relays 6 and 7 and should be co-ordinated with the slower of these two relays.34 times and t9a = 0.6.09 s.5 kV line protected by the instantaneous element of the overcurrent relay associated with breaker 6 Given %= where Ki = and Ks = Zsource Zelement Isc. At 6.34/3060.87 times.39 × (34. For relay 6 PSM = 1100 × 5/200 × 1/2 = 13.94 Protection of electricity distribution networks Relay 7: PSM and DIAL as given in the example data. then top = 0.18 ⇒ 0.87 times and with a time dial setting of 0.15 Ks (1 − Ki ) − 1 Ki .34 × (34.5/115) × (5/250) × (1/3) = 4. Iprim.5 A secondary current which is less than the setting of relay 6.22.69 + 0.17 times and with a time dial setting of 0.32 = 0. and PSM8b = 2170.4 = 1. no instantaneous setting is applied to relay 8 for the reasons given in Section 5. Here also.17 times. PSM9a = 2170. the time dial setting = 0.pickup 1280 = 1. the contributions to relay 6 from substations G and M are not considered.72 s. trip = 17 × (250/5) = 850 A referred to 115 kV.

0 Amp Time dial =0.0 Amp Time dial =0.3 Inst =27. R3: Tap = 4.0 Amp Time dial =0.05 Inst =22.010 kA 0.6 Relay number 1. 3 4 5 6 7 8 9 Summary of settings for Example 5.0 Amp CT =100/5 Amp R4: Tap =5.010 s 0.0 kA–1(34 kV) 10 kA 100 kA 1000 kA 100 kA Figure 5.5 Amp CT =200/5 Amp R8: Tap =7.0 100/5 200/5 100/5 200/5 200/5 300/5 250/5 . TR2: Power transformer L1 Feeder cable 95 100 s TR1 TR2 10 s R-8 R1 R-7 R4 R-5 R-6 L1 R-9 1.100 s 0.100 kA 1. R2.100 kA 1.0 s R2 R3 0.2 Inst =26.22 Relay co-ordination curves for Example 5.010 kA 0.3 CT ratio Pick-up (A) 4 5 4 2 4 7 3 Time dial 1/2 1/2 3 6 5 1 2 Instantaneous Isec (A) 20.3 Table 5.0 Amp CT =200/5 Amp R7: Tap =4.2 Inst =17.5 – 17.0 Amp Time dial =0.0 27.0 Amp Time dial =0.0 Amp CT =100/5 Amp R6: Tap =2. 2.0 Amp CT =250/5 Amp Thermal limits TR1.00 Amp Time dial =0.05 Inst =Disable CT =200/5 Amp R5: Tap =4.1 Inst =Disable CT =300/5 Amp R9: Tap =3.0 kA – l(13 kV) 10 kA 0.Overcurrent protection 1000 s Relay settings R1.0 Amp Time dial =0.4 Inst = 32.0 32.0 – 26.

50/12.25 V2 34. which is called fuse saving.96 Protection of electricity distribution networks From the computer listing Zf = and Ks = 6.70 1. It is important to ensure that the settings of these units are checked for co-ordination with phase-only sensing devices such as downstream fuses and reclosers.6 Co-ordination with fuses When co-ordinating overcurrent relays it may be necessary to consider the time/current characteristics of fuses which are used to protect MV/LV substation transformers. and/or earth-fault relays.7 Co-ordination of negative-sequence units Sensitivity to phase-to-phase fault detection can be increased by the use of negativesequence relays (type 50/51 Q) because a balanced load has no negative-sequence (I2 ) current component. it is necessary to keep in mind that the magnitude of the current for a phase-to-phase fault is √ 3/2 (87 per cent) of the current for a three-phase fault at the same location. Ia = 0 √ and Ib = −Ic . For a phase-to-phase fault. 5. It is therefore necessary to consider the case of preventing fuse operation because of the problems of replacing them after they operate. This is also the situation for phase-to-earth faults if type 50/51 relays are used since a balanced load has no zero-sequence (I0 ) component. In these cases it may be preferable to forgo the selectivity of the protection system by not taking account of the fuse characteristic curve. the circuit is left in an open-circuit condition until the fuse is replaced. . as indicated in eqn. When a fuse operates. the 3 factors cancel. When the two factors ( 3/2 of √ and 1/ 3) are combined. Instantaneous overcurrent and time-delay overcurrent negative-sequence units are common features of the new multifunction relays.11 so that the instantaneous element covers 70 per cent of the line. Therefore the magnitude of the negative-sequence component is 1 / 3 √ (58 per cent) of the magnitude√ the phase-fault current.25) + 1 = 0. so that the fuse will then act as a back-up.2: Ia2 = 1/3(Ia + a 2 Ib + aIc ).50 Therefore %= 0.50 P 183.11. 5. 5. To determine the pick-up setting of the negative-sequence element. leaving a 1/2 factor. On the other hand.50(1 − 1. the magnitude of the negative-sequence component for a phase-to-phase fault can be obtained from the following expression taken from Section 2.52 = = 6.93 = 0.

which are generally referred to as type 51V relays. On the other hand. These devices are used to improve the reliability of the relay by ensuring that it operates before the generator current becomes too low.732 × 150 = 259. Since the voltage-controlled type has a fixed pick-up. especially if the generators are isolated and the faults are severe.75 × 0.8 Overcurrent relays with voltage control Faults close to generator terminals may result in voltage drop and fault current reduction. with a pick-up value of 1. the voltage-restrained type will be less . the relay can be set for 175 per cent of generator rated current with rated voltage applied.and earth-fault devices it is necessary to adjust the negative-sequence element pick-up value by a multiplier that is the ratio of the fault current to the negative-sequence current. in generation protection it is important to have voltage control on the overcurrent timedelay units to ensure proper operation and co-ordination. Consider a downstream time-delay phase overcurrent element with a pick-up value of 100 A. The negativesequence pick-up value should be multiplied by a value greater than 1. For example. The varying pick-up level makes it more difficult to co-ordinate the relay with other fixed pick-up overcurrent relays. and an upstream negative-sequence time-delay relay with a pick-up value of 150 A. The voltage-restrained (51/27R) feature causes the pick-up to decrease with reducing voltage.8 A. It should be set to function below about 80 per cent of rated voltage with a current pick-up of about 50 per cent of generator rated current. 5.25 = 0. Generally. and no multiplying factor is involved.732.732 for a phase-to-phase-to-earth fault.Overcurrent protection 97 From the above it is recommended that negative-sequence elements be set by taking one half of the phase pick-up setting in order to achieve equal sensitivity to phase-to-phase faults as to three-phase faults. compared to the voltage-controlled relay. Therefore. phase-to-phase faults are the most critical with all other fault types resulting in an equal or greater shift of the time/current characteristic curve to the right on the co-ordination graph. In order to check the co-ordination between these two units for a phase-to-phase fault. it can be more readily coordinated with external relays than the voltage-restrained type. and by a factor of 3 for a phase-to-earth fault. For a phase-to-phase fault this is 1. The voltage-controlled approach typically inhibits operation until the voltage drops below a pre-set value.44 times rated). To plot the negative-sequence time/current characteristics on the same diagram as the phase. the phase overcurrent element plot should be shifted to the right by a factor of 1. for co-ordination with downstream phase overcurrent devices. Since no negative-sequence current flows for a three-phase fault. The voltage-controlled (51/27C) feature allows the relays to be set below rated current. as shown in Figure 5. negative-sequence relay operation does not take place.732. At 25 per cent voltage the relay picks up at 25 per cent of the relay setting (1. There are two types of overcurrent relays with this feature – voltage-controlled and voltage-restrained.23. and operation is blocked until the voltage falls well below normal voltage.

2 and 3 in this case). The program should define the settings in accordance with the criteria given in Section 5. The algorithm then establishes pairs of relays and identifies . the margins and constraints of the system and the available settings of the relays being co-ordinated. determining first which one is furthest away from the source. as illustrated in the past sections. Verify that the requirements given in Section 5. 5. The overall process consists basically of three steps: 1. The single line diagram given in Figure 5. or new relays should be tried. This section introduces a very simple procedure to set overcurrent relays using different algorithms. otherwise the process should be repeated with lower discrimination margins. especially if different topologies have to be analysed. In addition. The algorithm files the discrimination margins required. However.3.9 Setting overcurrent relays using software techniques The procedure for determining the settings of overcurrent relays. (1. Identify the pairs of relays to be set. 3.4 are fulfilled. 2. and which is acting as the back-up. the settings of those relays closest to the loads and at the boundaries of other networks have to be considered.98 Protection of electricity distribution networks 100 % IPickup 75 50 25 25 50 75 % VRestraint 100 Figure 5. and those for relay 9 which corresponds to the only boundary utility. The entry data required are the short-circuit currents for faults at all busbars. the setting of the relays closest to the loads. the procedure becomes cumbersome if performed manually and therefore software techniques are required. Locate the fault and obtain the current for setting the relays. for large systems.20 can be used to illustrate simply how a typical computer program can tackle a co-ordination problem. is relatively simple for radial or medium-sized interconnected systems.23 Pick-up setting of 51/27R relay susceptible to operation on swings or motor-starting conditions that depress the voltage below the voltage-controlled undervoltage unit drop-out point.

and hardware outputs. Similar procedures are carried out for those relays associated with the rest of the lines connected to busbar 6. output. Relay logic equations entered and saved in the relay system’s nonvolatile memory integrate the selected or enabled protection in order to provide the operating settings that control the relay pick-up threshold and time delay values. 2 and 3. control. virtual outputs. the algorithm carries out all the necessary checks in accordance with the restraints given in the data entry. If any requirement is not fulfilled.2 Principles of digital logic Digital systems are constructed by using three basic logic gates. for small systems and fault-case analysis. the algorithm determines the slowest of these relays which then has to be co-ordinated with relay 8. normally-contained function blocks that have all of the inputs and outputs of its discrete component counterpart. This reduces the curve drawing process.10. which should include blocks with control inputs. The more interconnected the system. the algorithm will determine which is the slowest relay of 1. 5. and finally relay 8 is co-ordinated with relay 9. protection. and co-ordinate this with relay 4 located upstream. virtual outputs. which is very time consuming. monitoring and reporting capabilities of multifunction protection relay systems.Overcurrent protection 99 which relay acts as a back-up within each pair. OR and NOT. the larger and more complicated will be the critical route and computer programs are being used more and more for large systems. be co-ordinated with relay 5. but separately programmed functions. Each relay system has multiple. such as the . However. These gates are designated AND.10. should be identified. When the process is finished. Numerical relays can be configured to suit a particular specification by defining the operating settings (pick-up thresholds and time delays) of the individual protection and control functions. During the execution of the program the critical route. Changing logic settings is similar to rewiring a panel.10 Use of digital logic in numerical relaying 5.20. After this. A group of logic equations defining the function of the multifunction relay is called a logic scheme. There also exist other logical gates. Each independent function block interacts with control inputs. the process is started again with a lower discrimination margin or using relays with different characteristics until adequate co-ordination is achieved. For the system shown in Figure 5. and hardware outputs based on logic variables defined in equation form with relay logic. which corresponds to the one with the highest number of relay pairs. manual methods are still used with the help of software editors containing libraries with relay curves from many manufacturers. and is used for managing the input.1 General When using numerical relays it is necessary to provide a suitable method for handling the relay logic capabilities. and relay 5 with relay 6. by ensuring that the required time discrimination margin is maintained in all cases. Operating settings and logic settings are interdependent. in turn. 5. Relay 4 will.

10. but not both. Each scheme is configured for a typical protection application and virtually eliminates the need for start-from-scratch programming. Figure 5. where the features of each logic scheme are broken down into functional groups.24. If the variable is A. The EOR – the Exclusive OR gate – is a circuit that will give a high output if either. The NOR gate is a NOT-OR circuit that is equal to an OR circuit followed by a NOT circuit. is used to show the EOR operation.3 Logic schemes Normally numerical relays have several pre-programmed logic schemes which are stored in the relay memory. This logic scheme provides basic time and instantaneous overcurrent protection. ⊕. The OR gate gives a high output if one or more of its inputs are high. and negative-sequence overcurrent protection. The basic operations and logic gate symbols are summarised in Figure 5. virtual breaker control. as in Figure 5. Note that a truth table with n inputs has 2n rows. 5. of its two inputs are high. Functions such as breaker failure.24 Logic gate symbols NAND and EOR gates. Alternatively customised schemes can be created using the relay logic capabilities. Protection scheme designers may select from a number of pre-programmed logic schemes that perform the most common protection and control requirements.25 shows a typical pre-programmed logic scheme provided with a numerical relay. Table 5. . The NAND is a NOT-AND circuit that is equal to an AND circuit followed by a NOT circuit. A dot (.7 shows the input/output combinations for the gate functions mentioned above. The NOT gate produces an inverted version of the input at its output. The protective elements include phase.) is used to show the AND operation but this dot is usually omitted. The AND gate is an electronic circuit that gives a high output only if all its inputs are high. The outputs of all NAND circuits are high if any of the inputs are low.100 Protection of electricity distribution networks A B AND A B NAND A B EOR A+B A NOT AB A B NOR A AB A B OR A+B A+B Figure 5. A plus sign (+) is used to show the OR operation. the inverted output is known as NOT A and is shown as A. neutral. An encircled plus sign. It is also known as an inverter. The outputs of all NOR gates are low if any of the inputs are high.24.

Overcurrent protection 101 Table 5. However. OUT2 operates for instantaneous phase overcurrent conditions. these features may be achieved by appropriate design of the relay logic. OUT4. . OUT3. 143. but the logic inputs can be programmed to provide this function. and OUT5 operates for timed neutral and negative-sequence overcurrent conditions. neutral and negative-sequence elements are activated to provide timed (51) and instantaneous (50) overcurrent protection in this scheme. Input 1 IN1 is typically assigned to monitor breaker status (52b). Automatic setting group changes are normally based on current level and duration. A setting group can be selected automatically or by using the communication ports or the front panel HMI. This numerical relay has 4 programmable inputs IN1 to IN4. OUT3 trips for timed phase overcurrent situations. and OUT5 are designated to specific function blocks. five programmable outputs OUT1 to OUT5. Setting group changes initiated by contact sensing inputs are not accommodated in this scheme. ten virtual outputs VO6 to VO15. When VO11 becomes TRUE. Contact outputs OUT2. four virtual selection switches 43. 343 and four protection setting groups with external or automatic selection modes. The phase.7a Input/output combinations for various gate functions Inputs A 0 0 1 1 B 0 1 0 1 Outputs AND 0 0 0 1 OR 0 1 1 1 NAND 1 1 1 0 NOR 1 0 0 0 EOR 0 1 1 0 Table 5. Virtual output VO11 is assigned for all protective trips. A function block is disabled by setting the pick-up set-point at zero in each of the four setting groups. OUT4 operates for instantaneous neutral and negative-sequence overcurrent conditions. one programmable alarm output OUTA. OUT1 will operate and trip the breaker. 243.7b Input/output combinations for the NOT gate Output 1 0 Input 0 1 automatic reclosing and protective voltage features are not enabled in this scheme.

11 Adaptive protection with group settings change This section presents several examples of logic scheme customisation to provide functions that normally are not incorporated in numerical relays as part of a manufacturer’s package. the grid infeed not available. is essential. from the coordination curve R1 in Figure 5. one of the transformers out of service for maintenance. adaptive protection. Figure 5. affect the short-circuit levels and therefore an incorrect co-ordination might arise if the relays are not reset for the prevailing power system conditions. Topology changes.25 Typical pre-programmed logic scheme of a numerical relay (reproduced by permission of Basler Electric) 5. Figure 5. For a fault on one of the feeders. To overcome this.27 the feeder relay would operate in a time t1 . With both transformers in service the fault current passing through each transformer would . which can be implemented by using the multiple setting groups feature included in most numerical relays. multiple variables going to the same OR gate are shown by a single line into the OR gate. for example.102 Protection of electricity distribution networks VOA – alarm D0 D1 D2 D3 Auto Sg0 Sg1 Sg2 Sg3 Co-outx VO11 PROT TRIP 6 Output logic Outa 0 Active setting group control VO1 – 52TC Co-outx Output logic Out1 BLK Phase IOC (50TP) Neutral IOC (50TN) Neg seq IOC (50TQ) Phase TOC (51P) Neutral TOC (51N) Neg seq TOC (51Q) 50TPT 50TPPU 50TNT 50TNPU 50TQT 50TQPU 51PT 51PPU 51NT 51NPU Co-outx 51QT 51QPU VO12 PROT PU 6 Co-outx VO5 51N T + 51QT Output logic Out5 VO2 – 50TPT Co-outx Output logic Out2 + – + – + – + – OPTO OPTO OPTO OPTO IN1 IN2 IN3 BLK IN4 BLK VO3 – 51PT Co-outx VO4 50TNT + 50TQT Output logic Out4 Output logic Out3 BLK CO-43 CO-143 CO-243 CO-343 43 143 243 343 BLK BLK Note: For clarity. the in-house generator out of service.26 shows a portion of a power system that might have four scenarios as follows: • • • • system normal. resulting in a fault current of If .

5If If If Figure 5.27 Overcurrent co-ordination curves considering adaptive relaying .26 Electrical system to illustrate setting group change R2a R2 t2a t2 R1 t1 t1 t2 0.Overcurrent protection 103 Grid 4 3 G2 T1 2 T2 1 Figure 5.

the transformer relay tripping time would then reduce from t2 to t2 .2 kV transformers T1 and T3. However.m. For these calculations assume that the L/R ratio is 0.2 Calculate the pick-up setting.2. in order to maintain discrimination with the feeder relays.87 MVA).s. pick-up and time dial settings for the phase relays in order to guarantee a co-ordinated protection system. determine the setting of relays 1. 3. 3 and 4 if the same type of relay is used. The considerations used for Examples 5. The CT in breaker 6 is 100/5. 5 and 8 can be subjected.1 Consider a power system with the same single line diagram as used for Example 5. CT ratios and other data are shown in the same diagram. 2. time dial setting and the instantaneous setting of the phase relays installed in the high voltage and low voltage sides of the 115/13. a) The maximum peak values to which breakers 1. 5. The maximum values of short-circuit current for three-phase faults at busbars A. in the substation illustrated in Figure 5. The short-circuit levels.m. 5. . giving a discrimination margin of (t2 − t1 ).3 For the system shown in Figure 5. leading to incorrect relay operation.2. allowing a time discrimination margin of 0.2 and 5. 5 and 8 can withstand for 5 cycles for guarantee purposes. The instantaneous. when one transformer is out of service. symmetrical (2570. and from then on are in multiples of 100. which is faster than the feeder relay operating time t1 .s. The feeders from substation C each have a capacity of 10 MVA. b) The r. 5. If the short-circuit level at substation A is 1400 MVA. The relay R2 on the low voltage side of the transformer would then operate in time t2 .3 also apply here.5If . 2.89 A r. 5.5 MVA transformer.5 kV line that is protected by the instantaneous element of the overcurrent relay associated with breaker 5. The percentage of the 34. From curve R2 .5 kV 58. taking into account that busbar D has a fault level of 12906. B and C. 4. carry out the following calculations: 1. It is thus necessary to move the transformer low voltage relay curve upwards to R2a .29. where the relay operating time at If will be t2a . but with a 115/34.4 s.28. asymmetrical values that breakers 1. the current through the remaining transformer increases to If . The transformer relays can be pre-programmed to ensure this shift of the relay curve takes place automatically when one transformer is out of service.104 Protection of electricity distribution networks be 0. Take into account that the secondaries are rated at 5 A and that the ratios available in the primaries are multiples of 50 up to 400.12 Exercises 5. The turns ratios of the CTs associated with breakers 1 to 8.

1 115/13.63 115/13.91 kA 1PH SC 1 50N 51N CO-11 3 900/5A 51 W CO-11 1 50N 51N W CO-11 13.9 kA 3PH SC 24 kA 1PH SC (5 MVA) Typical Feeder 3 50 51 W CO-11 50–600/5A 50N 51N W CO-11 Fault Figure 5.09 kA 1PH SC 14.7 kA 3PH SC 2000/5A 51 CO-11 7.2 kV Busbar 21.21 kA 3PH SC 7.2 .6 kA 3PH SC 0 1PH SC 0 1PH SC 300/5A 3 1 50 50N 51N 51 CO-11 CO-11 150/5A 3 50 51 W CO-11 T3 MITSU 41.7 MVA Z% = 10.2 kV 3 16.Overcurrent protection 105 115 kV Busbar 0.28 Single line diagram for Exercise 5.2 kV T1 ASGEN 20 MVA Z% = 9.82 kA 3PH SC 1.

625 Ω/km 34.7 7 34.25 MVA Z % = 6.2 km jO.3 .2 kV 1 2 A 2.5 kV 10.5 kV 5 C 6 (CT 100/5) D Yy 14.106 Protection of electricity distribution networks 2570.87 MVASC 8 115/34.5/13.625 MVA 2.29 Single line diagram for Exercise 5.5 MVA Z % = 11.0 3 13.5 kV 4 B Dy 34.2 kV 5.625 MVA Figure 5.

instantaneous setting 1000 A primary current. with the following characteristics: Pick-up: 1 to 12 in steps of 2 A Time dial: as in Figure 5.Overcurrent protection 107 Bear in mind the following additional information: • • The settings of relay 6 are as follows: pick-up 7 A.5 Isc on busbar A. . time dial 5.15 Instantaneous element: 6 to 144 in steps of 1 A Calculate the setting of the instantaneous elements of the relays associated with the feeders assuming 0. All the relays are inverse time type.

.

then the recloser will stay open after a preset number of operations.1 Equipment The devices most used for distribution system protection are: • • • • overcurrent relays. reclosers are designed to have up to three openclose operations and. prevents a distribution circuit being left out of service for temporary faults.Chapter 6 Fuses.1. even though there are no specific standards for the overall protection of distribution networks. at the most. 6. Thus. In an overhead distribution system between 80 to 95 per cent of the faults are of a temporary nature and last. to interrupt the circuit if the overcurrent persists after a predetermined time. the recloser. and this chapter will cover the other three devices referred to above. sectionalisers. The co-ordination of overcurrent relays was dealt with in detail in the previous chapter. with its opening/closing characteristic. Typically. some general indication of how these systems work can be made. thus isolating the faulted section from the rest of the system. for a few cycles or seconds. 6. after these. reclosers.1 Reclosers A recloser is a device with the ability to detect phase and phase-to-earth overcurrent conditions. reclosers and sectionalisers A wide variety of equipment is used to protect distribution networks. fuses. and then to automatically reclose to re-energise the line. a final open operation to lock out the sequence. The particular type of protection used depends on the system element being protected and the system voltage level. If the fault that originated the operation still exists. . and.

03 0.0 Time.3 0. B and C. The operating time/current characteristic curves of reclosers normally incorporate three curves.1 shows a typical set of time/current curves for reclosers.07 0. designated as A.7 0.01 50 70 100 200 300 400 500 700 1000 2000 3000 4000 5000 Current.1 Time/current curves for reclosers . The counting mechanisms register operations of the phase or earth-fault units which can also be initiated by externally controlled devices when appropriate communication means are available. However.5 0.02 A 0. new reclosers with microprocessor-based controls may have keyboard-selectable 20 10 7 5 3 2 1. respectively.05 0. A Figure 6.110 Protection of electricity distribution networks One further closing operation by manual means is usually allowed. s 0.10 B 0.2 C 0. Figure 6. one fast and two delayed.

the time-delay operation allows other protection devices nearer to the fault to open. depending on its setting. limiting the amount of the network being disconnected. Reclosers with hydraulic operating mechanisms have a disconnecting coil in series with the line. Vacuum and SF6 reclosers have the advantage of requiring less maintenance. The time characteristic and operating sequence of the recloser are dependent on the flow of oil in different chambers. the coil attracts a piston that opens the recloser main contacts and interrupts the circuit. Generally. One method is to use CTs connected residually so that the resultant residual current under normal conditions is approximately zero. vacuum or SF6 . when a fault occurs. When the current exceeds the predetermined setting. as would occur during earth faults. Single-phase reclosers are used when the load is predominantly single-phase. the smallest section of the circuit is disconnected to minimise disruption of supplies to customers. This allows reprogramming of the characteristics to tailor an arrangement to a customer’s specific needs without the need to change components. Reclosers can be classified as follows: • • • single-phase and three-phase. The three later ones operate in a timed manner with predetermined time settings. reclosers and sectionalisers 111 time/current curves which enable an engineer to produce any curve to suit the coordination requirements for both phase and earth-faults. The first shot is carried out in instantaneous mode to clear temporary faults before they cause damage to the lines. therefore. After selecting the size and sequence of operation of the recloser. Oil reclosers use the oil to extinguish the arc and also to act as the basic insulation. The same oil can be used in the control mechanism. In such a case. Reclosers with electronic operating mechanisms use a coil or motor mechanism to close the contacts. . when a single-phase fault occurs the recloser should permanently disconnect the faulted phase so that supplies are maintained on the other phases. oil. If the fault is permanent. Co-ordination with other protection devices is important in order to ensure that. a delayed shot is initiated that finally results in a tripping signal being transmitted to the recloser control mechanism.2. Three-phase reclosers are used when it is necessary to disconnect all three phases in order to prevent unbalanced loading on the system. The control circuit determines the subsequent opening and closing of the mechanism. When the current exceeds the setting value. The recloser should operate when the residual current exceeds the setting value.Fuses. mechanisms with hydraulic or electronic operation. it is important that the recloser has an appropriate sensitivity to detect them. the devices downstream are adjusted in order to achieve correct co-ordination. Earth faults are less severe than phase faults and. the time characteristic and the sequence of operation of the recloser are selected to co-ordinate with mechanisms upstream towards the source. The electronic type of control mechanism is normally located outside the recloser and receives current signals from a CT-type bushing. A typical sequence of a recloser operation for a permanent fault is shown in Figure 6.

2 Typical sequence for recloser operation .Reclosing intervals (contacts open) Load current (contacts closed) 112 Protection of electricity distribution networks Start of fault Fault current Timed operation (contacts closed) Rapid operation (contacts closed) Figure 6.

1. Sectionalisers are constructed in single. Co-ordination with other mechanisms located upstream towards the source.2 Sectionalisers A sectionaliser is a device that automatically isolates faulted sections of a distribution circuit once an upstream breaker or recloser has interrupted the fault current and is usually installed downstream of a recloser. Sectionalisers count the number of operations of the recloser during fault conditions. System voltage. A sectionaliser does not have a current/time operating characteristic. and downstream towards the load. in order to permit the sectioning of long lines and thus prevent the loss of a complete circuit due to a fault towards the end of the circuit. 6. When installing reclosers it is necessary to take into account the following factors: 1. This permits the recloser to close and re-establish supplies to those areas free of faults.or three-phase arrangements with hydraulic or electronic operating mechanisms. to provide primary protection for a circuit. 6. Short-circuit level. 2. Sectionalisers with hydraulic operating mechanisms have an operating coil in series with the line. in main feeder circuits. to prevent the tripping of the main circuit due to faults on the spurs. the operating mechanism of the sectionaliser is reset. Sensitivity of operation for earth faults. Minimum short-circuit current within the zone to be protected by the recloser. the sectionaliser opens and isolates the faulty section of line. the values that exist at the point of installation. 3. After a preselected number of recloser openings. The voltage rating and the short-circuit capacity of the recloser should be equal to. If the fault is temporary. they must be used with a back-up device that has fault current breaking capacity. 5. Maximum load current. It is also necessary to ensure that the fault current at the end of the line being protected is high enough to cause operation of the recloser. in branches or spurs. the sectionaliser contacts are opened by means of .Fuses. and can be used between two protective devices whose operating curves are very close and where an additional step in co-ordination is not practicable. reclosers and sectionalisers 113 Reclosers are used at the following points on a distribution network: • • • in substations. Each time an overcurrent occurs the coil drives a piston that activates a counting mechanism when the circuit is opened and the current is zero by the displacement of oil across the chambers of the sectionaliser. or greater than. 4. The same criteria should be applied to the current capability of the recloser in respect of the maximum load current to be carried by the circuit. and while the recloser is open. Since sectionalisers have no capacity to break fault current. After a prearranged number of circuit openings.

The majority of fuses used in distribution systems operate on the expulsion principle. The presence of de-ionising gases. there are several sections of ANSI/UL 198-1982 standards that cover low voltage fuses of 600 V or less. with the interior covered with de-ionising fibre. the lower limit based on the minimum time required for the fusing of the element (minimum melting time) with the upper limit determined by the maximum total time that the fuse takes to clear the fault. time/current characteristics.1. In the presence of a fault. In this way. and the turbulence within the tube. eliminate the arc established during the destruction of the element and then maintain circuit conditions open with nominal voltage applied to its terminals. 6. The zone of operation is limited by two factors. The arc is compressed and expelled out of the tube. A suitably selected fuse should open the circuit by the destruction of the fuse element. 3. Maximum short-circuit level.114 Protection of electricity distribution networks pretensioned springs. This type of sectionaliser is constructed with manual or motor closing.e.e. ensure that the fault current is not re-established after the current passes through zero point. in addition. For medium and high voltage fuses within . The load current is measured by means of CTs and the secondary current is fed to a control circuit which counts the number of operations of the recloser or the associated interrupter and then sends a tripping signal to the opening mechanism. This type of sectionaliser can be closed manually. rated currents. the arc is extinguished when current zero is reached. Co-ordination factors that need to be taken into account include the starting current setting and the number of operations of the associated interrupter before opening. and a fusible element. The nominal voltage and current of a sectionaliser should be equal to or greater than the maximum values of voltage or load at the point of installation. Maximum load current. The maximum clearance time of the associated interrupter should not be permitted to exceed the short-circuit rating of the sectionaliser. i. no arcing across the fuse element). manufacturing features and other considerations. it possesses an element that is directly heated by the passage of current and is destroyed when the current exceeds a predetermined value. Sectionalisers with electronic operating mechanisms are more flexible in operation and easier to set. System voltage. For example. Co-ordination with protection devices installed upstream and downstream. 4. the escape of gas from the ends of the tube causes the particles that sustain the arc to be expelled. 2. they have a tube to confine the arc. There are a number of standards to classify fuses according to the rated voltages.3 Fuses A fuse is an overcurrent protection device. (i. The following factors should be considered when selecting a sectionaliser: 1. the interior fibre is heated up when the fusible element melts and produces de-ionising gases which accumulate in the tube. The shortcircuit capacity (momentary rating) of a sectionaliser should be equal to or greater than the fault level at the point of installation.

01 100 200 300 500 700 1000 2000 3000 5000 7000 10000 Current. in addition. s Time. respectively.2 0. A Figure 6.7 0.07 0.5 0.02 0.1 s to the minimummelt current for 300 s operation.03 0.07 0. Figure 6.03 0.5 0. 46. s 100 200 300 500 700 1000 2000 3000 5000 7000 10000 1 0.40. is very popular.3–138 kV. standards such as ANSI/IEEE C37. The speed ratio is the ratio of minimum melt current that causes fuse operation at 0. a speed ratio (SR) of 6-8 is defined and. 10-13.3 shows the comparative operating characteristics of type 200 K and 200 T fuse links. For the K link.3 0.05 0. A Current. fuse manufacturers have their own classifications and designations. (b) 200 T fuse link .3 Time/current characteristics of typical fuse links: (a) 200 K fuse link.1 0. the use of fuse links designated K and T for fast and slow types.01 0.3 0. For the 200 K fuse a 4400 A current is required (a) 1000 700 500 300 200 (b) 1000 700 500 300 200 100 70 50 30 20 100 70 50 30 20 10 7 5 3 2 10 7 5 3 2 Time. reclosers and sectionalisers 115 the range 2.02 0. 47 and 48 apply. Other organisations and countries have their own standards. In distribution systems.Fuses. for a T link.7 0. 41. 42.2 1 0.05 0.1 0. depending on the speed ratio.

In unearthed systems. Load current.86. for this case. 4. or greater than.116 Protection of electricity distribution networks for 0. the maximum phase-to-phase voltage. Therefore. or greater than. 6500 A is required for 0. the maximum line-to-earth voltage and for threephase loads the nominal voltage is selected on the basis of the line-to-line voltage. The following information is required in order to select a suitable fuse for use on the distribution system: 1. and 520 A for 300 s. For the 200 T fuse. Voltage and insulation level. The above four factors determine the fuse nominal current. Selection of nominal current The nominal current of the fuse should be greater than the maximum continuous load current at which the fuse will operate. the nominal voltage of the fuse. the SR is 12.5. the following criteria should be used: 1. 2. 2. . Such a voltage should be equal to. Some manufacturers have produced tables to assist in the proper fuse selection for different ratings and connection arrangements. the symmetrical fault current calculated for the point of installation of the fuse. number of phases (three or one). for single-phase loads the nominal voltage should be equal to. In the case of power transformers. Type of system. Maximum short-circuit level. 3.1 s clearance time and 560 A for 300 s. or greater than. The system characteristics determine the voltage seen by the fuse at the moment when the fault current is interrupted. type of earthing. voltage and shortcircuit capability characteristics.1 s clearance. fuses should be selected such that the time/current characteristic is above the inrush curve of the transformer and below its thermal limit. An overload percentage should be allowed according to the protected-equipment conditions. Selection of nominal voltage The nominal voltage of the fuse is determined by the following system characteristics: • • • maximum phase-to-phase or phase-to-earth voltage. the nominal voltage should be equal to. Selection of short-circuit capacity The symmetrical short-circuit capacity of the fuse should be equal to. giving an SR of 7. In three-phase earthed systems. or less than.

However.4. is called the back up. towards the source.Fuses.2 Recloser-fuse co-ordination The criteria for determining recloser-fuse co-ordination depend on the relative locations of these devices.e.) 2.2.2. This ensures that the main fuse interrupts and clears the fault before the back-up fuse is affected in any way. 6. The main protection should clear a permanent or temporary fault before the backup protection operates. the graphic method is still popular not only because it gives more information but also because computer-assisted design tools make it much easier to draw out the various characteristics. The factor of 75 per cent compensates for effects such as load current and ambient temperature. and that upstream. if the main protection is a fuse and the back-up protection is a recloser. The criteria for co-ordinating them will be discussed later. However. Loss of supply caused by permanent faults should be restricted to the smallest part of the system for the shortest time possible. These possibilities are treated in the following paragraphs. as indicated in Figure 6.2. . (See Section 6. which proved to be an easy and accurate method. The co-ordination between two or more consecutive fuses can be achieved by drawing their time/current characteristics. co-ordination tables with data of the available fuses were also used. if the fault is not cleared. it is normally acceptable to co-ordinate the fast operating curve or curves of the recloser to operate first. or continue to operate until the circuit is disconnected. reclosers and sectionalisers 117 Fuse notation When two or more fuses are used on a system. i.1 Fuse-fuse co-ordination The essential criterion when using fuses is that the maximum clearance time for a main fuse should not exceed 75 per cent of the minimum melting time of the backup fuse. the device nearest to the load is called the main protection. 6. or vice versa. followed by the fuse. 6. In the following sections criteria and recommendations are given for the co-ordination of different devices used on distribution systems. or fatigue in the fuse element caused by the heating effect of fault currents that have passed through the fuse to a fault downstream but were not sufficiently large enough to melt the fuse. for the same current level.2 Criteria for co-ordination of time/current devices in distribution systems The following basic criteria should be employed when co-ordinating time/current devices in distribution systems: 1. whether the fuse is at the source side and then backs up the operation of the recloser that is at the load side. normally on log-log paper as for overcurrent relays.2. In the past.

5 should be taken into account. Fuses at the load side The procedure to co-ordinate a recloser and a fuse. even so. Normally it is easier to shift the fuse curve.75 t2 Fuse at the source side When the fuse is at the source side. Some values proposed by Cooper Power Systems are reproduced in Table 6. all the recloser operations should be faster than the minimum melting time of the fuse. either the fuse or the recloser curve should be shifted horizontally on the current axis to allow for the transformer turns ratio.118 Protection of electricity distribution networks t2 Time. the considerations given in Section 5. The recloser opening curve modified by the appropriate factor then becomes slower but. when the latter is at the load side. A Figure 6. given in Table 6. This is illustrated in Figure 6.5.4 Criteria for fuse-fuse co-ordination: t1 < 0. s t1 Current. is carried out with the following rules: • the minimum melting time of the fuse must be greater than the fast curve of the recloser times the multiplying factor. if the transformer connection group is delta-star. should be faster than the fuse curve. The multiplying factors referred to above depend on the reclosing time in cycles and on the number of the reclosing attempts.2 and taken from the same reference as above.1. It is convenient to mention that if the fuse is at the high voltage side of a power transformer and the recloser at the low voltage side. based on the transformer tap that produces the highest current on the high voltage side. . On the other hand. This can be achieved through the use of multiplying factors on the recloser time/current curve to compensate for the fatigue of the fuse link produced by the cumulative heating effect generated by successive recloser operations.

50 2.20 3. reclosers and sectionalisers 119 R Fuse Transformer R Recloser C C C times k Time.35 four delayed sequence Reclosing time in cycles 25 30 50 90 120 240 600 2.40 1.80 1. two delayed sequence one fast.10 1.70 3.5 Criteria for source-side fuse and recloser co-ordination Table 6.10 1.40 1.35 The k factor is used to multiply the time values of the delayed curve of the recloser.70 2.70 1.70 2.90 1.85 1.50 2. .35 3.1 k factor for source-side fuse link Multipliers for: two fast.60 2. s A Isc max at recloser installation Current.10 2. A Figure 6.20 1. three delayed sequence 3.45 1.Fuses.

2. A less effective coordination is obtained using one instantaneous operation followed by three timed operations. separation greater than 12 cycles ensures nonsimultaneous operation.25 1.80 1.3 Recloser-recloser co-ordination The co-ordination between reclosers is obtained by appropriately selecting the amperes setting of the trip coil in the hydraulic reclosers.25 1.35 1. while the second will clear a further 10 per cent . or of the pick-ups in electronic reclosers. In small reclosers. • the maximum clearing time of the fuse must be smaller than the delayed curve of the recloser without any multiplying factor.6. Hydraulic reclosers The co-ordination margins with hydraulic reclosers depend upon the type of equipment used. separation of the curves by between two and 12 cycles could result in simultaneous operation.120 Protection of electricity distribution networks Table 6. 6.2 k factor for the load-side fuse link Multipliers for: one fast operation 25–30 60 90 120 1. The load fuses are set to operate before the third opening.35 1. where the current coil and its piston produce the opening of the contacts.35 Reclosing time in cycles The k factor is used to multiply the time values of the recloser fast curve.25 two fast operations 1. the first opening of a recloser will clear 80 per cent of the temporary faults. . Better co-ordination between a recloser and fuses is obtained by setting the recloser to give two instantaneous operations followed by two timed operations. In general. The application of the two rules is illustrated in Figure 6. clearing permanent faults. the recloser should have at least two or more delayed operations to prevent loss of service in case the recloser trips when the fuse operates. the following criteria must be taken into account: • • • separation of the curves by less than two cycles always results in simultaneous operation.25 1.

the co-ordination margins are as follows: • • separation of the curves by less than two cycles always results in simultaneous operation.6 Criteria for recloser and load-side fuse co-ordination With large capacity reclosers. . a separation of more than eight cycles guarantees non-simultaneous operation. s Isc max at fuse installation Current. A Figure 6. The principle of co-ordination between two large units in series is based on the time of separation between the operating characteristics. In such cases. in the same way as for small units. the piston associated with the current coil only actuates the opening mechanism. reclosers and sectionalisers 121 R R Recloser Fuse C A A times k A Time.Fuses.

4 Recloser-relay co-ordination Two factors should be taken into account for the co-ordination of these devices.5 per cent. and the clearance time of the downstream recloser plus its tolerance should be lower than the upstream recloser clearance time less its tolerance.5 per cent . so that the net percentage of relay operation after the third opening of the recloser = 50 per cent − 12. The percentage of relay reset that takes place during the recloser interval is (2 s/16 s) × 100 per cent = 12.6 s) × 100 per cent = 5 per cent . The downstream recloser must be faster than the upstream recloser.).5 per cent = 37.122 Protection of electricity distribution networks Electronically-controlled reclosers Adjacent reclosers of this type can be co-ordinated more closely since there are no inherent errors such as those that exist with electromechanical mechanisms (due to overspeed. which is required to co-ordinate with an inverse time-delay overcurrent relay that takes 0.30 s. consider a recloser with two fast and two delayed sequence with reclosing intervals of two seconds.3 sec. 6. The latter should be set with the same. The percentage of the relay operation during which each of the two rapid recloser openings takes place is (0. and the delayed operating time is 0./0. and 16 s to completely reset.6 s to close its contacts at the fault level under question. etc. It should be noted that the criteria of spacing between the time/current characteristics of electronically controlled reclosers are different to those used for hydraulically controlled reclosers. The rapid operating time of the recloser is 0. Normally. The impulse margin time of the relay is neglected for the sake of this illustration. in order to clear temporary faults on the line between the substation and the load recloser. the relay will move towards its operating point from this partially reset position. The relay reset for the third opening of the recloser = 12. or a larger. the setting of the recloser at the substation is used to achieve at least one fast reclosure. and the total percentage of the relay operation after the fourth opening of the recloser = 37. inertia. and therefore co-ordination is guaranteed. . as previously. The reset time of the relay is normally long and. the relay completely resets after both of the two rapid openings of the recloser.2. The percentage of the relay operation during the second time delay opening of the recloser takes place = (0.5 per cent.6 sec) × 100 per cent = 50 per cent . For example.6 s) × 100 per cent = 50 per cent. From the above analysis it can be concluded that the relay does not reach 100 per cent operation by the time the final opening shot starts. The percentage of the relay operation during the first time-delay opening of the recloser is (0.030 s.3 s/0.03 s/0. the interrupter opens the circuit some cycles after the associated relay trips.5 per cent. and the relay has to integrate the clearance time of the recloser. number of rapid operations as the recloser at the substation.5 per cent + 50 per cent = 87. if the fault current is re-applied before the relay has completely reset. Therefore.

If a permanent fault occurs beyond the sectionaliser. These operations can be any combination of rapid or timed shots as mentioned previously. its fuse should operate first. the k factor for two fast . If the fault is still not cleared. reclosers and sectionalisers 123 6. are shown in Figure 6.1 Figure 6.8.2. their coordination does not require an analysis of these curves. The time/current curves for the transformer and branch fuses. being backed up by the recloser fast operating shots.7 shows a portion of a 13. For a fault at the distribution transformer. the sequence of operation of the recloser should be adjusted in order to obtain the appropriate co-ordination for faults beyond the fuse by following the criteria already mentioned. The sectionaliser will isolate the faulted section of the network after the full number of counts has elapsed. In turn. which results in operation of the branch fuse in 0.2 kV distribution feeder that is protected by a set of overcurrent relays at the substation location.2 kV is 4.75 (6.1.5 Recloser-sectionaliser co-ordination Since the sectionalisers have no time/current operating characteristic. as indicated in Section 6. The recloser will then re-energise the section to restore the circuit. a 6 T fuse was selected on the basis of allowing a 20 per cent overload. then the branch fuse should operate next followed by the delayed opening shots of the recloser and finally by the operation of the feeder relay. At the branch fuse location the short-circuit current is 2224 A. A fault beyond the last sectionaliser results in the operation of the recloser and the start of the counters in all the sectionalisers. for example three disconnections in this case.75 factor is used in order to guarantee the co-ordination of the branch and transformer fuses.02 s.5 kVA distribution transformer at 13. The recloser chosen has two fast and two delayed operations with 90 cycles intervals. for example two fast and two delayed.2. the recloser and the relays. The fast curve of the recloser was chosen with the help of the following expression based on the criteria already given.2. The co-ordination criteria in this case are based upon the number of operations of the back-up recloser. A recloser and a sectionaliser have been installed downstream to improve the reliability of supply to customers. The sectionaliser should be set for one shot less than those of the recloser. the furthest recloser should be adjusted for a smaller number of counts.6 Recloser-sectionaliser-fuse co-ordination Each one of the devices should be adjusted in order to co-ordinate with the recloser.Fuses. Example 6. 6. which guarantees that it lies in between the curves of both fuses: trecloser × k ≤ tMMT of branch fuse × 0. If additional sectionalisers are installed in series.1) where tMMT of branch fuse is the minimum melting time.9 A. As the nominal current of the 112. leaving that part of the feeder upstream still in service. the sectionaliser will open and isolate the fault after the third opening of the recloser. From Table 6. The 0.2.

124 Protection of electricity distribution networks 31993 ASC 1 50 51 31993 ASC 2 50 51 7066 ASC 3 Recloser R 2 fast 2 delayed S 3 Operations 2224 ASC 4 Branch fuse 2097 ASC 5 Transf. fuse 112.7 Portion of a distribution feeder for Example 6.1 .5 kVA R Recloser Transformer S Sectionaliser Fuse Figure 6.

fuse 0. relay Feeder relay Recloser fast Recloser (delay) Transf. but the curves of Figure 6. 6.3 = 1.0 Tap 2000/5CT Inst.8 Phase-current curves for Example 6. overridden Feeder relay Siemens 7SK-88 VI 0.75/1. reclosers and sectionalisers 125 1000 5 4 1 LV Transf. and so that the relay reset time is considered.5 × Icond × RTC 600/5 CT 100 2 3A 2 10 1 3A Recloser (delay) I(Start) = 240 I(Start) = 1.02 × 0.35) = 0. and the pick-up current of the recloser. This time.8 shows that adequate co-ordination has been achieved.1 Branch fuse LVtransf.01 0.1 the maximum time for the recloser operation is (0. Detailed calculations have not been given for this particular example since the procedure was indicated in Chapter 5. load 4 5 Branch fuse Kearney T 40E Transf. The feeder relay curve is selected so that it is above that of the delayed curve of the recloser.1 operations and a reclosing time of 90 cycles is 1. from eqn. fuse Kearney T 6E Time in seconds 1 3B 0.5 × Inom. relay West CO-11 2.011 s. . With these values.5 1 10 100 1000 10000 Current in Amperes × 10 Figure 6.30 TD 1 Tap 4.35. determines the fast curve of the recloser.5 × Inom. load 3B Recloser fast I(Start) = 240 I(Start) = 1.Fuses.0 TD 6.

.

and when bi-directional overcurrent protection could produce unnecessary disconnection of circuits. The use of directional overcurrent relays in the two situations is shown in Figure 7.1 Construction Directional overcurrent relays are constructed using a normal overcurrent unit plus a unit that can determine the direction of the power flow in the associated distribution system element. (b) multisource system . In addition to the relay current this second unit usually requires a reference signal to measure the angle of the fault and thus determine whether or not (a) (b) Figure 7.1 Application of directional overcurrent relays: (a) ring system. 7.1.Chapter 7 Directional overcurrent relays Directional overcurrent protection is used when it is necessary to protect the system against fault currents that could circulate in both directions through a system element. This can happen in ring or mesh-type systems and in systems with a number of infeed points.

then the fluxes are out of phase by 90◦ . However. since the overcurrent unit only picks up when the flow is in the correct direction. 1 being proportional to the current and 2 proportional to the voltage. the voltage of that phase might collapse. the angle for maximum torque occurs when the current and voltage of the relay are in phase. common practice to use the current from a different phase. 7. With the supervision method. with the angle between 1 and 2 . If I and V are in phase. . for a fault on one phase. AMT. both cases are illustrated in Figure 7. In addition. Basically. The angle of maximum torque.128 Protection of electricity distribution networks (a) OC D OC D D OC (b) D OC Figure 7. the overcurrent unit can pick up for the wrong power flow direction. This can be obtained very simply by using the current and voltage from the same phase.2 where D indicates the directional unit and OC the overcurrent unit.3 Relay connections The connection of a directional relay is defined on the basis of the degrees by which a current at unity power factor leads the polarisation voltage. It is better to use the control system to determine the direction of the power flow. It is. It should be noted that is in phase with I but lagging with respect to the voltage since V = −(d )/dt. (b) by control the relay should operate. the reference or polarisation signal is a voltage but this can also be a current input. when a breaker is opened in a ring system the current flows will change and this could lead to the consequential possibility of loss of co-ordination. there are two methods for obtaining the direction of the power flow – supervision and control.2 Obtaining the direction of power flow: (a) by supervision. 7. Therefore. The torque is positive if 0 < < 180◦ and negative if 180◦ < < 360◦ .2 Principle of operation The operating torque can be defined by T = K 1 2 sin where 1 and 2 are the polarising values. is the angle for which this displacement produces the maximum torque and therefore is always aligned with the polarisation voltage. therefore. this is not practical since. Generally.

Vac B: Ibc . and because they offer no advantages compared to the previous case. Angle of operation: current angles from 60◦ leading to 120◦ lagging. or Ia leading 30◦ or lagging 150◦ at unity power factor. provided that it has three elements.1 30◦ connection (0◦ AMT) Feeding the relays: A: Ia . Use: it is recommended that relays with this connection be used solely on feeders (see Figure 7. since two phase elements and one earth element can give rise to poor operation (see Figure 7. i.2 60◦ connection (0◦ AMT) Feeding the relays: A: Iab . Ia lags Va by 60◦ at unity power factor. For this reason. Iab lags Vac by 60◦ . Vac B: Ib .4). The three-phase unit arrangement should not be used in transformer circuits where some faults can result in a reverse current flow in one or more phases leading to relay mal-operation.3 Vector diagram for 30◦ connection (0◦ AMT) 7. one per phase. However. Use: this type of connection should always be used on feeders. Angle of operation: current Iab from 30◦ leading to 150◦ lagging. they are little used. 7.3.3).e.3. . Vba C: Ica Vcb Maximum torque: when the current lags the phase-to-neutral voltage by 60◦ .Directional overcurrent relays 129 Ia Va Zero torque line 60° 30° AMT Vac Vc Vb Figure 7. they have the disadvantage that the CTs have to be connected in delta. Vba C: Ic Vcb Maximum torque: when the phase current lags the phase-neutral voltage by 30◦ .

Use: in feeders where the source of zero-sequence components is behind the point of connection of the relay (see Figure 7. . Vca + 30◦ C: Ic Vab + 30◦ Maximum torque: when the current lags the phase-to-neutral voltage by 60◦ .4 Vector diagram for 60◦ connection (0◦ AMT) Iab Zero torque line 30° Vbc 30° Vbc 30° Vc 30° Vb Va AMT Figure 7. Angle of operation: current angles from 30◦ leading to 150◦ lagging. Vbc + 30◦ B: Ib .3.5).130 Protection of electricity distribution networks Iab Va Zero torque line 60° AMT Vac 30° Vc Vb Figure 7.3 90◦ connection (30◦ AMT) Feeding the relays: A: Ia .5 Vector diagram for 90◦ connection (30◦ AMT) 7.

This configuration should always be used when single-phase directional relays are applied to circuits that can have a 2-1-1 current distribution. however. Angle of operation: current angles from 45◦ leading to 135◦ lagging.4 90◦ connection (45◦ AMT) Feeding the relays: A: Ia . Use: this arrangement is recommended for protecting transformers or feeders that have a source of zero-sequence components in front of the relay (see Figure 7. This is based on the principle that the neutral current always flows towards the system from earth whereas. current polarisation is employed using the earth current from a local transformer connected to earth. . This connection is essential in the case of transformers in parallel. 7.6). and displaced with respect to the residual current for the angle characteristic of the source impedance. or transformer feeders.4 Directional earth-fault relays Directional earth-fault relays are constructed on the basis that the residual voltage is equal to three times the zero-sequence voltage drop in the source impedance. It should be stressed. Vbc + 45◦ B: Ib . especially for guaranteeing correct operation of the relays for faults beyond Yd transformers.3. that the possibility of the failure of a voltage-polarised directional protection relay is minimal and it is therefore recommended that this arrangement should be used wherever possible. Vab + 45◦ Maximum torque: when the current lags the phase-to-neutral voltage by 45◦ .Directional overcurrent relays 131 Ia Zero torque line Va 45° Vbc AMT 30° Vc Vbc Vb 15° Figure 7.6 Vector diagram for 90◦ connection (45◦ AMT) 7. depending on the fault. the residual current may flow in any direction. When a set of suitable VTs is not available for obtaining the polarisation voltage. Vca + 45◦ C: Ic .

the installation of a directional overcurrent relay is justified by assuming that this circuit would be part of a ring in a future system.3 X2 = 0. The bases at the generator location are 13. Although the system is radial.B.11 X0 = 0.8 Z0 = 0. VB = 0. VBA C: IC VCB In addition. Determine the current and voltage signals (in amps and volts) that go into each one of the directional relays that have a 30◦ connection and are fed as indicated below: A: IA .2 kV 67 115000/ √3 115√3 G X1 = 0.102 .5 ⇒ Z0 = 0.132 Protection of electricity distribution networks T1 A 500/5 F L B T2 G 13.3 Figure 7.2/115 kV 115/13.2 T2 X1 = 0.2 kV and 100 MVA.15 X2 = 0.1 A solid earth-fault on phases B and C is represented by the arrow at the point F in the power system in Figure 7. are IA = 0.11 X2 = 0. VAC B: IB .11 1. In the solution.8. B-C-N.11 L X1 = 0.7. The equivalent circuit is obtained by connecting the three sequence networks in parallel as shown in Figure 7. indicate which relays operate on the occurrence of the fault.09 T1 X1 = 0. ignore load currents and assume a pre-fault voltage equal to 1 p. Solution The conditions for a double phase-to-earth fault. From Figure 7.4 X0 = 1.4 X2 = 0.3 X0 = 0.1 Example 7.u. VC = 0 The three sequence networks are shown in Figure 7.9.7 Single line diagram of system for Example 7. N.15 X0 = 0.

Directional overcurrent relays 133 j0.4 B j0.4 B j0.u.102 IA0 1 p.26 + ( j0.9 Equivalent circuit for Figure 7.26 = j2.11 A F j1.845 j0.15 j0. j0.8 so that the three sequence currents in the network are 1 = −j3.09 j0.u.102 = j0.0) j0.0p.1 IA1 j0.3 Figure 7.15 j0.26 + j0.0) IA0 = −(−j3.102 .155 j0.2 B j0.362)) IA2 = −(−j3.3 1 p.102 j0.102)/( j0. IA1 = ( j0.26 IA2 j0.u.11 A F j0.11 A F j0.26 + j0.3 j0.26 j0. Figure 7.8 Sequence networks for Example 7.26 × j0.

14◦ × 502.155 ⇒ IA = 0 as was to be expected for a B-C-N fault IB = a 2 IA1 + aIA2 + IA0 = 1 240◦ (−j3.2315 = 4.86◦ A VBA = −0.155 = −3.845) + j2.66 VBA = VB − VA = −VA = −0.22 = 0.66 VB = VC = 0 VAC = VA − VC = VA − 0 = 0.86◦ IC = aIA1 + a 2 IA2 + IA0 = 1 120◦ (−j3.64 135.64 44.134 Protection of electricity distribution networks At the point of the fault: IA = IA1 + IA2 + IA0 = −j3.66 VCB = VC − VB = 0 The bases at the point of fault are V = 115 kV.66 × √ 3 3 115000 IB = 4.2315 = 4.845) + j2.04 × (5/500) = 23.33 + j3.155 = 3.04 A 3V 3 × 115 × 103 Therefore.0) + 1 120◦ ( j0. the values at the point of the fault are IA = 0 √ 3 115000 115 ×√ × = 43.26) = 0.220 VA = VA1 + VA2 + VA0 = 3VA1 = 3 × 0.0) + 1 240◦ ( j0.14◦ At the point of fault on the network (not at the relay): VA1 = VA2 = VA0 = −IA2 ( j0.64 135.04 × (5/500) = 23.82 0◦ V VAC = 0.33 + j3.86◦ × 502.66 × 1 115000 × = 43.82 180◦ V √ 1000 3 IC = 4.14◦ A VCB = 0 .845 + j2.29 44.0 + j0. P = 100 MVA 100 × 106 P IBase = √ =√ = 502.29 135.64 44.

22 = 0.33 + j3.u.845 210◦ + 2.6077 j0.0) + 1 240◦ ( j0.077 = 4.6077) = −j4.66 VBA = VB − VA = −VA = −0.26◦ IC = aIA1 + a 2 IA2 + IA0 = 1 120◦ (−j3.11) = 0.22 VA0 = −IA0 ( j0.33 + j3. in this case.48 j0.0 150◦ + 0.26) = 0. as there are at the fault point.155 and it will be seen that. so that at the relay IA = IA1 + IA2 + IA0 = −j3.66 VB = VC = 0 The CT is fed from the same fault point.0 ( j0.845 + j2 = −j0.61) = j2 p.26 = 0.845 −30◦ + 2.11) = −j2. the zero-sequence current in the relay itself is different because of the division of current.0 + j0.73◦ VA1 = VA2 = IA2 ( j0.. IAB = j2. so that VAC = VA − VC = VA − 0 = 0.0 90◦ = −3. In the relay.66 VCB = VC − VB = 0 1 = −j1.845) + j2 = 3.22 VA = VA1 + VA2 + VA0 = 3VA1 = 3 × 0.0) + 1 120◦ ( j0.26 + j0.5/1.102 1 = −j3.845) + j2 = 3. however.0 30◦ + 0.534 137. IA = 0: IB = a 2 IA1 + aIA2 + IA0 = 1 240◦ (−j3.and negativesequence current values.26 + j0.823 At the point where the relay is located there will be equal positive.534 42.26 I1φ = IA1 + IA2 + IA0 = 3IA1 .154 (1.845 × 0.0 90◦ = 3.077 = 4.Directional overcurrent relays 135 Corollary I3φ = and IA1 = so that I1φ = 3 × (−j1.

76 137.66 × C 1 115000 × = 43.04 × (5/500) = 22.26◦ × 502.728 −90◦ A VAC = 43.82 0◦ V 3 115000/ 3 B IB = 4.10) thus creating some doubt about the operation of its directional unit.04 × (5/500) = 0.155 −90◦ × 502. −90◦ < angle of IA < 90◦ .73◦ × 502.136 Protection of electricity distribution networks The signals that feed the relay are A IA = 0.82 −0◦ V For operation.10 Analysis of operation of relay for phase A . +90˚ VRT –90˚ IR Figure 7.76 42.66 × √ √ = 43.73◦ A VBC = 0 Analysis of operation of directional relays: Polarisation: A B C IA VAC IB VBA IC VCB Phase A relay: IA = 0.534 137.04 × (5/500) = 22.534 42.26◦ A VBA = −0.82 180◦ V √ 1000 3 IC = 4.728 −90◦ A √ 115 3 115000 × VAC = 0. The relay in phase A is at the limit of functioning (see Figure 7.

taking into account the DC transient component of the current.Directional overcurrent relays 137 IS 137. 7. Example 7. multiplied by a safety overload factor in order to maintain co-ordination. the relays installed at substations adjacent to the source substation should never register any current from the substation towards the source. see Figure 7. The criterion used here is the same as for setting bi-directional overcurrent relays protecting lines between substations.76 137.82 180◦ V For operation. since the angle of IB is 137.11 Analysis of operation of relay for phase B Phase B relay: IB = 22. A lower value should not be used as this could result in a false trip if the directional element picks up inadvertently under severe load transfer conditions.26◦ A VBA = 43.12 determine the maximum load currents. When the ring has only one source.26◦ . 90◦ < angle of IB < 270◦ .5 times the maximum load current.5 Co-ordination of instantaneous units The calculations for the setting of an instantaneous unit in a ring system are carried out using the short-circuit level at the next relay downstream. with the ring open. for these relays. it is recommended that. the instantaneous units should be set at 1. Phase C relay: this does not operate because VCB = 0. With these settings the instantaneous units then have the same pick-up current value as that of the time-delay units so that co-ordination is not compromised. the transformation ratios of the CTs and the current setting of the instantaneous units . Therefore. The relay in phase B operates.26° 90° VSR 180° 270° Figure 7.2 For the system shown in Figure 7.11.

The instantaneous settings should be given in primary amperes.38 A √ total load current from K to L = (7 MVA)/ 3(13200 V) = 306.08 A Imax 3 = 437.12 Single-line diagram for a ring system with one infeed for Example 7. since reference to specific relay models is not given. and thereafter in multiples of 200.38 A Imax 4 = 437.2 to guarantee a co-ordinated protection scheme.08 A √ total load current from L to K = (8 MVA)/√3(13200 V) = 349.38 A Imax 5 = 349.9 A total load current from K to J = (5 MVA)/ 3(13200 V) = 218.08 A total load current from J to K = (10 MVA)/ 3(13200 V) = 437.17 A total load flowing from L =0A Relays (7) (8) (5) (6) (3) (4) (1) (2) (1) (2) (3) (4) (5) (6) (7) (8) Selection of CTs It is assumed that the available CTs have primary turns in multiples of 100 up to 600.2 kV B 8 L 3Ω 4Ω 7 7 MVA 4 K 5 6 3 MVA Figure 7.138 Protection of electricity distribution networks J 2 5 MVA 2Ω 3 2Ω 150 MVA sc I 1 A 13.08 A Imax 2 = 656. the CT ratios selected are: Imax 1 = 656.90 A CT ratio = 800/5 CT ratio = 800/5 CT ratio = 500/5 CT ratio = 500/5 CT ratio = 400/5 . The CT ratios are calculated for the maximum load conditions first.69 A total load current flowing from J =0A With the ring open at B: √ total load current flowing through A = (15 MVA)/√3(13200 V) = 656. Calculation of the maximum load currents With the ring open at A: √ total load current flowing through B = (7 + 3 + 5 MVA)/ 3(13200 V) √ = (15 MVA)/ 3(13200 V) = 656. Thus.

16 + 1.16 Ω 1.79 × (9 ÷ 11) = 2226.64) Apportioning the current flows in the inverse proportion of the circuit impedances: IscJ (on the right hand side) = 2721.1 A IscJ = √ 3(1. IscJ = √ 3(1.8 A Zsource = (13200 V)2 /(150 MVA) = 1.16 Ω I 9Ω 1.64 Ω 13.2) .Directional overcurrent relays 139 Imax 6 = 349.2 kV √3 Figure 7.16 + 9) 1.05 × Isc is well below the number of primary turns for each one.87 A IscJ (on the left hand side) = 2721.79 A.92 A With breaker A open: 13200 = 750.08 A Imax 8 = 656. it is necessary to check whether or not saturation is present at the maximum fault level at each breaker. Calculation of fault currents Busbar I √ Isc = (150 MVA)/ 3(13200 V) = 6560.2 kV √3 2Ω 13.79 × (2 ÷ 11) = 494.90 A Imax 7 = 656.16 Busbar J The equivalent circuit for a fault at J is shown in Figure 7. With the most critical fault values. with the ring closed.13: 13200 = 2721.13 Equivalent circuit for a fault at J (Example 7. it can be shown that no CT is saturated since the value of 0. using the above CT ratios. as calculated in the following section.08 A CT ratio = 400/5 CT ratio = 800/5 CT ratio = 800/5 In order to confirm this selection.

16 + 3) .74 A With breaker A open: 13200 = 933.94 A IscK = √ 3(1.16 + 4) Busbar L The equivalent circuit for a fault at L is shown in Figure 7.74 × (3 ÷ 11) = 622.16 + 7) With breaker B open: 13200 = 1476.99 A IscK (on the left hand side) = 2059.74 × (4 ÷ 11) = 1310. with the ring closed.14: 13200 = 2059.45 A IscL (on the left hand side) = 2281. IscL = √ 3(1.2 kV √3 7Ω 1.2) With breaker B open: IscJ = √ 13200 3(1.16 + 2. with the ring closed.94 A IscK = √ 3(1.16 + 2) = 2411.74 × (8 ÷ 11) = 1659.14 Equivalent circuit for a fault at K (Example 7.29 A With breaker A open: 13200 = 1831.97 A IscL = √ 3(1.54 Ω 4Ω 13.74 × (4 ÷ 11) = 748.15: 13200 = 2281.16 + 2.140 Protection of electricity distribution networks 1. IscK = √ 3(1.54) IscK (on the right hand side) = 2059.2 kV √3 Figure 7.18) IscL (on the right hand side) = 2281.16 Ω 2.16 Ω I 13.74 A.74 A.71 A Busbar K The equivalent circuit for a fault at K is shown in Figure 7.

17 A Relay 1: 1. multiplied by .16 Ω 2.15 Equivalent circuit for a fault at L (Example 7.2 kV √3 Figure 7.25 × 1831.12 A Relay 4: 1.96 A 7.62 A Relay 6: 1.1 Pick-up setting The pick-up setting of a directional overcurrent relay is calculated by considering the maximum load transfer that can be seen by the relay in any direction.16 Ω 3Ω 1.6.2 kV √3 8Ω 13.94 A = 1846.99 A = 1039.25 × 933.08 A = 984.94 A = 1167.25 × 831.08 A = 984.97 A = 2289.Directional overcurrent relays 141 1.50 × 656.71 A = 3014.10 A = 937.64 A Anticlockwise direction: Relay 2: 1.99 A Relay 3: 1.6 Setting of time-delay directional overcurrent units As in the case of bi-directional overcurrent relays.25 × 750.42 A Relay 8: 1.25 × 1476.12 A Relay 5: 1. the time-delay units of directional overcurrent relays in a ring are set by selecting suitable values of the pick-up current and time dial settings.2) With breaker B open: IscL = √ 13200 3(1.16 + 8) = 831.25 × 2411. The procedure for each one is indicated in the following sections.18 Ω I 13.99 A Setting of instantaneous unit Clockwise direction: Relay 7: 1. 7.50 × 656.

Next. 3.2. normally starting from the relays associated with the main source busbar. especially under heavy transfer conditions. for this condition. if co-ordination is not achieved then the time dial values should be increased. For this case it is important to take . consider a fault at the opposite end of the line with the ring open and.3.3. the operating time of the relay on the breaker at the opposite end of the line should be calculated as well as the times for its back-up relays. then the time dial values of the relays at the adjacent substations should be modified. it has to be emphasised that both methods guarantee proper co-ordination although the first one can produce slightly higher time dial values and is more used in simple systems or when the locations of the co-ordination curves are not critical. 2. As in the previous case. the time dial value is adjusted so that. requires an iterative process as detailed below: 1. As in the case of bi-directional relays. both clockwise and anticlockwise. 7. The application of this method is illustrated in the instantaneous settings calculated for the relays in Example 7. Determine the initial time dial values of the relays on the ring such that co-ordination is guaranteed with the relays associated with the lines and machines fed by the adjacent busbar in the direction of trip. For this condition a check should be made to ensure that there is adequate discrimination between the chosen relay and the back-up relays at adjacent substations. taking account of the contact travel of the timer units. with the ring closed. whereas the second takes account of contact travel and is more rigorous since it requires fault calculations for various ring topologies.142 Protection of electricity distribution networks the overload factor referred to in Section 5. In addition. However. Time dial setting considering contact travel The time dial setting of directional relays. Calculate the time required for the first relay to operate for a fault at its associated breaker terminals.2 Time dial setting The time dial setting can be defined by means of two procedures. taking the instantaneous current setting given to the relay downstream. The load transfer in both directions is taken into account to avoid the possibility of relay mal-operation if the directional unit is incorrectly activated by the wrong polarisation. Any relay can be chosen as the first. calculate the operating time of the relay closest to the fault and check that there is adequate discrimination between it and the back-up relays at the adjacent substations. although it is usual to take one of the relays associated with equipment connected to the main source busbar. its operating time is above that of the downstream relay by the required discrimination time margin.6. Time dial setting by direct method The setting of time dial units by the direct method is based on the fault values used to set the instantaneous units. If not. The first one is based on instantaneous setting values. This procedure should be carried out for all the relays on the ring.

After setting the pick-up and instantaneous current values. for a fault at the various busbars with the ring closed. The short-circuit current value for a fault at the breaker terminals associated with relay R11 is calculated. 3. with the ring open.e. with a 34. 6. i. i. From Figure 7. The time dial values of the relays are initially set in such a way that co-ordination is guaranteed with those relays associated with lines or machines supplied from the three busbars of the ring. 5.Directional overcurrent relays 143 into account the contact travel during the fault before the ring is opened by the operation of the first relay.17a. the steps to set the time dial values by considering the travel contact are as indicated below: 1. The procedure is completed when no further time dial setting changes are required. To illustrate the above procedure. since the setting is carried out for the most severe conditions. The operating times of relays R13 and R14 should be checked to ensure that there is adequate co-ordination with relay R21 . 4. this value is 2471 A.e.16. From Figure 7. and with the ring open. their time dial settings should be increased. consider the system shown in Figure 7. the three-phase short-circuit level is 5157 A. It should be noted that calculating the time dial setting based on contact travel guarantees proper co-ordination of the relays on a ring. by considering a fault at its associated breaker terminals with the ring closed. the following expressions should be used: trelay next to fault = tadjacent relay with ring closed + trelay next to fault with ring open 1 − and tback-up relay = tadjacent relay with ring closed + tback-up relay with ring open 1 − tremote relay with ring closed tback-up relay with ring closed tadjacent relay with ring closed trelay next to fault with ring closed tback-up relay ≥ trelay next to fault + t discrimination margin 4. The same procedure is repeated for each relay. A fault at the breaker terminals associated with relay R21 with the ring closed is considered first. and then for a fault at the opposite end of the line with the ring open. The operating times of relay R11 and its back-up R22 are calculated to ensure that the specified fault will be cleared with the ring closed.17b. The operating times of relay R11 and its back-up relay R22 are calculated for this condition with the following expressions: trelay R11 next to fault = tR21 adjacent relay with ring closed + tR11 relay next to fault with ring open 1 − tR21 adjacent relay with ring closed tR11 relay next to fault with ring closed .5 kV ring connecting three busbars. To do this. If not. 2.

7.5/11.1 summarises the steps as a guideline to completing this co-ordination exercise.67 kV 35 MVA YN ynd Zps:9. . the time dial setting of relay R22 should be checked to confirm that it satisfies the following expression: tR22 ≥ tR11 + tdiscrim. Table 7.5 kV 800/5 R23 600/5 5.5 kV Figure 7. Finally. t = operating time of the relay considering the new topology and taking account of the contact travel.62% 1400/5 R14 B R21 7 km 34.margin This procedure is then repeated for the rest of the relays of the ring. t = operating time of the relay with the new topology after the first relay operates.16 Ring system to illustrate overcurrent directional relay setting procedure and tR22 back-up relay = tR21 remote relay with ring closed + tR22 back-up relay with ring open 1 − tR21 remote relay with ring closed tR22 back-up relay with ring closed where: t = operating time of the relay for the initial fault with the ring closed.144 Protection of electricity distribution networks A R24 115 kV 400/5 110/34.5 kV 800/5 34.7 km 4 km R11 C 800/5 R12 800/5 R22 D 600/5 R13 34.

62% R14 B 34.5 kV R21 115 kV R24 110/34.5/11.62% R23 R14 B 2471 A R11 C R12 34.5 kV R22 D R13 34.5 kV R22 D R13 34.67 kV 35 MVA YN ynd Zps:9.67 kV 35 MVA YN ynd Zps:9.Directional overcurrent relays 145 (a) A 115 kV R24 110/34. (b) fault associated with R11 with ring open at R21 .5 kV (b) A Indicates the location of the opening of the line 34.17 Short-circuit currents for faults at breaker: (a) fault associated with R21 with ring closed.5 kV Figure 7.5 kV R21 5157 A R23 R11 C R12 34.5/11.

tR 22 tR21 tR 23 tR12 tR . tR . 11 tR14 . 560 A fault at B = 2. 23 tR23 . tR . 13 tR23 tR 21 tR14 . tR . tR . 14 tR12 . 22 tR22 . 22 tR13 . tR . 14 tR22 tR . tR .16 Topology of the ring closed open closed open closed open closed open closed open closed open Calculation Co-ordination check tR13 . tR 12 Fault at B21 B11 B12 B22 B13 B23 B11 B21 B22 B12 B23 B13 tR21 . 12 tR23 .7 Exercises Exercise 7.1 The following short-circuit data refer to the ring system in Figure 7. tR . 23 tR11 . 21 tR13 . 235 A (contribution from 1) plus 496 A from 2 622 A (from 1) + 1. 21 tR22 .18: With the ring closed: fault at A = 6. 12 tR22 tR 11 tR23 tR 22 tR14 . 658 A (from 2) . 731 A fault at C = 2. 280 A 2. 14 tR21 . tR . 11 tR12 . tR . tR . tR .18 Network for Exercise 7. 13 tR11 .16 Ω A B 3 2Ω 4 6Ω 1Ω 5 1 2 6 C Figure 7.146 Protection of electricity distribution networks Table 7. tR . tR . tR .1 7. 13 tR21 tR 12 tR12 tR 13 tR14 tR11 tR 23 tR13 tR tR tR 11 21 14 tR11 150 MVASC ZSOURCE = 1.1 Summary of the procedures for the time dial settings for ring system in Figure 7.

and assume that each line is already open at the time the fault occurs. F2 .5 kV ring system are shown in Figure 7. F3 and F5 correspond to faults with the ring open at the ends indicated by the symbol • • . Exercise 7.5 kV F6 = 4440 A F5 = 2880 A C F6 F1 Indicates the location of the opening of the line F4 = 2160 A F1 = 1690 A F4 B F3 = 2300 A F2 A Figure 7. calculate the setting for the instantaneous element of relay 4 at substation A. B and C. The tap of each relay was calculated beforehand and a value of 5 was selected for the three relays. F1 .Directional overcurrent relays 147 115 kV 34.2 The short-circuit levels for different three-phase fault conditions on a 34. The fault currents F1 . F4 and F6 correspond to faults with the ring closed With the ring open: fault at B = 750 A = 2411 A fault at C = 1832 A = 832 A (with 1 open) (with 2 open) (with 1 open) (with 2 open) If the instantaneous element 6 at substation C is set to 424 A and protects up to the middle of the line A-C with the ring closed.2. F2 = 1350 A F5 F3 .19 Network for Exercise 7. F3 and F5 take account of those lines that are open at the points indicated for each fault.19. Determine the relay time dial and instantaneous current settings for the overcurrent relays associated with breakers A.

The relay has the following characteristics: Pick-up: 1 to 12 A in steps of 1 A Time dial: as in the diagram Instantaneous settings: 6 to 80 A in steps of 1 A . shown in Figure 5.15.148 Protection of electricity distribution networks Use the characteristic curves of the inverse-time relay.

under normal load conditions or when there is a fault outside the protection K1I1 K2I2 I1 Protected element C T1 I2 C T2 Figure 8.Chapter 8 Differential protection 8. provided that both sets of CTs have appropriate ratios and connections then. Almost any type of relay can function as differential protection – it is not so much the construction of the relay that is important but rather its method of connection in the circuit. the difference lies in the fact that the operating signal is derived from a voltage across a shunt resistance. The majority of the applications of differential relays are of the current-differential type. Although the currents I1 and I2 may be different. operating on the same principle as the current relays.1. but they can also be of the voltage-differential type.1 Differential protection – current balance . The secondaries of the current transformers (CTs) are interconnected and the coil of an overcurrent relay is connected across these.1 General Differential protection functions when the vector difference of two or more similar electrical magnitudes exceeds a predetermined value. A simple example of a differential arrangement is shown in Figure 8.

The relay restraining force increases with the magnitude of (I1 + I2 ). the secondary currents will circulate between the two CTs and will not flow through the overcurrent relay. If the relays do not have these variable tappings. Since this type of relay is most common in differential relaying schemes. CTs. the restraining torque is increased in the presence of through-fault currents producing a more stable operating characteristic and preventing relay mal-operation. which is the same as if (I1 + I2 )/2 flowed through all of the restraining coil. One arrangement that is extensively used is the differential relay with a variablepercentage characteristic. will not have identical secondary currents for the same primary currents because of slight differences in their magnetising characteristics. although produced to the same specification.2 Differential relay with variable-percentage characteristic .2. alternatively described as being percentage biased. The operating characteristic of the relay with this form of restraint is shown in Figure 8. I1 C T1 Protected element I1 NR NR1 NR2 Nop I2 C T2 I2 NR Restraint coil Nop Operating coil Figure 8. thus preventing unnecessary tripping due to any CT unbalance errors. the tappings can be set to balance out any currents due to differences in the CTs. In addition. In all cases the current in the differential relay would be proportional to the vector difference between the currents that enter and leave the protected element. The current in the operating coil is proportional to (I1 − I2 ). if a fault occurs in the section between the two CTs the fault current would flow towards the short-circuit from both sides and the sum of the secondary currents would flow through the differential relay.150 Protection of electricity distribution networks zone of the element. which has an additional unit. then the currents leaving the CTs should match closely in order to avoid mal-operation of the relays. then the total ampere-turns are equal to I1 (N /2) plus I2 (N/2). the restraining coil. If N is equal to the number of turns of the restraining coil with the operating coil connected to the mid-point of the restraining coil. In relays that have variable tappings in the restraint coil circuits. if the current through the differential relay exceeds the setting value then the relay will operate. in addition to the operating coil as shown in Figure 8. However.3. all references to differential relays in this chapter will be on the basis of using this type of relay.

Thus. There are also other types of differential relays which use directional or overvoltage elements in place of the overcurrent elements.4. as shown in Figure 8. these add arithmetically.3 Relay characteristic (variable-percentage type) Restraint coil Operating coil Element with 3 terminals Figure 8. The slope of the operating characteristic of each relay varies depending on the current distribution in the three restraining coils.4 Differential protection for element with three terminals Differential relays can be used for power system elements that have more than two terminals. Each of the three restraining coils has the same number of turns and each coil produces a restraint that is independent of the others.Differential protection 151 I1 – I2 Operation m= NR Nop = Iop Ires No operation (I1 + I2)/2 Figure 8. all types are extensions of the fundamental principles that have been described above. .

and the fact that equivalent ampere-turns are developed in the primary and secondary windings of the transformer.152 Protection of electricity distribution networks 8.2 Classification of differential protection Differential protection can be classified according to the type of element to be protected. The CTs on the primary and secondary sides of the transformer are connected in such a way that they form a circulating current system. I1 I2 87 IR1 Differential relay IR2 Figure 8.5 Transformer differential protection . Faults on the terminals or in the windings are within the transformer protection zone and should be cleared as quickly as possible in order to avoid internal stress and the danger of fire. The main faults in this group are core faults. lines and busbars. as illustrated in Figure 8.3 Transformer differential protection A differential system can protect a transformer effectively because of the inherent reliability of the relays. 8. generators and rotating machines. as follows: • • • transformers. However. An internal fault that does not constitute an immediate danger is designated an incipient fault and. This type of protection not only responds to phase-to-phase and phaseto-earth faults but also in some degree to faults between turns. phase-tophase faults between the windings of a three-phase transformer are less common. caused by the deterioration of the insulation between the laminations that make up the core. could degenerate into a major fault. which are highly efficient in operation.5. The principal cause of these faults is arcing inside the bushings and faults in the tap changer. Differential protection can also detect and clear insulation faults in the transformer windings. with the severity depending on the design of the transformer and the type of earthing. The majority of internal faults that occur in the windings are to earth (across to the core) or between turns. if not detected in time.

the CTs on the delta side of the transformer should be connected in star. . their nominal secondary values should be multiplied by 3 so that the currents flowing in the delta are balanced by the secondary currents of the CTs connected in star. it is necessary to have some method of distinguishing between the magnetising current and the fault current. the following factors should be taken into account: Transformation ratio The nominal currents in the primary and secondary sides of the transformer vary in inverse ratio to the corresponding voltages. For the same reason. the zero-sequence current that flows in the star side of the transformer will not induce current in the delta winding on the other side. Furthermore. The magnetising inrush produces a current flow into the primary winding that does not have any equivalent in the secondary winding.Differential protection 153 8. Using a differential relay with a suitable sensitivity to cope with the magnetising current.3. The zero-sequence current can therefore be eliminated from the star side by connecting the CTs in delta. For this reason it is necessary to include some form of biasing in the protection system together with some identifying markings of the higher current input terminals. and any differential protection system should be able to cope with this variation. Tap changer If the transformer has the benefit of a tap changer it is possible to vary its transformation ratio. Transformer connections When a transformer is connected in star/delta. This should be compensated for by using different transformation ratios for the CTs on the primary and secondary sides of the transformer. This shift can be offset by suitable secondary CT connections. Magnetisation inrush This phenomenon occurs when a transformer is energised. usually obtained by a unit that introduces a time delay to cover the period of the initial inrush peak. depending on the vector group. Since it is not practical to vary the CT transformation ratios. the secondary current has a phase shift of a multiple of 30◦ relative to the primary.1 Basic considerations In order to apply the principles of differential protection to three-phase transformers. These methods include: 1. or when the primary voltage returns to its normal value after the clearance of an external fault. the differential protection should have a suitable tolerance range in order to be able to modify the sensitivity of its response of operation. When CTs are connected in √ delta. Since the differential relay sees the magnetising current as an internal fault. The net effect is thus similar to the situation when there is an internal fault on the transformer.

3. and the connection of the differential relays as applied to transformer protection schemes. The following examples show the connections of the CTs. in conjunction with a differential unit. Determine the transformation ratio and connections of the CTs required in order to set the differential relays. In general. If there are more than two windings. then compensation transformers can be used to offset the phase shift across the transformer. Inhibiting the differential relay during the energising of the transformer.5/69 kV Yd1 152 252 3 87 Figure 8.6. CTs with ratios in steps of 50/5 up to 250/5. Use relays with a variable-percentage 30 MVA 11. Example 8. If there are more than two windings it is necessary to consider all combinations. taking two windings at a time and the nominal power of the primary winding. all combinations should be considered.1 Consider a 30 MVA. If the available CT ratios do not enable adequate compensation to be made for any variation in secondary current from CTs. This arrangement compensates for the phase shift across the transformer and blocks the zero-sequence current in the event of external faults to earth. taking two windings at a time.2 Selection and connection of CTs The following factors should be taken into account when considering the application of differential protection systems: 1.6 Single-line diagram for Example 8. Yd1 transformer as shown in the single-line diagram in Figure 8.154 Protection of electricity distribution networks 2. and those on the delta side should be connected in star. 11. and in steps of 100/5 thereafter.5/69 kV. the calculation of their transformation ratios.1 . should be used. 2. 8.3. or a supervisory unit. 3. The CT ratios should be selected in order to produce the maximum possible balance between the secondary currents of both sides of the transformer under maximum load conditions. the CTs on the star side of a star/delta transformer should be connected in delta. The relays should be connected to accept the load current entering one side of the transformer and leaving by the other side. Using a harmonic-restraint unit.

For a throughput of 30 MVA the load currents are 30 MVA Iload (69 kV) = √ = 251. and 5.0-7.Differential protection 155 characteristic.0. the CT ratio at 11.0 A.5 kV is selected as close as possible to the maximum load current and. 5. 5.0 A 3 × 69 kV 30 MVA = 1506. The currents in the windings and in the lines are drawn and show that the restraint currents on the star and delta sides of the relay are in phase.5.5 kV In order to increase the sensitivity.0-5.0-6.0.0-5. 5.7 Three-phase connection diagram for Example 8. 5.5 kV) of IA 1500/5 Ia IB B Ib IC C Ic IC IC 30 MVA Yd1 Ic – Ib IB IB IC – IB Ib – Ia 115 kV IA 69 kV IA – IC 250/5 Ia – Ic IB – IA A IA Ia – Ic Ib – Ia Ic – Ib R R R OP OP OP R R R Ia – Ic Ib – Ia Ic – Ib Figure 8.1 . 5.0-8. The available current taps are: 5. Solution Figure 8.0-6.3.0-9.5 kV) = √ 3 × 11.7 shows the complete schematic of the three-phase connections.0. 5.6. therefore.0-10. a CT ratio (11.0.13 A Iload (11.

indicating how the currents circulate in order to check the correct functioning of the differential relay.9. In this case a value approximating to the nominal current is preferred and the ratio 250/5 is selected. The completed schematic is given in Figure 8. When calculating the√ other CT ratio. This would suggest using a CT ratio of 150/5. The load currents are √ Iload (13. a balance of currents has to be achieved. Finally. With the two ratios selected in this way.8. taking into account the fact that the differential relay has several taps.69 A Therefore the tap range 5. Use differential relays of the same type as for Example 8.5 kV = 1506.02 A √ Irelay at 11. the currents in the windings of the relay for nominal conditions are Irelay at 69 kV = 251 × (5/250) = 5. 1506.5 MVA 115/13. Once this is obtained.8 Single-line diagram for Example 8.13 × (5/1500) × 3 = 8. i.75 A .156 Protection of electricity distribution networks 12.e.2 1500/5 is chosen.1.0-9.2 For the transformer shown in Figure 8. it is not necessary to have exactly the same current values at its terminals and therefore another CT ratio can be used.0 A should be selected.5 MVA)/( 3 × 115 kV) = 62. Example 8. the complete schematic diagram of the connections can be drawn.5 MVA)/( 3 × 13. Solution Knowing the vector group of the transformer it is possible to determine the connections of the windings. Draw the complete schematic for the three-phase connections and identify the currents in each of the elements.13 × (5/1500) × 3 = 251 × (5/X)A ⇒ X = 144.2 kV) = (12.2 kV Dy1 100/5 600/5 87 Compensation transformer Figure 8.7 A √ Iload (115 kV) = (12. this ratio is checked to see if it is compatible with the taps that are available on the relay.2 kV) = 546. However. determine the transformation ratios and the connections required for the compensation transformers.

from which a = 2. Example 8. let us assume that this value of a is not included on commercially √ available relays but that a value of a = 3 is.0-7.75 × (5/100) = 3. Produce the complete three-phase connection schematic and identify the currents in each of the elements.2 In order to select the ratio of the compensation transformers it is necessary to obtain a correct balance of currents. .13 5 from which X = 7.75 A × (5/100) = 546. given the power rating and voltage ratio of the power transformer and the CT ratios. determine the transformation ratios and the connections of the compensation transformers required in order to set the differential relays. which are not provided with taps. this being a very typical value provided on this type of relay.26 A.2 kV) = 546.2 kV IA IB IC 600/5 Ia Ib Ic Ia – Ib Ib – Ic Ic – Ia R R R OP OP OP R R R Ia – I b Ib – Ic Ic – Ia Ia Ib Ic a:1 Ia Ib Ic Dy1 Figure 8.51. 62.3 A.55 X = . 3.7 A × (5/600) × (1/a) × 3.13 A √ √ Irelay (13.10. However. i.9 Three-phase connection diagram for Example 8.3 In the system in Figure 8. taking into account the delta connection on one √ of the sides.55 A Selection of the tap setting: 4.Differential protection 157 IA – IB IB – IC IC – IA 100/5 Ia – I b Ib – Ic Ic – Ia 115 kV IA IB IC 13. so that for these conditions the tap to be selected is 5. On this basis the currents feeding into the relay are: Irelay (115 kV) = 62.7 × (5/600) × (1/ 3) × 3 = 4.e.

578 = 1.11. When determining the ratios of the compensation transformers.3 Solution The three-phase diagram of connections is given in Figure 8. 2 13.5 kV side should be equal to 4. Considering the currents on the primary and secondary sides.10 Single-line diagram for Example 8.18 A.18 A. The current in the relay associated with the 34. Irelay (115 kV) = √ [25 MVA/( 3 × 115 kV)] × (5/150) = 125. trfr. provided that the delta side has the higher number of turns. If the differential relays are connected on the Y side. and a Yy0 connection for compensation transformer 1 is therefore the most appropriate arrangement.51 × (5/150) = 4. thus obtaining a correct balance of magnitudes. This connection is equivalent to taking the primary and the tertiary and treating them as a transformer with two windings so . the calculations involving the primary and secondary windings should be carried out on the basis of the main transformer having only two windings with no current circulating in the tertiary.158 Protection of electricity distribution networks 150/5 115 kV 25 MVA 34.18 A = √ 500 a2 3 × 34. The calculations involving the primary and tertiary windings should be treated in a similar way.5 kV 25 MVA 500/5 Comp. trfr.2 kV side is calculated assuming that the power of the tertiary winding is equal to that of the primary winding. the connections of the windings of the power transformer can be obtained from the vector group.2 kV 5 MVA Primary–secondary Yd5 Primary–tertiary Yy0 300/5 87 Comp. The turns ratio of compensator 2 is therefore 1/0. 1 Figure 8. If several earthing points are used circulating currents could be induced during external faults causing the relays to pick up inadvertently.73. There is no need for phase compensation between the primary and tertiary winding. The current in the relay for the 13. This method ensures that a correct selection is made which will cover any fault or load current distribution. It should be noted that the common point of the operating coils and the neutral points of the compensation transformers and CTs must be connected to only one earthing point in order to avoid maloperation during external faults.5 kV) = 4. then compensation transformer number 2 should be connected in Yd5 in order to compensate for the phase difference between the primary and secondary currents. so that 5 1 25 MVA 1 × ×√ × Irelay (34.5 kV 3 from which a2 = 0.578.

. involving X per cent turns.18 A = (25 MVA/( 3 × 13. especially in the case of single-phase faults. in which the star winding has been earthed via a resistor. 2 Yd5 a2:1 Ia IB – IA IC – IB IA – IC 13.11 Three-phase connection diagram for Example 8.12a.36.3 Percentage of winding protected by the differential relay during an earth fault Although differential protection is very reliable for protecting power transformers.2 kV) = 4.2 kV)) × (5/300) × (1/a1 ). With this setting.3 √ that Irelay (13. for any load distribution.Differential protection 159 IA 150/5 Ia IB Ib IC Ic IC Yd5 Ia Ib Ic R OP R OP R OP R R R R R R Ic Comp.trfr.5 kV IA IB IA IB IC Ib – Ia Ic – Ib Ia – Ic Comp.1 Yy0 a1:1 Ic Ib Ia 115 kV 34. from which a1 = 4. Assume an internal earth fault occurs at point F at a distance X from the neutral point. Consider the case of a delta/star transformer as shown in Figure 8.3. the windings are not always fully protected.2 kV IA IB IC Yy0 Ia Ib Ic IA IB IC Ib Ia Ib Ic Ia Ib Ic Figure 8. it can be shown that. the primary restraint current is equal to the sum of the secondary and tertiary restraint currents. 8.trfr.

(with full line-to-neutral voltage applied between phase and earth).6 0.3 0.5 0. the . with a fault on the star side: (a) connection diagram.1 0 20 40 Winding not protected 60 80 100 Relay pick up IS IL Distance X of fault from neutral (%) Figure 8.9 0. by direct ratio. Therefore. star winding earthed via a resistor.8 0. The secondary current for a fault at F is produced by X per cent of the line-toneutral voltage. In addition.4 0.12 Delta/star transformer.160 Protection of electricity distribution networks (a) A Source B X C IL Ip IL = IP Is F Np Ns Rf (b) 1.2 0.0 0.7 Fault current I (per-unit) 0. (b) fault current values for primary and secondary and that the resistor has been set so that nominal current Inom will flow for a fault on the terminals. the current will be XI nom . The numbers of primary and secondary turns are Np and Ns respectively.

Differential protection 161 number of turns involved in the fault is XN s . The distribution of current in the delta side for an earth fault on the star side results in a line current IL equal to the phase current. Therefore IL = XInom × (XNs /Np ) = X2 Inom (Ns /Np ) Under normal conditions, the line current in the delta side, IL , is √ IL = 3Inom × (Ns /Np ) (8.1)

(8.2)

If the differential relay is set to operate for 20 per cent of the nominal line current then, for operation of the relay, the following should apply: IL ≥ 0.2 × IL i.e. √ X2 Inom (Ns /Np ) ≥ 0.2 × 3 × Inom × (Ns /Np ) √ X2 ≥ 0.2 3, i.e. X ≥ 59 per cent

Therefore, 59 per cent of the secondary winding will remain unprotected. It should be noted that to protect 80 per cent of the winding (X ≥ 0.2) would require an effective relay setting of 2.3 per cent of the nominal primary current. This level of setting can be very difficult to achieve with certain types of differential relays. Figure 8.12b illustrates typical primary and secondary currents for delta/star transformers, where the secondary star winding is earthed via a resistor, and also the effect of the location of a fault along the star winding on the pick up of the differential relays.

8.3.4

Determination of the slope

The setting of the slope of differential relays is carried out with the aim of ensuring that there will be no mal-operations because of differences in the currents in the restraint windings due to the transformation ratios of the CTs and the operation of the tap changer under load conditions. In order to determine the slope, the restraint and operating torques are calculated on the basis of the currents and the number of turns in the respective coils as set out below: Tres = I1 NR1 + I2 NR2 Top = |I1 − I2 |Nop where: I1 , I2 = currents in the secondaries of the CTs; NR1 , NR2 = number of turns on the restraint coils (see Figure 8.2). In order for the relay to operate, Top > Tres , i.e. |I1 − I2 |Nop > I1 NR1 + I2 NR2 . If NR1 = NR2 = NR /2, this then gives Tres = (I1 + I2 )NR /2. For relay operation, the slope Nres |I1 − I2 | ≥ =m 0.5|I1 + I2 | Nop A typical operating curve is shown in Figure 8.3.

162 Protection of electricity distribution networks

8.3.5

Distribution of fault current in power transformers

When considering the operation of differential protection it is very important to take into account the distribution of fault current in all the windings in order to ensure that the settings that have been selected have a suitable sensitivity. This is particularly critical for single-phase faults in transformers that are earthed via an impedance. The following example illustrates the procedure. Example 8.4 A 115/13.2 kV Dy1 transformer rated at 25 MVA has differential protection as indicated in Figure 8.13. The transformer is connected to a radial system, with the source on the 115 kV side. The minimum operating current of the relays is 1 A. The transformer 13.2 kV winding is earthed via a resistor which is set so that the current for a single-phase fault on its secondary terminals is equal to the nominal load current. Draw the complete three-phase diagram and indicate on it the current values in all the elements for: (i) (ii) Full load conditions. When a fault occurs at the middle of the winding on phase C, on the 13.2 kV side, assuming that the transformer is not loaded. For both cases indicate if there is any relay operation. Solution Full load conditions The full load conditions for the maximum load of the transformer are as follows: 25 × 106 VA = 1093.47 A Inom(13.2 kV) = √ 3 × 13.2 × 103 V and 25 × 106 VA = 125.51 A Inom(115 kV) = √ 3 × 115 × 103 V It should be noted that, based on the primary currents given, the phase rotation A-B-C is negative, i.e. clockwise. Therefore, the respective currents at the secondary lead the primary currents by 30◦ in order to provide the required phase shifting for the transformer vector group (Dy1). Figure 8.13 shows the current values through the HV and LV connections, and it is clear that balanced currents are presented to the differential relays, which therefore do not pick up, as expected. Fault at the middle of 13.2 kV winding C Since the transformer is earthed through a resistor that limits the current for faults at the transformer 13.2 kV bushings to the rating of the winding, and since the fault is at the middle of the winding, the fault current is then equal to half the rated value as follows: Ifault = (Inom(13.2 kV) )/2 = 1093.47/2 = 546.74 A

Supply 0° Ω 2250/5 2.43 30° 72.46 30° 120° Ω 72.46 150° 240° Ω 72.46 270° 2.43 270° R 1093.47 270° Ω 2.43 150° 1093.47 150° Ω 25 MVA Dy1 1093.47 30° Ω

150/5

125.51

Load

A

125.51

B

125.51

C

4.18 0° Ω R R OP R OP 0.03 R 0.03

4.21 0° Ω 4.21 120° Ω R R OP 0.03 0.03 ≤ 1A => Does not operate 4.21 240° Ω

4.18 120° Ω

4.18 240° Ω

Differential protection 163

Figure 8.13

Three-phase connection diagram for Example 8.4

164 Protection of electricity distribution networks The primary current within the delta winding is Iprim = Ifault × Then Iprim = Ifault × V2 13.2 = 18.12 A = 546.47 × √ √ 2 × 3 × V1 2 × 3 × 115 (N2 /2) N2 V2 , since =√ N1 N1 3 × V1

Figure 8.14 shows the current values through the HV and LV connections, from which it can be seen that, for this case also, the differential relays do not operate since the current through their operating coils is only 0.6 A, which is less than the 1 A required for relay operation.

8.4

Differential protection for generators and rotating machines

Differential protection for generators and other rotating machines is similar to that for transformers in several ways. Internal generator winding faults include phase-tophase short-circuits, short-circuited turns, open circuits and faults to earth, and should be disconnected by opening the circuit as quickly as possible. In order to obtain the most effective form of differential protection the neutral of the generator should be well earthed, either solidly or via a resistor or a reactor. The differential protection should satisfy the following requirements: 1. It should be sensitive enough to detect damage in the winding of the generator stator, and yet not operate for faults outside the machine. 2. It should operate quickly in such a way that the generator is disconnected before any serious damage can result. 3. It should be designed so that the main breaker is opened, as well as the neutral breaker and the field circuit breaker. The arrangements of the CTs and the differential relays for a machine connected in star can be seen in Figure 8.15, and in Figure 8.16 for a delta connection. If the neutral connection is made inside the generator and the neutral taken outside as shown in Figure 8.17, the differential protection provided will only cover earth faults.

8.5

Line differential protection

The form of differential protection using only one set of relays as illustrated in Figure 8.2 is not suitable for long overhead lines since the ends of a line are too far apart to be able to interconnect the CT secondaries satisfactorily. It is therefore necessary to install a set of relays at each end of the circuit and interconnect them by some suitable communication link. Pilot protection is an adaptation of the principles of differential protection that can be used on such lines, the term pilot indicating that there is an interconnecting channel between the ends of the lines through which information

Supply 25 MVA Dy1 R

150/5

Load 2250/5

A

B 18.12 A

18.12 A

S

0.6 A

18.12 A 548.7 R

C

T

0.6 A

R R OP R OP 0.6 R R

0.6

0.6

R OP 0.6

Differential protection 165

Figure 8.14

Conditions for a fault at the middle of the winding on phase C on the 13.2 kV side (Example 8.4)

166 Protection of electricity distribution networks
a b c

OP 52 OP Generator windings R R R R

R

R Differential relay

OP

Figure 8.15

Differential protection for a generator connected in star, with four terminals

can be transmitted. There are three different types of interconnecting channels – pilot wires, current-carrying wires and centimetric-wave systems. A pilot wire arrangement generally consists of two telephone-line type wires, either overhead or cable. In a current-carrying pilot system, high frequency low tension currents are transmitted along the length of a power conductor to a receiver at the other end; the earth or guard wire generally acts as the return conductor. Centimetric waves are transmitted by means of a very high frequency radio system, usually operating above 900 MHz. The principle of operation of pilot differential protection is similar to the differential systems for protecting generators and transformers, but the relays have different settings because the breakers at the ends of the line are more widely separated and a single relay should not be used to operate two tripping circuits. This method of protection is ideal from a theoretical point of view as both ends of the line should open instantaneously for faults wherever they occur on the line. In addition, the system should not operate for faults outside the section and is therefore inherently selective. Many of the operational difficulties with conventional schemes due to induced currents have been overcome by the use of fibre optics, which has greatly improved the reliability of this type of protection.

Differential protection 167
a b c

OP 52 OP OP R R R R

R R Differential relay

Generator windings

Figure 8.16

Differential protection for a generator connected in delta
a b c

R 52 Generator windings OP

R

R

Figure 8.17

Differential protection for a generator connected in star

There can be many circuits connected to the busbar. high impedance relays predominate because of their better performance.6. Although some busbar differential protection schemes still use multiple restraint features. a large proportion of busbar faults are the result of human error rather than faults on switchgear equipment. The majority of faults on a busbar involve one phase and earth and are due to many causes such as lightning and imperfections in the insulation of switchgear equipment. The secondaries of OP R R R R R OP R OP R OP R 52 52 52 52 Busbar Figure 8.18 Multiple restraint busbar differential protection . Under normal system conditions the power that enters a busbar is identical to the power that leaves. 8. which then results in the tripping of all the breakers associated with that busbar.1 Differential system with multiple restraint Differential relays with a variable-percentage characteristic should be used in a scheme involving multiple restraint. In busbar differential schemes that involve bushing type CTs. a fault inside the differential circuit unbalances the system and current thus flows in the operating coil of the relay. as shown in Figure 8.18.168 Protection of electricity distribution networks 8. However.6 Busbar differential protection Busbar differential protection is based on the same principles as transformer and generator differential protection. six to eight secondaries can usually be connected in parallel without difficulty. which necessarily implies the connection of a number of CT secondaries in parallel.

and the transformation ratios of the CTs. During internal faults the voltage across the relay terminals is higher and results in the operation of the instantaneous overvoltage unit which sends a tripping signal to the appropriate breakers. during an external fault. Consequently. and thus avoids mal-operation for external fault or overload conditions when the secondary currents of all the CTs are not the same because of differences in the magnetisation characteristics. the voltage across the terminals of the relay is relatively low and does not initiate any relay operation. plus a safety margin. Busbar 87 a b c High impedance differential relay Figure 8. In order for the scheme to be effective.2 High impedance differential system The high impedance arrangement tends to force any incorrect differential current to flow through the CTs instead of through the operating coil of the relay.Differential protection 169 the CTs on the feeders on the outgoing side of the busbar are connected in parallel and across the differential relay. taking into account the maximum primary fault current. The relay basically consists of an instantaneous overvoltage unit which is set by calculating the maximum voltage at the relay terminals for an external fault.6.19 High impedance differential protection scheme with CTs in parallel . the resistance of the secondary windings and the wiring. together with the secondaries of CTs of the circuits on the incoming side of the busbar which also are connected in parallel. 8. the resistance of the CT secondary wiring should be as low as possible. with the connections made in this way. the busbar is only protected from earth faults. connected across the terminals of the CT secondaries which are connected in parallel to a set of CTs per circuit.19. as shown in Figure 8. However. Connecting CTs in parallel This arrangement requires only one high impedance relay.

8.1 for the transformer in Figure 8.20 High impedance differential protection arrangement using a common bus for each phase Using common buses for each phase In this scheme balanced groups of CTs are formed in each one of the phases. using a common bus to feed three elements of a high impedance differential relay as illustrated in Figure 8. Exercise 8.0-8. so that the busbar is protected against both phase and earth faults.9-3. and have been selected in such a way that there is no saturation.2 Carry out the same calculations as for Exercise 8.2-4.20.8-4. using a differential relay with the same taps for each restraint circuit.2-3. determine the taps for each restraint coil.21 is protected by a differential relay provided with the following taps in each restraint circuit: 2.5-3. .7 Exercises Exercise 8.6-5.22.1 The transformer shown in Figure 8.7 If the current transformers have the ratios indicated in the diagram.170 Protection of electricity distribution networks Busbar A B C N 87A 87B 87C Figure 8.

23 Diagram for Exercise 8.5/13.23.5 kV side of the transformer have a ratio of 200/5.21 Diagram for Exercise 8.2 kV side a ratio of 900/5. The CTs on the 34. Dy1 transformer whose connections are given in Figure 8.1 161 kV 30/40 MVA 69 kV 30/40 MVA 200/5 400/5 500/5 12.5/13.5/10 MVA Figure 8.5 kA 200/5 Dy1 10 MVA 34.3 Exercise 8.4 kV 7. 34.4/69 kV 16/20 MVA Figure 8.2 kV. and those on the 13.Differential protection 171 1000/5 200/5 12.2 kV Figure 8.2 900/5 IB 4500 IC 4500 R OP R OP R R OP R R 4.22 Diagram for Exercise 8. .3 Consider the 10 MVA.

determine the transformation ratios and connections of the compensation transformers in order to connect the differential relays. 69/12.5 kA. determine if these result in the differential relays operating for a fault between phases b and c in the secondary windings of the transformer which are carrying a fault current of 4. Note: The differential relays operate for a current equal to 20 per cent of nominal current. taking into account that IA . which is 5 A. they operate for currents above 1 A.4 kV. The primary and secondary windings of the compensators can only be connected in star or delta. Draw the three-phase schematic diagram of the connections and identify the currents in each of the elements of the system. Yd7 transformer shown in Figure 8.24 Diagram for Exercise 8.172 Protection of electricity distribution networks IA Yd7 69/12. which are not provided with taps to vary the settings.4 kV 1200/5 600/5 IB IC OP R R R R OP R OP R a:1 Figure 8. Exercise 8. Therefore. .24. IB and IC feed into the transformer primary. If the answer is negative. Check if the operation of the differential relays is appropriate. indicate what correction should be applied.4 For the 16/20 MVA. In addition.4 Calculate the magnitude and direction of the currents in the primary circuits and the CT secondaries.

and for these reasons is extensively used on power system networks.1.1 General It is essential that any faults on a power system circuit are cleared quickly. For the system shown in Figure 9. as seen by the relay. against the voltage at the relay location to determine the impedance down the line to the fault. this involves comparing the fault current. Distance protection meets the requirements of reliability and speed needed to protect these circuits. a relay located at A uses the A F1 B F2 A F1 B F2 Figure 9. otherwise they could result in the disconnection of customers. Distance protection is a nonunit type of protection and has the ability to discriminate between faults occurring in different parts of the system. Essentially.1 Faults occurring on different parts of a power system . depending on the impedance measured. loss of stability in the system and damage to equipment.Chapter 9 Distance protection 9.

9.2 Relay based on an amplitude comparator . the distance relay has a fixed reach. which leads to incorrect r.2. or when I ZR ≥ V . although the quantities compared would be different in each case.m. this condition can be obtained in the amplitude comparator that operates when the ampere-turns of the current circuit are greater than the ampere-turns of the voltage circuit. The value of the impedance Z for a fault at F1 would be ZAF1 .2 Types of distance relays Distance relays are classified depending on their characteristics in the R-X plane. The main advantage of using a distance relay is that its zone of protection depends on the impedance of the protected line that is a constant virtually independent of the magnitudes of the voltage and current. The most common type compares the magnitude or phase of the two incoming signals in order to obtain the operating characteristics.174 Protection of electricity distribution networks line current and the line voltage to evaluate Z = V /I . Zs R If + Vf Vf = If Zf – To trip Fault If ZL Relay Restraint coil Operating coil Figure 9. If ZR is the impedance setting of the distance relay. the number of incoming signals and the methods used to compare the incoming signals. As shown in Figure 9. in contrast to overcurrent units where the reach varies depending on system conditions.s. However. it is more convenient to compare two signals by their phase difference than by their amplitudes. which are straight or circular lines when drawn in the R-X plane. the use of amplitude comparators is limited. it is difficult to provide an amplitude comparator that functions correctly under fault conditions when the phase displacement between V and I tends to be 90◦ and transients are present. Thus. it should operate when ZR ≥ V /I . and (ZAB + ZBF2 ) for a fault at F2 . For these reasons. values of V and I which are required to evaluate I ZR ≥ V . Any type of characteristic obtainable with one type of comparator can also be obtained with the other.

4 is satisfied when Given that C −90◦ ≤ α − β ≤ +90◦ (9.4 is satisfied in the semiplane on the right. obtained for use with an amplitude comparator. The relationship between the signals is as follows: S0 = S1 + S2 Sr = S1 − S2 From eqns. as shown in Figure 9. 9.5) The above relationships demonstrate that two signals. it can be seen that condition 9. there exist two other signals S1 and S2 that can be compared by phase. Imaginary Operating zone C– 1 C C+ 1 –1 +1 Real Figure 9.3 Comparison of phases in a complex plane: C = S1 /S2 .2) Drawing C in the R-X plane.3 can be expressed as |C + 1| ≥ |C − 1| (9. points C = (S1 α)/(S2 β). S0 and Sr .Distance protection 175 The following analysis shows that for two signals. can be converted in order to be used by a phase angle comparator.1) (9.1 (S0 + Sr ) 2 (S0 − Sr ) S2 = 2 S1 = The comparison of the amplitudes is given by: |S0 | ≥ |Sr | |S1 + S2 | ≥ |S1 − S2 | Defining S1 /S2 = C. the relationship 9.4) (9. then the relationship 9. The signals to be compared are analysed in the following paragraphs to obtain the operating characteristics of the main types of distance relays.3. which are to be compared in magnitude. This semiplane is defined for all the in such a way that −90◦ ≤ ≤ +90◦ .3) (9.

the angle between S1 and S2 is 90◦ if |Z| = |ZR /K|. is given by −90◦ ≤ ≤ +90◦ .8 in the R-X plane.e. for all the points inside the circle. Therefore point C is the limit of the operating zone.7) . which then will not operate. The vector AB represents the impedance in front of the relay between its location at A and the end of the line AB. 9. the following signals should be assigned to S0 and Sr : S0 = I ZR Sr = KV (9. on the other hand. i. Note that the magnitudes of signals S1 and S2 have been changed when dividing by KI . while the vector AC represents the impedance of the line behind the site of the relay. The construction is shown in Figure 9.1 Impedance relay The impedance relay does not take into account the phase angle between the voltage and the current applied to the relay and. or when I ZR ≥ V . and the locus of point C for the different values of Z is a circle of radius ZR /K. If. it is required that the relay will operate when ZR ≥ V /I . Being nondirectional. the situation shown in Figure 9. along the vector AC. Thus. If Z < ZR /K. In order for an impedance relay to work as a phase comparator. for this reason. (9. then is greater than 90◦ and Z is outside the operating zone of the relay. Drawing ZR /K and the eqns.8 give the origin of the rhomboid OABC which has diagonals of S1 and S2 .7) and for all faults behind the busbar. as in Figure 9. gives S1 = Z + ZR /K S2 = −Z + ZR /K where Z = V /I .2. It should be noted that drawing S1 and S2 in one or the other scale does not affect the phase relationship between the two signals. In this case is less than 90◦ and consequently the vector for Z is inside the relay operating zone. if ZR is the impedance setting. However.176 Protection of electricity distribution networks 9.6. this is not important since the main purpose is to retain the phase difference between them. Z > ZR /K.4. 9. the operating characteristic of the relay is determined by the locus of the points Z such that . The relay operates for all values of impedance less than the setting.e. the phase angle between S1 and S2 . 9. From the properties of the rhomboid.8) (9.7 by KI . the impedance relay will operate for all faults along the vector AB (see Figure 9.5 is obtained.6) The constant K takes into account the transformation ratios of the CTs and VTs. The corresponding signals for a phase comparator are S1 = KV + I ZR S2 = −KV + I ZR Dividing eqns. its operating characteristic in the R-X plane is a circle with its centre at the origin of the co-ordinates and a radius equal to the setting in ohms. i. Eqns.

4 Operating characteristic of an impedance relay obtained using a phase comparator X S1 Θ < 90° ZR K S2 Z O Operating zone R Figure 9.5 Impedance Z inside the operating zone of an impedance relay .Distance protection 177 X A ZR K S2 C Z O R B Θ = 90° Limit of operating zone S1 Figure 9.

7 Impedance relay characteristic in the complex plane .178 Protection of electricity distribution networks X S1 Θ > 90° ZR K S2 Z O R Figure 9.6 Impedance Z outside the operating zone of an impedance relay Line AC C A X B 21 Line AB B A R C Figure 9.

9) The operating zone of the directional relay is defined by the values of Z and ZR . 9.8 Operating zone of a directional relay .Distance protection 179 The impedance relay has three main disadvantages: 1.2 Directional relay Directional relays are elements that produce tripping when the impedance measured is situated in one half of the R-X plane. 3. 2. It is highly sensitive to oscillations on the power system. It is affected by the arc resistance. and defining Z = V /I . The construction of the characteristic is shown in Figure 9. which result in a phase difference between S1 and S2 of less than 90◦ . it will see faults in front and behind its location and therefore requires a directional element in order to obtain correct discrimination. It is not directional. gives S1 = Z S2 = ZR /K (9.8. They are commonly used together with impedance relays in order to limit the operating zone to a semi-circle. This can be obtained by adding an independent directional relay to restrict or prevent the tripping of the distance relay when power flows out of the protected zone during a fault.10) (9. The operating characteristic is obtained from a phase comparison of the following signals: S1 = KV S2 = ZR I Dividing by KI . due to the large area covered by its circular characteristic. in which S1 and S2 are drawn.2. X ZR = S2 K Operating zone Z = S1 Θ < 90° R Figure 9.

drawn for a reactance setting of XR /K. consequently. dividing by KI . the impedance measured by the relay.180 Protection of electricity distribution networks 9.12 in the complex plane and determining those values of Z for which is less than 90◦ . In a radial system this is generally true.3 Reactance relay The reactance relay is designed to measure only the reactive component of the line impedance. This apparent reactance can be positive or negative and add to. From the diagram it will be seen that the relay will measure a value that is smaller than the actual reactance between the relay point and the fault. thus affecting its operation.10 shows the voltage seen by the relay in the presence of faults with arc resistance and two infeeds.9 Operating zone of a reactance relay . If the fault resistance is large in comparison to the line reactance the effect could be serious and this type of relay should not be used. but not necessarily if the fault is fed from two or more points since the voltage drop in the fault resistance is added to the drop in the line and affects the relay voltage. the limit of the operating zone is a straight line parallel to the resistance axis. Unless the current in the relay is exactly in phase with the fault current. 9.12) (9.9. producing an effect similar to the line reactance.11) The operating characteristics are obtained by drawing eqns. Figure 9.2. the voltage drop in the fault resistance will result in a component 90◦ out of phase to the relay current. its setting is achieved by using a value determined by the reactance XR . The construction is shown in Figure 9. As the impedance of the fault is almost always resistive. or subtract from. X Θ < 90° XR K S1 Z 0 R Operating zone S2 = Figure 9. gives S1 = −Z + XR /K S2 = XR /K (9. In this case the pair of equations for S1 and S2 is as follows: S1 = −KV + XR I S2 = XR I and. here. it might be assumed that the fault resistance has no effect on the reactance relays.

(b) vector diagram 9. For values of Z located inside the circumference. .11. In this case the limit of the operating zone ( = 90◦ ).2. is traced by a circle with a diameter of ZR /K and a circumference that passes through the origin of the co-ordinates. as shown in Figure 9. and this will result in operation of the relay. drawn in the R-X plane.14) (9. the relay characteristic is determined by the locus for the values of Z that are fulfilled when is less than 90◦ . will be less than 90◦ .10 Voltage seen by a relay in the presence of faults with arc resistance and two infeeds: (a) schematic of circuit. in addition it has the advantage that the reach of the relay varies with the fault angle.4 Mho relay The mho relay combines the properties of impedance and directional relays.13) Drawing ZR /K and the eqns. 9. Its characteristic is inherently directional and the relay only operates for faults in front of the relay location.Distance protection 181 (a) Relay I1 R1 + jX1 Rf I2 R2 + jX2 Fault resistance (I1 + I2) Rf I1X1 Voltage seen by the relay I1 + I2 (b) I1 I1R1 I2 : Error due to the fault resistance I1 + I2 Figure 9. as is shown in Figure 9. The characteristic.14 in the R-X plane. is a circle with a circumference that passes through the origin of the co-ordinates and is obtained by assigning the signals the following values: S1 = −KV + ZR I S2 = KV from which S1 = −Z + ZR /K S2 = Z (9.12.

2. when it is used on long lines and the reach does not cover the section sufficiently along the resistance axis. . then it is incapable of detecting faults with high arc or fault resistances.11 Operating characteristic of a mho relay X ZR K S1 Θ < 90° Z = S2 O R Figure 9.5 Completely polarised mho relay One of the disadvantages of the autopolarised mho relay is that.182 Protection of electricity distribution networks X Limit of operating zone ZR K S1 Θ = 90° Z = S2 O R Figure 9. The problem is aggravated in the case of short lines since the setting is low and the amount of the R axis covered by the mho circle is small in relation to the values of arc resistance expected.12 Impedance Z inside the operating zone of a mho relay 9.

6 Relays with lens characteristics Distance relays with lens characteristics are very useful for protecting high impedance lines that have high power transfers. The polygonal tripping characteristic is obtained from three independent measuring elements – reactance.13.13 Operating characteristic of a completely polarised mho relay One practical solution to this problem is to use a completely polarised mho relay where the circular characteristic is extended along the R axis for all unbalanced faults.15) where: V = voltage at the location of the relay. not involved with the fault. This offset lens characteristic. In order to achieve this . which is fed by the following signals: S1 = Vpol S2 = V − I ZR (9. 9. I = fault current.14. This characteristic can be obtained by means of a phase comparator. on the faulted phase or phases. which can be adjusted to offset the mho circular characteristic as shown in Figure 9. or phases. 9. ZR = setting of the distance relay. which are equal to V 2 /S. as illustrated in Figure 9.7 Relays with polygonal characteristics Relays with polygonal characteristics provide an extended reach in order to cover the fault resistance. become small and get close to the impedance characteristics of the relay. since the position of the resistance line can be set in the tripping characteristic (see line 2 in Figure 9. in particular for short lines. Vpol = polarisation voltage taken from the phase.Distance protection 183 X Mho circle ZR K Operating characteristic for unbalanced fault O R Figure 9. is common in some relays.2. Under these conditions the line impedance values.2.15 which shows a typical polygonal operating characteristic). especially that of zone 3. resistance and directional.

14 Zone 3 offset lens characteristic X X4 X3 X2 Line Z4 characteristic Z3 Z2 Zone 4 (T4) Zone 3 (T3) X1 Z1 Zone 2 (T2) Zone 1 (T1) R1 R2 R3 R4 R Zone 5 (T5) Z5 Figure 9.15 Polygonal relay operating characteristic . Z2 Figure 9.184 Protection of electricity distribution networks X Zone 3 Zone 2 Zone 1 R Z1.

this value is obtained by using system currents and voltages from the measurement transformers that feed the relays.3 Setting the reach and operating time of distance relays Distance relays are set on the basis of the positive-sequence impedance from the relay location up to the point on the line to be protected. However. (9. the measuring elements are suitably combined. a mho circle that passes through ZK is employed.5.8 Relays with combined characteristics A typical combined operating characteristic is defined in the impedance plane by lines running parallel to the resistive and reactive axes which cross each other at the setting point for ZK . respectively. 9. the following expression is used: Vprim Vsec × VTR = Zprim = Iprim Isec × CTR Thus Zsec = Zprim × (CTR/VTR) (9. 9. as shown in Figure 9. starting from the location of the relay. Line impedances are proportional to the line lengths and it is this property that is used to determine the position of the fault. in this way the required polygonal characteristic is obtained. (b) R/X ratio = 2 characteristic.Distance protection 185 (a) X ZK (b) X ZK R R Figure 9.16) . in order to convert the primary impedance into a secondary value.16. In relays with this characteristic the reaches in the resistive and reactive directions have the same range of settings and can be adjusted independently of each other.2.16 Typical combined operating characteristics: (a) R/X ratio = 0. The relay is tripped only when all three elements have operated.17) where CTR and VTR are the transformation ratios of the current and voltage transformers. Therefore. which is used for the setting of the distance relay. In order to achieve the required directionality.

The remaining 20–15 per cent provides a factor of safety in order to mitigate against errors introduced by the measurement transformers and line impedance calculations. if necessary. Since the reach and therefore the operating time of the .17 Distance relay protection zones for a radial system Normally. is normally set by the manufacturer to trip instantaneously since any fault on the protected line detected by the zone 1 unit should be cleared immediately without the need to wait for any other device to operate. the operating time for the zone 3 units should also be set at a value that will ensure that system stability is maintained and therefore. which operates in t2 s. zone 3: this is set to cover all the protected line plus 100 per cent of the second longest line. However. The 20–15 per cent to the end of the line is protected by zone 2. In addition to the unit for setting the reach.2 s to avoid incorrect co-ordination. the latter acting as a back-up to protect the busbars. three protection zones in the direction of the fault are used in order to cover a section of line and to provide back-up protection to remote sections (see Figure 9. In the case of zone 3. each zone unit has a timer unit.17). t1 . The operating time for zone 1.17) so it is set to cover only 80–85 per cent of the protected line. Since the tripping produced by zone 1 is instantaneous.4 s. it should not reach as far as the busbar at the end of the first line (see Figure 9. then the timer for the zone 3 of the furthest relay should be increased by at least 0. Some relays have one or two additional zones in the direction of the fault plus another in the opposite sense. The operating time for zone 2 is usually of the order of 0. when the settings of relays at different locations overlap.25 to 0. zone 2: this is set to cover all the protected line plus 50 per cent of the shortest next line.2 s over the tripping time of any associated transformer overcurrent protection. In the majority of cases the setting of the reach of the three main protection zones is made in accordance with the following criteria: • • • zone 1: this is set to cover between 80 and 85 per cent of the length of the protected line. When there are power transformers at adjacent substations the zone 2 timer should have a margin of 0.0 s. plus 25 per cent of the shortest next line. Zone 3 provides the back-up and operates with a delay of t3 s.186 Protection of electricity distribution networks t3 t x A Relay 80% AB B t2 AB + 50% BC C AB + BC + 25% CD D Figure 9.6 to 1. and that of zone 3 is in the range of 0. consideration may have to be given to reducing the zone 3 operating time in such circumstances.

In particular. For a fault on line BC. In this case the times for zones 2 and 3 should not have a fixed value.19. their co-ordination is much easier than that for overcurrent relays. both the zone 2 and zone 3 units should operate. no specific recommendation is . the relay at A should provide an opportunity for the breaker at B to clear the fault. it is for this reason that the zone 2 and zone 3 units operate with an appropriate time delay in order to obtain discrimination between faults on lines AB and BC. In order to illustrate the philosophy referred to earlier. Since the same philosophy is used as the basis for either method. but within the cover of the zone 2 unit at A. All three units should operate for a fault within the operating characteristic of zone 1. it is necessary to have three relays at A to set the three zones. Some methods for setting distance relays use different criteria to those already mentioned. Since there is also distance protection at substation B.18 in which it is required to protect the lines AB and BC. there is the method where it is recommended that the reach of zone 2 should be 120 per cent of the impedance of the line to be protected.Distance protection 187 A B C X Zone 3 unit C B Zone 2 unit Zone 1 unit R A Figure 9. mainly with regard to the reach of zones 2 and 3. For this. The diagram of operating times is shown in Figure 9. and that for zone 3 should be 120 per cent of the sum of impedances of the protected line and of its longest adjacent line. but should be based on the opening time of the breakers and the reach of the relays to guarantee that there will be no overlap in the same zones covered by adjacent relays. consider the case of the system in Figure 9.18 Operating characteristic of distance protection located at A distance relays are fixed.

The time delay for zones 4 and 5 is normally the same as that for zone 3 but increased by a margin of. zones 3 and 4 provide cover only in the forward direction and zone 5 in the backward direction. i.19 Operating times for distance protection at A made to use one or the other.188 Protection of electricity distribution networks A B C Distance Zone 1 (A) t2 t3 Operating time Zone 2 (A) Zone 3 (A) Figure 9. or load impedance values. but the settings for zones 4 and 5 can vary from utility to utility. Care should be taken to ensure that the zones with the higher settings. In fact. in addition to the line impedance ZA . typically. In these cases.20. Analysing the case illustrated in Figure 9. which implies that its reach is reduced. do not overlap different voltage levels through step-up or step-down transformers. it can be appreciated that the impedance seen by the distance relay at A for a fault beyond busbar B is greater than actually occurs. especially the numerical types. 9. as shown in Figure 9. Some accepted criteria suggest setting zone 4 at 120 per cent of zone 3. 400 ms. The setting for the first three zones is the same as discussed before. Modern distance relays. if a solid earth-fault is present at F.15.19) The relay th