Turn-To-Turn Fault Detection in Transformers Using Negative Sequence Currents

TURN-TO-TURN FAULT DETECTION IN
TRANSFORMERS USING NEGATIVE SEQUENCE

CURRENTS
A Thesis
Submitted to the College of Graduate Studies and Research
In Partial Fulfillment of the Requirements
For the Degree of Master of Science
In the Department of Electrical and Computer Engineering
University of Saskatchewan
Saskatoon, Saskatchewan, Canada
by
MARIYA BABIY
© Copyright Mariya Babiy, September 2010. All Rights Reserved.

PERMISSION TO USE
I agree that the Libraries of the University of Saskatchewan may make this thesis freely available
for inspection. I also agree that permission for the copying of this thesis in any manner, in part
or whole, for scholarly purposes may be granted by the professor who supervised the thesis work
or, in his absence, by the Head of the Department of Electrical Engineering or the Dean of the
College of Engineering. Any publication, use of this thesis or copying for financial gain is
prohibited without my written permission and approval by the University of Saskatchewan. It is
also understood that due recognition shall be given to the author and the University of
Saskatchewan in any scholarly use which may be made of any material in this thesis.
Request for permission to copy or to make other use of material in this thesis in whole or part
should be addressed to:
Head of the Department of Electrical Engineering,

57 Campus Drive
University of Saskatchewan,
Saskatoon, Saskatchewan
Canada S7N 5A9
i
ABSTRACT
A power transformer is one of the most important and expensive components in any power
system. Power transformers can be exposed to a wide variety of abnormal conditions and faults.
Internal turn-to-turn faults are the most difficult types of faults to detect within the power
transformer. The IEEE Standards documents have revealed that there is no one standard way to
protect all power transformers against minor internal faults such as turn-to-turn faults and at the
same time to satisfy basic protection requirements: sensitivity, selectivity, and speed.
This thesis presents a new, simple and efficient protection technique which is based on negative
sequence currents. Using this protection technique, it is possible to detect minor internal turn-to-
turn faults in power transformers. Also, it can differentiate between internal and external faults.
The discrimination is achieved by comparing the phase shift between two phasors of total
negative sequence current. The new protection technique is being studied via an extensive
1
simulation study using PSCAD®/EMTDC™ software in a three-phase power system and is
also being compared with a traditional differential algorithm.
Relay performance under different numbers of shorted turns of the power transformer, different
connections of the transformer, different values of the fault resistances, and different values of
the system parameters was investigated. The results indicate that the new technique can provide
a fast and sensitive approach for identifying minor internal turn-to-turn faults in power
transformers.
1
PSCAD®/EMTDCTM : Trademark of the Manitoba HVDC Research Centre.
ii
ACKNOWLEDGEMENTS
I would like to express my sincere gratitude to my supervisor, Dr. Ramakrishna Gokaraju, for his
guidance, supervision and suggestions.
I also wish to express my sincere thanks to Dr. M. Ramamoorty, Former Director-General,

Central Power Research Institute, India and Dr. Neelima Tambey, Crompton Greaves Limited
for providing the initial literature and guidance for pursuing this research project.
I also would like to thank Mr. Juan Carlos Garcia and Dr. Dharshana Muthumuni for their help
and guidance related to transformer modeling in PSCAD/EMTDC.
I specially would like to express my deepest gratitude to Donald Robinson who was with me
from start to finish. Thank you very much for your support, guidance, help, and encouraging me
throughout my studies at the University of Saskatchewan.
Also, I am very grateful to one of my colleagues, Mohamed El-Samahy, my one-of-a kind

mentor and dear friend who will never be forgotten. Thank you very much for many useful
discussions I had with you. I would also like to thank all my colleagues in the Electrical and
Computer Engineering Department.
Many Thanks to my amazing friends Mariya, Olga, Alex, Nelya, Stacey, Iryna, Anastasiya,
Olga, Andriy, Tetyana and so many thanks to others, especially to Ellen, Voula and John.
Finally, I would like to express heartfelt appreciation to my parents, Oksana and Mykhailo, and
my brother Vasyl for all their love, patience, understanding and support throughout my years of
studies.
iii
DEDICATION
This thesis is dedicated to my parents, Oksana and Mykhailo, who have raised me to be the
person who I am now. Also, this thesis is dedicated to Donald Robinson, who believed in me,
inspired me and has supported me all the way from the beginning of my graduate studies.
Thank you for all your support, guidance, love and help.
iv
TABLE OF CONTENTS
Page
PERMISSION TO USE…………………………………………………………………... i
ABSTRACT……………………………………………….……………………………..... ii
ACKNOWLEDGEMENTS…………………………….…………………………...….... iii
DEDICATION……………………………………………………………………………. iv
TABLE OF CONTENTS………………………………..…………………………..….... v
LIST OF FIGURES…………………………………..………….……………………….. ix
LIST OF TABLES……………………………………………….……………...………... xvi
LIST OF SYMBOLS AND ABBREVITIONS …………………………………………. xviii
CHAPTER 1 INTRODUCTION…………………………………………….…………... 1
1.1 Introduction to the power transformer……………………………………………...... 1
1.1.1 Principles of the power transformer…………………………..…………...….... 2
1.1.2 Types of transformer failures…………………………………………………... 5
1.2 Transformer internal turn-to-turn faults………………………………………….….. 6
1.3 Literature review………….…………………………………………..…………….... 12
1.4 Objectives of the research ……………………………………………………….…... 17
1.5 Outline of the thesis………………………………………………………………...... 17
v
CHAPTER 2 UNSYMMETRICAL FAULTS...……………………………………….... 19
2.1 Symmetrical components……………………………...………………………….….. 19
2.2 Sequence impedances for the power transformer…………………………………...... 24
2.3 Sequence networks connections and symmetrical components for various types of
common faults………………………………………………………………………... 27
2.4 Sequence network connections and symmetrical components for turn-to-turn faults
in power transformers………………………………………….…………………….... 30
2.5 Summary……………………………………………………………………………... 35
CHAPTER 3 SENSETIVE TURN-TO-TURN FAULT DETECTION USING

NEGATIVE SEQUENCE CURRENTS…………………………………………………. 36
3.1 Introduction to the new transformer protection method…………….……………...... 36
3.1.1 Flow of the negative sequence currents for transformer external faults……….. 37
3.1.2 Flow of the negative sequence currents for transformer internal faults………... 38
3.2 The description of the new negative sequence current based protection method……. 38
3.3 Summary……………………………………….…………………………..…….…... 41
CHAPTER 4 PSCAD/EMTDC MODELNG OF THE POWER TRANSFORMER

AND THE DIFFERENTIAL RELAY…………………………………………………....
42
4.1 Tools required to model the power system and relays……………………………...... 42
4.2 The modeled power system……………………………………………….……....…. 42
4.2.1 Voltage sources…………………………………………………………..…….. 43
4.2.2 Circuit breakers ….….…………………………………………………..……... 43
4.2.3 Current transformers……………………………………………………..…….. 44
4.2.4 Power transformer model for simulating internal turn-to-turn faults …..……… 45
vi
4.2.4.1 The calculation of the leakage reactance of the power transformer….... 47
4.3 Modeling differential relays in PSCAD/EMTDC™………………………………..... 48
4.3.1Traditional differential protection method…………………………..………….. 49
4.3.1.1 The percentage differential slope……………………………..……….... 49
4.3.2 Description of the negative sequence currents‟ fault detection component ……. 51
4.4 Summary…………………………………………………….……………...………... 54
CHAPTER 5 TEST RESULTS …………………………………………………….……. 56
5.1 Introduction………………………………………………………………………..…. 56
5.2 Performance of the traditional differential protection……………………………..…. 56
5.2.1 Differential current versus restraining current…………………………….....… 57
5.2.2 Primary and secondary phase currents during internal turn-to-turn faults…….. 61
5.3 Performance of the proposed method for internal turn-to-turn faults……………...... 68
5.3.1 Turn-to-turn faults on the secondary winding of the power transformer
connected in Y-Y………………………….………………………..………….. 68
5.3.1.1 Sensitivity analysis of the proposed relay for turn-to-turn faults on the
secondary winding (Y-Y)………………………………………..…….. 77
5.3.2 Turn-to-turn faults on the primary winding of the power transformer
connected in Y-Y………………………….………………………...………….. 80
primary winding (Y-Y)……………………………….…………….….. 87
5.4 Performance of the proposed method for external faults…………………………….. 89
5.4.1 External phase B-to-ground fault on the high voltage side of the power
transformer…………………………..…………………………………………. 90
5.4.2 External phase B-C fault on the low voltage side of the power
vii
transformer……………………………………………………………………... 92
5.5 Performance of the proposed method for small unbalances in the power system….... 95
5.6 Performance of the proposed method during the CT saturation……………………... 98
5.6.1 CT saturation during turn-to-turn faults……………………………...………... 98
5.6.2 CT saturation during other miscellaneous internal faults…………………….... 99
5.7 Performance of the proposed method during inrush current…….………………….... 105
5.8 Summary…………………………………………………………………..…...…….. 106
CHAPTER 6 SUMMARY AND CONCLUSIONS……………………………………... 108
6.1 Summary……………………………………………………………………….…….. 108
6.2 Conclusion……………………………………………………………………….….... 109
6.3 Contribution……………………………………………………………………….…. 109
6.4 Future work……………………………………………………………………….….. 110
REFERENCES……………………………………………………………………………. 111
APPENDIX A. DATA AND PARAMETERS OF THE MODELED POWER

SYSTEM…………………………………………………………………………………… 115
APPENDIX B. PERFORMANCE OF THE PROPOSED METHOD FOR Δ-Y

TRANSFORMER……………….……………………………………………………...…. 118
viii
LIST OF FIGURES
Page
Figure 1.1 The ideal transformer…………………………………………………... 3
Figure 1.2 The single-phase, two winding transformer……………………………. 7
Figure 1.3 The ampere-turn-balance principle…………………………………….. 7
Figure 1.4 Percentage differential relay……………………………………………. 9
Figure 1.5 Failure of transformer by shorted turns………………………………… 14
Figure 2.1 Unbalanced current phasors……………………………………………. 19
Figure 2.2 Positive sequence components……………………………………….… 19
Figure 2.3 Negative sequence components………………………………………... 20
Figure 2.4 Zero sequence components…………………………………………..… 20
Figure 2.5 Positive sequence components expressed in terms of phase A
quantities……………………………………………………………..… 21
Figure 2.6 Negative sequence components expressed in terms of phase A
quantities……………………………………………………………..… 21
Figure 2.7 Zero sequence components expressed in terms of phase A
quantities……………………………………………………………..… 21
Figure 2.8 Sum of sequence components………………………………………..… 22
Figure 2.9 The equivalent zero sequence circuit for Y-Y connections
with both neutrals grounded……………………………………………. 25
Figure 2.10 The equivalent zero sequence circuit for Y-Y connections
with the primary neutral grounded and the secondary
neutral isolated…………………………………………………………. 25
Figure 2.11 The equivalent zero sequence circuit for ∆-∆connections…………….. 26
Figure 2.12 The equivalent zero sequence circuit for Y-∆ connections
with neutral grounded………………………………………………….. 26
ix
Figure 2.13 The equivalent zero sequence circuit for Y-∆ connections
with neutral isolated……………………………………………………. 27
Figure 2.14 The sequence connection for phase-to-ground fault…………………... 28
Figure 2.15 The sequence connection for phase-to-phase fault…………………….. 29
Figure 2.16 The sequence connection for phase-to-phase-to-ground fault……….… 29
Figure 2.17 Shunt unbalance with a shorted turn in phase A……………………….. 30
Figure 2.18 Sequence network for a turn-to-turn fault in phase A………………….. 31
Figure 2.19 Phase currents and negative sequence currents during the steady state
condition and during the fault for 5% of shorted turns………………… 32
condition and during the fault for 3% shorted turns……………….…... 32
condition and during the fault for 1% shorted turns…………..………... 33
Figure 3.1 Flow of negative sequence currents for transformer external fault…..… 37
Figure 3.2 Flow of negative sequence currents for transformer internal fault…….. 38
Figure 3.3 Negative sequence current based logic………………………..……..… 39
Figure 4.1 Simulated power system model………………………………………… 42
Figure 4.2 The voltage source component model…………………………………. 43
Figure 4.3 The circuit breaker component model…………………………………. 44
Figure 4.4 The current transformer component model……………………………. 44
Figure 4.5 A single phase transformer with a turn-to-turn fault on secondary
winding ……………………………………………………...……….... 45
Figure 4.6 A graphic representation of a single phase transformer with internal
turn-to-turn faults………………………………………………………. 47
Figure 4.8 Magnitudes and phase angles of sequence components produced
by FFT blocks …………………………………………………………. 52
Figure 4.9 Comparison of negative sequence currents with a pre-set level……….. 53
Figure 4.10 Phase angle comparison between two phasors of negative
sequence currents………………………………………………………. 54
x
Figure 5.1 Internal turn-to-turn fault on the secondary winding of the
power transformer………………………………………………………. 57
Figure 5.2 Traditional differential relay response when 25% of turns are shorted... 58
Figure 5.5 Traditional differential relay response when 5% of turns are shorted…. 60
Figure 5.8 Phase currents on both sides when 25% of turns are shorted………..…. 62
Figure 5.9 Phase currents on both sides when 15% of turns are shorted………...… 63
Figure 5.10 Phase currents on both sides when 10% of turns are shorted………..…. 64
Figure 5.11 Phase currents on both sides when 5% of turns are shorted……………. 65
Figure 5.12 Phase currents on both sides when 3% of turns are shorted……………. 66
Figure 5.13 Phase currents on both sides when 1% of turns are shorted………….… 67
Figure 5.14 Negative sequence current magnitudes for 25% shorted turns on
secondary (Y-Y)……………………………………………….....…….. 69
Figure 5.15 Negative sequence current magnitudes for 15% shorted turns
on secondary (Y-Y)……………………...………………..…………..... 69
on secondary (Y-Y)…………………………….……………………..... 69
on secondary (Y-Y)………………………….…………………….….... 70
on secondary (Y-Y)…………………………….………………….….... 70
on secondary (Y-Y)…………..…………………...………………..…... 70
Figure 5.20 Zooming in on negative sequence current magnitudes for 25%
shorted turns on secondary (Y-Y)…………………….……………....… 71
shorted turns on secondary (Y-Y)…………………………………....… 71
xi
shorted turns on secondary (Y-Y)………………………………….....… 72
shorted turns on secondary (Y-Y)…………………….………………… 72
shorted turns on secondary (Y-Y)……………………………..……..… 72
Figure 5.25 Zooming in on negative sequence currents magnitudes for 1%
shorted turns on secondary (Y-Y)……………………………..……...… 73
Figure 5.26 Phase angle comparison between two negative sequence currents
for 25% shorted turns on secondary (Y-Y)……….…………………….. 73
for 15% shorted turns on secondary (Y-Y)…………........................…... 74
for 10% shorted turns on secondary (Y-Y)………..……………….....… 74
for 5% shorted turns on secondary (Y-Y)……...……………………..… 74
for 3% shorted turns on secondary (Y-Y)…………..…..……………..... 75
for 1% shorted turns on secondary (Y-Y)…………..….....................…. 75
Figure 5.32 Trip signal from the proposed relay for 25% shorted turns
shorted turns on secondary (Y-Y)…………………….…………...…… 76
Figure 5.33 Trip signal from the proposed relay for 15% shorted turns shorted turns
on secondary (Y-Y)………………………………………………..…… 76
on secondary (Y-Y)……………………………………………….….… 76
on secondary (Y-Y)……………………………………………..……… 76
on secondary (Y-Y)……………………………...………………...…… 77
xii
on secondary (Y-Y)……………………………...…………………...… 77
primary (Y-Y) ……….………………………………..….................…. 81
primary (Y-Y) ……….………………………………..….................….. 81
primary (Y-Y) ……….………………………………..……………..…. 82
primary (Y-Y) ……….……………………….………..…...…………... 82
primary (Y-Y) ……….………………………………..….................…. 82
shorted turns on primary (Y-Y) …………………………………...…… 83
shorted turns on primary (Y-Y) …………………………………...…… 83
shorted turns on primary (Y-Y) ……………………………………...… 84
shorted turns on primary (Y-Y) ……………………………………...… 84
shorted turns on primary (Y-Y) …………………..……………..…...… 84
for 25% shorted turns on primary (Y-Y)………………..…………….... 85
for 15% shorted turns on primary (Y-Y)……………...……………..…. 85
for 10% shorted turns on primary (Y-Y)……………….....................…. 86
for 3% shorted turns on primary (Y-Y)………...……..………………. 86
xiii
for 1% shorted turns on primary (Y-Y)………….....………………..…. 86
Figure 5.53 Phase currents on both sides of the power transformer for
external phase B-to-ground fault on HV side………………………..… 91
Figure 5.54 Negative sequence current magnitudes for external phase
B-to-ground fault on HV side…………………………...…..………..… 91
Figure 5.55 Phase angle between two phasors of negative sequence currents
for external phase B-to-ground fault on HV side…..…………...….. 92
Figure 5.56 Output signal from the proposed technique for external phase B-to-
ground fault on HV side ……………………………………………….. 92
Figure 5.57 Phase currents on both sides of the power transformer for external
phase B-C fault on LV side…………………………….…………...….. 93
Figure 5.58 Negative sequence current magnitudes for external phase B-C fault on
LV side……………………………………………………………….… 94
for external phase B-C fault on LV side………………………….……. 94
Figure 5.60 Output signal from the proposed technique for external phase B-C fault
on LV side………………………………………………………….….. 95
Figure 5.61 N Negative sequence current magnitudes during a small unbalance in
the power system…………………………...……………………...…… 96
Figure 5.62 P Phase angle between two phasors of negative sequence currents
during a small unbalance in the power system……………...………….. 97
Figure 5.63 T Output signal from the proposed technique for a small unbalance in
t the power system………………………………………………………. 97
Figure 5.64 S Secondary currents produced by CT for of non-saturated and
saturated CT cases……………………………………………………… 101
Figure 5.65 P Negative sequence current magnitudes for internal phase B-C
fault for non-saturated and saturated CT cases………………………… 102
Figure 5.66 P Phase angle between two phasors of negative sequence currents for
phase B-C fault for non-saturated and saturated CT cases………….…. 103
xiv
Figure 5.67 T Output signal from the proposed technique for phase B-C fault for
non-saturated and saturated CT cases…………………………….…..... 104
Figure 5.68 l Negative sequence currents magnitudes during inrush current
phenomenon…………………………………………….….......….….... 105
during inrush current phenomenon…………………………….………. 106
Figure 5.70 Output signal from the proposed technique during inrush current…....... 106
Figure B.1 N Negative sequence current magnitudes for 5% shorted turns on
secondary (Δ-Y)……………………………………..…………...…….. 118
secondary (Δ-Y)………………………………………...…………..….. 119
secondary (Δ-Y)………………………………………..........…….…… 119
Figure B.4 Phase angle comparison between two negative sequence currents
for 5% shorted turns on secondary (Δ-Y) with 30º phase shift ……….. 120
for 3% shorted turns on secondary (Δ-Y) with 30ºphase shift ………… 120
for 1% shorted turns on secondary (Δ-Y) with 30º phase shift………… 121
for 5% shorted turns on secondary (Δ-Y) without 30º phase shift..…..... 121
for 3% shorted turns on secondary (Δ-Y) without 30º phase shift …….. 122
for 1% shorted turns on secondary (Δ-Y) without 30º phase shift……... 122
xv
LIST OF TABLES
Page
Table 2.1 Magnitudes of the phase currents and negative sequence currents
during steady state condition and during an internal turn-to-turn fault
for 5% shorted turns……………………………………………………. 34
Table 2.2 Magnitudes of the phase currents sand negative sequence currents
for 3% shorted turns…………..…………..……………………………. 34
Table 2.3 Magnitudes of the phase currents and negative sequence currents
for 1% shorted turns………..……………….………………………….. 34
Table 5.1 T Turn-to-turn fault on the secondary winding of the power
transformer connected in Y-Y for various percentages of
shorted turns (Rf = 0Ω)………………………...…………………….…. 78
Table 5.2 T Turn-to-turn fault on the secondary winding of the power
shorted turns (Rf = 1Ω)……………………………………………....…. 79
Table 5.3 R Results using different system parameter values (Y-Y transformer,
turn-to-turn fault applied on the secondary winding)……………...…... 80
Table 5.4 T Turn-to-turn fault on the primary winding of the power
shorted turns (Rf = 0Ω)…………………………………………….….... 87
Table 5.5 T Turn-to-turn fault on the primary winding of the power
shorted turns (Rf = 1Ω)…………………………………………………. 88
xvi
Table 5.6 R Results using different system parameter values (Y-Y transformer,
turn-to-turn fault applied on the primary winding)…………………...... 89
Table 5.7 P Phase currents due to a small unbalance……………………………….. 96
Table A.1 P Parameters of source 1………………………………..………………... 115
Table A.2 P Parameters of source 2………………………………..………………… 115
Table A.3 P Parameters of CT on primary side………………………………..…….. 115
Table A.4 P Parameters of CT on secondary side………………………………..….. 115
Table A.5 P Parameters of a single phase power transformer……………………….. 116
Table A.6 Leakage reactance of a single phase transformer for internal turn-to-
turn fault on secondary winding of the power transformer…………….. 116
Table A.7 L Leakage reactance of a single phase transformer for internal turn-to-
turn fault on primary winding of the power transformer……………….. 117
xvii
LIST OF SYMBOLS AND ABBREVIATIONS
IEEE The Institute of Electrical and Electronics Engineers

PSCAD Power Systems Computer Aided Design
EMTDC The Electromagnetic Transient Program
DC Direct current
AC Alternating current
HV High voltage
LV Low voltage
kVA Kilo Volt Ampere
MVA Mega Volt Ampere
f Frequency
Hz Hertz
io Excitation current
Φ The mutual flux in the transformer‟s core
Vp The primary voltage of the transformer
Ep The voltage at the primary terminal of the transformer
Vs The secondary voltage of the transformer
Np The number of turns in the primary coil of the transformer
Ns The number of turns in the secondary coil of the transformer
Фmax The maximum value of the flux
ω Angular frequency
RMS Root mean square
Ip The primary current
Is The secondary current
Nx Number of shorted turns
Ix The additional current component caused by the induced voltage in the short
circuited turns
CT Current transformer
xviii
RC Restraining coil
OP Operating coil
DWT Discrete wavelet transform
ANN Artificial neural network
Ia Current of phase A
Ib Current of phase B
Ic Current of phase C
I1 Positive sequence current
I2 Negative sequence current
I0 Zero sequence current
IN Neutral current
a phasor operator equal to 120 degree
The inverse matrix
Z1 Positive sequence impedance
Z2 Negative sequence impedance
Z0 Zero sequence impedance
Zf Fault impedance
Zl Leakage impedance
Vf The pre fault voltage
Za Impedance of phase A
Zb Impedance of phase B
TR Transformer
ZNS_S1, ZNS_S2 The negative sequence impedances for the equivalent sources S1 and S2
ENS Fictitious negative sequence source
INS_P, INS_S The negative sequence currents on the primary and secondary side of TR
INS_P' The negative sequence current transformed from the fault side to the other side
of TR
INS_P , INS_S The negative sequence currents on the primary and secondary side of TR scaled
down using CTs
FFT Fast Fourier Transform
xix
Imin The minimum pre-set level
S1,S2 Voltage sources
BRK Circuit breaker
V Line-to-ground, peak voltage magnitude
S The maximum load
L1 self-inductance in the primary winding
La, Lb, Lc self-inductance of sub coils “a”, “b” and “c”
M1a , M1b, M1c mutual inductance between primary winding and sub coils “a”, “b” and “c”
Mab , Mac, Mbc mutual inductance between sub coils “a”, “b” and “c”
N Number of turns of the winding
l The mean length of the magnetic circuit
μ0 The permeability
Ag Cross-sectional area of core
L Inductance
Id Differential current
K Compensation factor
Ir Restraint current
SLP The slope of the percentage differential characteristic
TTC Transformer tap changing
M The percentage of the current‟s mismatch
IL, IH Relay input currents at the same kVA base for low and high
voltage sides respectively
TL, TH Relay tap settings for low and high voltage sides respectively
Mag+ Positive sequence magnitude component
Mag- Negative sequence magnitude component
Mag0 Zero sequence magnitude component
Ph+ Positive sequence phase component
Ph- Negative sequence phase component
Ph0 Zero sequence phase component
xx
IA1, IB1, IC1 Phase currents on the primary side of the power transformer in phase A, B, and
C respectively
IA2, IB2, IC2 Phase currents on the secondary side of the power transformer in phase A, B,
and C respectively
ISA1, ISB1, ISC1 Secondary phase currents produced by CT‟s on the primary side of the power
transformer in phase A, B, and C respectively
ISA2, ISB2, ISC2 Secondary phase currents produced by CT‟s on the secondary side of the
power transformer in phase A, B, and C respectively
INS_P_ph Phase angle of the negative sequence current on the primary side of TR is
output from FFT block
INS_S_ph Phase angle of the negative sequence current on the secondary side of TR is
output from FFT block
LCUR Line current unbalance ratio
RTDS ™ 2 Real Time Digital Simulator
2
Registered trademark of RTDS Technologies, Winnipeg, Canada.
xxi
CHAPTER 1
INTRODUCTION
1.1 Introduction to the power transformer
One of the most important and complex systems that has been built by human civilization is the
power system. The electric power system plays a key role in modern society. One of the most
important components of any power system is the power transformer.
The development of the first power transformer significantly changed transmission and
distribution systems. Before the power transformer was invented in 1885 by William Stanley,
power was produced and distributed as direct current (DC) at low voltage. The use of electricity
was only limited to urban areas because of voltage drop in the lines. Also, the use of DC voltage
would require the supply voltage to be the same as the supply voltage required by all electrical
equipment connected to the power system. However, it is quite difficult to transform DC power
to a lower current and high-voltage form efficiently. The alternating current (AC) system is able
to overcome the limitation of the DC system and distributes electricity efficiently over long
distances to consumers. The use of the AC system allows the use of a multi-voltage level energy
delivery system. The AC power, generated at a low voltage, can be stepped up by using the
power transformer for the transmission purpose to higher voltage and lower current. Thus,
voltage drops and transmission energy losses can be reduced.
The transformer is the link between the generator of the power system and the transmission lines.
A power transformer is defined in reference [1] as a static piece of apparatus with two or more
windings which by electromagnetic induction, transforms a system of alternating voltage and
current into another system of voltage and current usually of different values and at the same
frequency for the purpose of transmitting electrical power.
1
The power transformer makes it possible to transmit the power from the generating station to the
customers at a distance. The power transformer is the most important and critical component of
the power system that transfers electrical energy from one circuit to another through the power
transformer‟s coils. Power transformers are able to transfer 99.5% of their input power as their
output power.
The life-span of the power transformer is about 30 years in service. Transformers are divided
into three groups [2]:
1) Small power transformers 500 – 7500 kVA.
2) Medium power transformers 7500 kVA – 100 MVA.
3) Large power transformers 100 MVA and above.
However, the power transformer can sometimes fail due to different factors. One of the major
factors that can cause the power transformer to fail is increasing the power consumption.
Increasing the power consumption every year leads the load on the power transformer to grow.
When the load on the power transformer increases, the operating stresses increase as well.
According to the report on analysis of transformer failures presented at International Association
of Engineering Insurers, 36th Annual Conference in Stockholm in 2003 by William H. Bartley,
the largest transformer loss occurred at a power plant in 2000 and the total damage loss of this
power transformer was 86 million US dollars [3]. The continuity of the power transformer
operation is very important in maintaining the reliability of the power supply. Any unscheduled
repair work such as replacement of the faulty transformer is very expensive and time
consuming.
1.1.1 Principles of the power transformer
The operation of a power transformer is based on two principles:

1. Electric current produces the magnetic field.
2
The primary winding of the transformer is connected to a source of sinusoidal voltage of
frequency f Hz and draws a small excitation current io from the source. This excitation current
sets up the mutual flux in the transformer‟s core Φ.
2. A change in the magnetic field within a coil of wire induces a voltage across the ends of the
coil (electromagnetic induction).
Changing the current in the primary coil changes the magnitude of the applied magnetic field.
The changing magnetic flux extends to the secondary coil where a voltage is induced across its
ends. Therefore, the primary and secondary windings of the transformer are not connected
electrically, but magnetically [2]. Electric energy is transferred between two circuits without the
use of moving parts. The ideal power transformer is depicted in Figure 1.1.
IP NP : NS IS
VP EP ES VS
Figure 1.1 – The ideal transformer
For the ideal power transformer, it is assumed that resistance of the primary and the secondary
windings equal zero. Since the resistance is zero, there is no voltage drop in the transformer
winding, so the voltage at the power transformer terminals equals the induced voltage in the
winding (Vp = Ep).
The secondary voltage Vs, induced in the secondary winding, is proportional to the primary
voltage Vp, and is given by the ratio of the number of turns Ns in the secondary winding to the
number of turns Np in the primary winding [2]. The voltage induced in the secondary coil is:
3
(1.1)
Where Ns - the number of turns in the secondary coil,

Φ - the magnetic flux through one turn of the coil.
The same magnetic flux passes through the primary and the secondary coils of the power
transformer causing the instantaneous voltage across the primary winding to be equal to:
(1.2)
Where Np - the number of turns in the primary coil,

Φ - the magnetic flux through one turn of the coil.
The waveforms of the flux and the voltage are sinusoidal. The instantaneous value of the flux is:
(1.3)
Then the induced voltage is:
(1.4)
Where - the maximum value of the flux,

- angular frequency.
The RMS value of the induced voltage is:
(1.5)
Since the voltage drop in the winding resistance is neglected and then induced voltage equals the
terminal voltage. Then if a sinusoidal voltage is applied to the winding, sinusoidally varying
flux has to be established with the maximum value and satisfy the requirements that
induced voltage equals the RMS value of the terminal voltage.
4
(1.6)
As can be seen, the core flux depends on the applied voltage, the number of turns in the winding,
and the frequency. The core flux is fixed by the applied voltage and also by the required exciting
current which is determined by the magnetic properties of the core.
1.1.2 Types of transformer failures
Damages to the power transformer can be caused by different stresses, which are due to
overheating, open circuits and short circuits [4]. The major focus of power transformer
protection is short circuits because open circuits do not present a particular hazard. Short circuit
protection includes internal and external faults.
External faults occur outside the protection zone of the transformer and include:
1. Over voltage – stresses the insulation beyond its withstand capacities and therefore causing
breakdown of the transformer.
2. Overloads – lead to overheating of transformers‟ insulation and have the potential to cause
permanent damage.
3. Under frequency – caused by a system disturbance that causes an imbalance between
generation and load. The exciting current increases at low frequencies causing overfluxing
of the transformer iron core. Overfluxing may gradually lead to insulation breakdown of
the magnetic circuit.
Internal faults occur within the transformer protection zone. The internal faults can be divided
into two groups [3] - [6]:
1. Incipient faults develop slowly and may develop into major faults such as phase-to-ground
faults or three phase faults if the cause is not detected and corrected. Incipient faults can be
divided into three groups:
5
 Overheating – caused by loss of coolant which is due to leakage, poor internal
connection in the electric or magnetic circuits, and loss of fans which have to provide
cooling.
 Over fluxing – may lead to insulation breakdown of the magnetic circuit insulating
materials.
 Overpressure – occurs due to release of gases from the transformer.
2. Active faults – caused by the breakdown in insulation which creates a sudden stress. The
active fault occurs when the current flows from one phase conductor to another such as
phase-to-phase and phase-to-ground. These faults may occur suddenly and they require fast
action by protective relays. Active faults:
 Turn-to-turn short circuits.
 Phase-to-phase short circuits.
 Phase-to-ground short circuits.
 Tank faults.
 Core faults.
One of the most challenging problems in power transformer protection is detecting internal turn-
to-turn faults. The reason why it is difficult to detect minor internal faults within the power
transformer is explained in the following sections.
1.2 Transformer internal turn-to-turn faults
According to the IEEE Standard C37.91-2000, which gives failure statistics of transformers for
different time periods, more than 50% of the total numbers of failures are winding failures [7].
Winding failures are due to insulation between turns which can break down due to
electromagnetic and mechanical forces on the winding.
A single-phase, two winding transformer with an internal turn-to-turn fault is depicted in Figure
1.2 to explain the theory about turn-to-turn faults. A turn-to-turn short circuit is a short circuit of
6
a few turns in the transformer windings. Figure 1.2 shows that a turn-to-turn fault has occurred
in the secondary windings and involves Nx turns.
II1p II2s
N1 N2-Nx
Np Ns - Nx
Ix
Ix
Nx
N x
Figure 1.2 – The single-phase, two winding transformer
Where Ip, Is - the primary and secondary current,

Np, Ns - number of turns on the primary and secondary side of the power transformer,
Nx - number of shorted turns.
Ix is the additional current component which is caused by the induced voltage in the short
circuited turns. This current component is not measured by the differential relay to circulate
through the shorted turns. The additional current component will be lagging the voltage by 90
degrees because the impedance of the shorted circuit is inductive. The Ampere-turn-balance
principle for the coils which are located on the same magnetic core has to be fulfilled,
irrespective of the existence of a turn-to-turn fault. In order to fulfill this requirement, two
measured currents will give a certain angle shift δ (Figure 1.3).
IxNx
IpNp
δ (Ns-Nx)Is
Figure 1.3 – The ampere-turn-balance principle
7
The component IxNx is seen by the differential relay as the differential current. When turn-to-
turn faults occur within the power transformer and only a few turns are shorted, then the
differential current magnitude can be very small and might not be within the tripping region.
According to the analysis of power transformer failures, 94 power transformers failed for the
period of 1997 through 2001 [3]. The total loss of 94 power transformers was $ 286,628,811 US
dollars. According to the Annual Report presented at 36th Annual Conference in Stockholm in
2003, the largest numbers of transformer failures occur in the Utility Substation sector and the
cause of power transformer failures is insulation failure. For all different types of power
transformer failures, insulation failure has the highest risk to fail [8]. The cause of developing
turn-to-turn faults is the degradation of the insulation system. The insulation system of the
transformer consists of paper and oil. The paper and oil can be subject to aging, which is defined
as the change in the properties of an electrical insulation system. Degradation is the process of
reducing insulation quality, which causes a breakdown in the insulation and leads to
development of turn-to-turn faults. The degradation of the insulation system can be due to the
mechanical, electrical, and thermal stresses, and moisture. When failure of the power
transformer happens, immediate detection of the failure is necessary in order to protect the entire
power system and minimize the damage and repair cost. If the turn-to-turn fault has not been
quickly detected and cleared, then this turn-to-turn fault can develop into a more serious and
costly to repair fault such as a phase-to-ground fault.
For detecting internal turn-to-turn faults in the power transformer three characteristics can be
used. These three characteristics are [9,10]:
1) Gas formation which is caused by the fault arc.
2) An increase in the phase currents.
3) An increase in the differential current.
When internal faults occur within the power transformer, immediate detection of these faults is
necessary in order to protect the entire power system and minimize the damage and repair cost.
Usually three types of protection can be used to protect the transformer from the internal faults:
8
1. Sudden pressure relay
The sudden pressure relay operates on a rate of rise of gas in the transformer [10] - [12]. This
relay does not operate on pressure changes or static pressure which results from the normal
operating condition of the transformer.
2. Over current relay

Over current relays are seldom used because of their vulnerability to false operation [9] - [12].
The false operation of over current relays might be caused by mismatch errors of current
transformers (CT‟s), saturation errors, and magnetizing inrush current when the transformer is
energized. The over current relay responds to the magnitude of the input current and operates
when this magnitude exceeds the pre-set level – pickup current. Then the over current relay
contacts will close to energize the circuit breaker trip coil. The contact of the relay will remain
open if the magnitude of the input current is less than the pickup current. Over current relays are
used as primary protection where differential protection is not used. If the differential protection
is used, then the over current relay is used as back up protection. Typically, fuses are used as
primary protection for transformers rated below 10 MVA. For transformers rated above 10
MVA, differential relays are used as primary protection and over current relays as back up.
3. The percentage differential relay

Differential relaying is one of the most effective and reliable methods of providing protection
against internal faults in power transformers. The percentage differential relay is depicted in
Figure 1.4.
Figure 1.4 – Percentage differential relay
9
Differential relaying treats the power transformer as a unit, taking measurements at all of the
transformer terminals. This method is very convenient because the power transformer terminals
are located at the same terminal. As it can been seen from the Figure 1.4, the differential
protection principle is based on comparing current magnitudes which enter (I1) and leave (I2) the
protected zone. The protected zone is the zone between two current transformers. Under normal
conditions, the current I1 would be equal to the current I2. Therefore, during normal system
operation and during external faults (faults outside of the protection zone) the differential relay
operating current is very small or near zero. If the fault occurs between the two ends of the
protected zone, the magnitudes of the current I1 and I2 are no longer equal. Then the presence of
the fault can be detected [12].
It can be observed from Figure 1.4 that the percentage differential relay has two types of coils:
the restraining coil (RC) and the operating coil (OC) [9]. The percentage differential relay
operates when the differential current between currents entering and leaving the protected zone
exceed a pre-determined percentage of the through current before tripping can occur. This
through current is referred to as the restraining current. Operation of the differential relay occurs
when the operating current exceeds the restraining current.
The major component of the differential protection scheme is the current transformer (CT). The
current transformer is connected in series with the power transformer windings with the
secondary currents circulating between them. The relay is connected across the midpoint where
the voltage is theoretically nil. Therefore, no current passes through the relay, and hence no
operation for faults outside the protected zone. Under internal fault conditions the relay operates
and both CT secondary currents add up and pass through the relay. The percentage differential
relay can be instantaneous in operation as it does not have to coordinate with any other relay on
the network.
In order to correctly apply the differential protection, two requirements have to be satisfied [13]:
1. The current magnitude differences on different sides of the protected transformer have to be
compensated by correct selection of interposing CT ratios. It is necessary to properly
compensate the phase shift between the windings by correct selections of interposing CT
10
winding connections. Some differential relays might internally accommodate the phase shift
of the transformer.
2. It is important to compensate for zero sequence current by using the interposing CT
connection, which makes it possible to remove zero sequence current.
Percentage differential relays must be tolerant of worst-case addition of the mismatch errors.
Percentage differential relays allow larger percentage mismatches up to 70% during heavy
through currents [9]. Percentage differential relays can be affected by a number of factors,
which are explained below:
Magnetizing inrush current

Magnetizing inrush current is the current which may flow when the transformer is first energized
[14] - [15]. The inrush current may reach the instantaneous peaks of 8 to 30 times that for full-
load current, and then decays rapidly during the first few cycles before decaying slowly. The
magnitude of the inrush current is affected by the following factors: type of the magnetic
material in the core, residual flux in the power transformer before switching on, and the size of
the power transformer. Since the differential relay might see the inrush current as an internal
fault, some methods should be present, such as a harmonic restraint, for distinguishing between
faults and inrush current [16] - [17]. The inrush current has a second harmonic. This second
harmonic can be used to restraint or block a relay during energization and avoid undesired
tripping. The second harmonic blocking can respond to magnitudes and phase angles of the
second harmonic and the fundamental frequency currents. The differential element can correctly
distinguish between the power transformer energization and faults.
Transformer tap-change operation

Transformer tap change operation makes a large contribution to current mismatch [15]. A
typical load tap changer range of 10% in voltage gives a 10% variation in current. When on load
tap-changer moves from one position to another, amplitude mismatch between the transformer
windings will result in a false differential current. This is a significant mismatch for which the
relay must not operate.
11
Current transformer (CT) saturation
Fault currents can lead to CT saturation [15]. Current transformer saturation might cause the
relay operating current to flow due to the distortion of the saturated current. Current transformer
saturation reduces the secondary output currents from the current transformer and causes a false
differential current to appear to the relay. To minimize problems due to the saturation, the CT‟s
ratios have to be selected properly.
Over excitation
Transformers are designed to operate below the flux saturation level [18]. An increase from the
maximum voltage level will lead to saturation of the core and an increase in the excitation
current drawn by the transformer. When a transformer core is over excited, the core will operate
in a non-linear magnetic region. This will create harmonic components in the exciting current.
There is a significant amount of current at the fifth harmonic that is characteristic of over
excitation.
30º phase angle shift introduced by transformer Y-∆ connection.

The connection of the power transformer windings produces a phase displacement from the
primary currents and voltages to the secondary currents and voltages.
Besides detecting the fault, percentage differential relay also has to be sensitive to all of the
above mentioned factors which can cause misoperation of the relay.
1.3 Literature review
Internal faults involve a magnitude of fault current which is low relative to the power
transformer base current. This indicates a need for high speed and high sensitivity to ensure
good protection. According to the IEEE Standard documents, there is no one standard way to
protect all power transformers against minor internal faults and at the same time to satisfy basic
protection requirements: sensitivity, selectivity, and speed [7].
12
The most difficult internal turn-to-turn fault is the fault which initially involves only a few turns.
The IEEE Standard C37.91-2000 indicates that as much as 10% of the transformer winding
might be shorted to cause a detectable change in the terminal current. Therefore, when fewer
numbers of turns are shorted, it will result in an undetectable amount of current. There is no
limit to the maximum internal fault that can flow, other than the protection system capability [7].
According to the IEEE Committee Report [7], the main and most commonly used protection
provided for internal faults for power transformers of approximately 10 MVA three-phase is the
percentage differential relay. Although the percentage differential relay is the most commonly
used protection, it is not thoroughly efficient for detecting minor internal turn-to-turn faults in
power transformers. It is difficult to detect minor internal turn-to-turn faults using the
percentage differential relay because the changes in the transformer terminal‟s current will be
quite small because the ratio of transformation between the whole winding and the short-
circuited turns is quite small. The traditional transformer differential percentage protection is not
sensitive enough to detect minor internal winding faults. Low-level turn-to-turn faults cannot be
detected with overall sensitivity represented by the restraint characteristic of the percentage
differential relay. For example, if the restraint characteristic of the percentage differential relay
has been set to 20% and a minor internal fault causes a differential current of 10%, thus an
internal fault cannot be detected until this fault evolves into a more severe fault with a higher
differential current. And for this reason, the conventional percentage differential protection is
not sensitive enough to determine low-level turn-to-turn faults.
Alternatively, minor turn-to-turn faults can be detected by the sudden pressure relay. However,
the sudden pressure relay is slow to operate unless the internal fault is severe. The sudden
pressure relay can detect internal faults with a delay of typically 50-100 ms that often causes the
fault to become more serious. Reference [6] gives a real case study on transformer failure due to
shorted turns. Failure of transformer by shorted turns is depicted in Figure 1.5. A turn-to-turn
fault was developed in a 30 year old power transformer as a result of severe winding conductor
insulation aging. The power transformer was tripped on sudden pressure relay. After the fault
was cleared, fault diagnosis showed extensive loss of conductors and conductor insulations in the
faulted phase, which was unlikely to be economically repairable. As it can be seen from this real
13
case study, the internal turn-to-turn fault was detected but with a time delay that caused severe
damage of the faulted winding.
Figure 1.5 – Failure of transformer by shorted turns (reproduce from reference [6] with
permission)
The previous methods which have been proposed towards improving the differential relay and
detecting internal turn-to-turn faults within the power transformer are outlined in the paragraphs
below.
Sachdev, Sidhu and Wood presented a digital relaying algorithm for detecting transformer
winding faults [19]. This principle is based on checking whether the differential equation
combined from the equations at the primary and secondary windings is valid. The electro-
magnetic equations of a transformer are differential equations of currents, voltages and mutual
flux linkages. The differential equations are valid during normal operating conditions,
magnetizing inrush and external faults. However, they are not valid during internal faults. If the
power transformer is connected in Y-Y, the fault can be detected directly because the winding
current is available. However, if the power transformer is connected in Y-∆, then each of the
three equations contains the unavailable ∆ winding current causing this algorithm to decompose
14
the winding currents into non-circulating and circulating currents. According to the results
presented in this paper, the proposed technique was effective for 5% of shorted turns of the
winding.
Kang, Lee and et al described a transformer protection based on the increment of the flux
linkages [20]. The ratio of the increments of the primary and the secondary winding flux
linkages is equal to the turn ratio during normal operating conditions, magnetizing inrush current
and over-excitation. During internal winding faults, it will differ from the turn‟s ratio.
However, the ratio of the increments of the flux linkages is not always equal to the turn ratio
because the increments of the flux linkages of the primary and secondary windings are
instantaneous values and consequently pass through zero. According to the presented result, the
proposed technique is effectively limited to a turn-to-turn fault involving more than 10% of the
winding.
Jiang, Bo and et al proposed power transformer protection based on transient detection using
Discrete Wavelet Transform (DWT) [21]. The protection technique in this paper focuses on the
features of fault transient currents instead of the fault under fundamental frequency. To detect
the transformer fault, only the dominant transient within the certain bands plays the important
role. The average and differential currents between the primary and the secondary currents are
derived from the DWT outputs. To obtain the restraining and operating current, the spectral
energies of the average and differential currents are calculated using the moving average process.
To determine whether the fault is internal or external to the transformer, the relay compares the
levels of the operating and restraining current.
Ngaopitakkul and Kunakorn presented an algorithm based on a combination of Discrete Wavelet

Transform and neural networks for detecting and classification of internal faults in two winding
three phase transformer [22]. According to this method, the current waveforms are extracted to
several scales with the Wavelet transform, and the coefficients of the first scale from the Wavelet
transformer are investigated. The comparison of the coefficients was performed and used as
inputs for the training process of the neural networks.
15
Khorashadi-Zadeh and Li presented a sensitive artificial neural network (ANN) based
differential relay for fault identification in power transformer protection [23]. In this proposed
algorithm, different transient states are considered as different patterns. These different patterns
are recognized by an artificial neural network. Inputs to an artificial neural network are the
harmonics of the positive sequence of differential current. In this method the inputs are the 1 st,
2nd, and 5th harmonic components of the positive sequence differential current. The drawback of
using methods based on neural networks is that these methods require a large number of training
patterns which are produced by simulation of various cases and this method is not generalized to
be applied to different power transformers.
In 2005, Gajic, Brncic, Hillstrom and et al from ABB (Sweden) in a conference paper discussed
about a new differential protection method based on negative sequence currents for detecting
turn-to-turn faults within the power transformer [24]. According to the new method, the relay
will look into the phase angle shift between negative sequence current components from
different sides of the transformer and will make decisions based on this phase angle shift.
According to reference [24], the new sensitive method can be used for detecting minor internal
turn-to-turn faults in the power transformer. Studies presented in this paper were very limited –
it dealt with one particular configuration and one system condition. The results given in this
paper were also not convincing because of limited studies performed for only one set of system
parameters and no detailed investigations.
Past researchers faced problems while using negative sequence currents for detecting minor turn-
to-turn faults within the power transformer. According to reference [25], the changes in
impedance of the total phase circuit for a shorted turn is very difficult to calculate because of the
power transformer actions. In accordance with their estimations when the impedance in the
faulted phase changes by 3%, the negative and zero sequence currents will not be sufficiently
large enough and will be less than 1%. Their rational is as follows: the magnitudes of the
currents largely depend on the total reactor impedance as the source impedance is relatively low
compared to the reactor impedance. Also, it is worth noting that removing the ground from the
unit will not affect the magnitudes of the positive and negative sequence currents significantly,
16
but it will eliminate the zero sequence currents because there will not be a path for the zero
sequence currents to flow through.
The literature review discussed in previous paragraphs showed that detecting minor internal turn-
to-turn faults within the power transformer is a very difficult problem. Some special schemes or
relays have to be designed and can be used as customized application.
The thesis focus work is on developing a negative sequence protection scheme for detecting
internal turn-to-turn faults within the power transformer for various operating conditions and
different configurations of the power transformer. The main motivation of this research was to
verify if this proposed method is able to perform better than the traditional differential protection
method.
1.4 Objectives of the research
As described above, the primary objective of this thesis is to investigate the possibility of using
the protection technique, which is based on negative sequence currents, for detecting minor turn-
to-turn faults in transformers in order to improve the performance of the traditional transformer
differential protection. The second objective is to verify how accurately internal turn-to-turn
faults can be detected within the power transformer using this new protection technique. The
final objective is to investigate the performance of the new protection method based on negative
sequence currents under different numbers of shorted turns of the transformer, different
connections of the power transformer, different values of the fault resistances, different values of
the system parameters, and during the CT saturation.
1.5 Outline of the thesis
This thesis is organized in six Chapters and two Appendices. A brief introduction to the power
transformers and a brief overview about existing transformer protection schemes, which are
17
currently used for protection of power transformers, is described in the first chapter. The
problem with detecting minor internal faults within the power transformer is stated. Objectives
of the research are drawn from the literature review on transformer protection against internal
turn-to-turn faults. The organization of the thesis is discussed as well.
In Chapter 2, a detailed description about symmetrical components and sequence networks for
various types of common faults and an internal turn-to-turn fault in power transformers is
presented.
The new protection method is presented in Chapter 3. Also, logic of the proposed technique
based on negative sequence currents is introduced in this chapter.
Chapter 4 discusses simulation of the power system. Furthermore, the detailed description of the
model of the proposed technique based on negative sequence currents and the traditional
differential protection model are presented in this chapter.
Chapter 5 shows the studies of the performance of the traditional differential protection and the
proposed technique based on negative sequence currents. The performance of the proposed
technique under different connections of the power transformer, different numbers of shorted
turns of the power transformer, different values of the fault resistances, different values of the
system parameters, during the CT saturation, and the inrush current is investigated, as well as the
performance of the traditional differential protection.
Chapter 6 summarizes the research work described in the thesis and presents general
conclusions.
Data and parameters of the modelled power system are presented in Appendix A.
Performance of the proposed method for the power transformer connected in Δ/Y is given in
Appendix B.
18
CAHPTER 2
UNSYMMETRICAL FAULTS
2.1 Symmetrical components
This research introduces the protection technique, which can improve differential relay
sensitivity for minor internal turn-to-turn faults. This technique is based on the theory of
symmetrical components, more exact, on negative sequence currents. According to Fortescue‟s
theorem, three unbalanced phasors of a three phase system can be resolved into three balanced
systems of phasors [26] - [29]. The unbalanced set of current phasors is depicted in Figure 2.1
Ic
Ia
Ib
Figure 2.1 – Unbalanced current phasors
The balanced sets of components are (Figures 2.2 - 2.4):
1. Positive sequence components

This sequence consists of three phasors which are equal in magnitude, displaced from each
other by 120 degrees in phase, and have the same phase sequence as the original Iabc.
Ic1
120 0
Ia1
Ib1
Figure 2.2 – Positive sequence components
19
2. Negative sequence components
This sequence consists of three phasors which are equal in magnitude, displaced from each
other by 120 degrees in phase, and have the phase sequence Iacb which is opposite to the
original phasors.
Ib2 1200 Ia2
Ic2
Figure 2.3 – Negative sequence components
3. Zero sequence components

This sequence consists of three phasors which are equal in magnitude and with zero phase
displacement from each other.
Ia0
Ib0
Ic0
Figure 2.4 – Zero sequence components
This method converts three unbalanced phases into three independent sources.
For example, a vector for three phase currents can be written as:
(2.1)
Where Ia - the current in phase A,

Ib - the current in phase B,
Ic - the current in phase C.
The three symmetrical components can be written as:
20
(2.2)
Where I 0 - the zero sequence component,

I 1 - the positive sequence component,
I 2 - the negative sequence component.
Phases are rotated forward by 120 degrees. A phase rotation operator a is defined to rotate a
phasor vector. Using the a operator, the double subscript notation which was used in Figures 2.2
- 2.4 can be eliminated. This can be done by expressing each phasor in terms of phase A phasor.
120 0
Figure 2.5 – Positive sequence components expressed in terms of phase A quantities
120 0
Figure 2.6 – Negative sequence components expressed in terms of phase A quantities
Ia0=I0
Ib0=I0
Ic0=I0
Figure 2.7 – Zero sequence components expressed in terms of phase A quantities
21
The symmetrical components have to satisfy the constraint which states that the vector sum of
symmetrical components equals the original set of unbalanced phasors [26]. The sum of the
sequence components is depicted in Figure 2.8.
I0
Ic Ia
I0
Ib
I0 I2
I1
Figure 2.8 – Sum of sequence components
The unbalanced currents can be expressed as the sum of their components:
(2.3)
(2.4)
(2.5)
A matrix A can be defined using this operator to transform the phase vector (ABC) into
symmetrical components (012).
(2.6)
The phase currents generated by the sequence components can be written as:
(2.7)
22
Conversely, the sequence components which are generated from the phase currents can be
written as:
= ∙ (2.8)
Where - the inverse matrix.
(2.9)
Then equation (2.8) can be written as:
· (2.10)
Then equation 2.10 can be written as:

(2.11)
(2.12)
(2.13)
In a three phase system, the sum of the line currents (Ia, Ib, Ic) is equal to the current IN in the
return path through the neutral [27] - [29]:
(2.14)
As it can be seen from Equation 2.11:

(2.15)
If there is no path through the neutral of a three phase system, then the current IN is zero and the
line currents contain no zero components.
23
2.2 Sequence impedances for the power transformer
When positive sequence currents are present, the impedance is called the positive sequence
impedance. When negative sequence currents are present, the impedance is called the negative
sequence impedance. And when only zero sequence currents are present, then the impedance is
called the zero sequence impedance [26] - [29].
For all power transformers, the positive, negative, and zero sequence impedances are the same.
(2.16)
Where Z1 - the positive sequence impedance,

Z2 - the negative sequence impedance,
Z0 - the zero sequence impedance,
Zl - the leakage impedance.
The sequence impedance of the power transformer is the total leakage impedance of the
transformer.
The primary and the secondary windings of the power transformer can be connected in either
delta (∆) or wye (Y) configurations. There are four possible combinations of connection: ∆-∆,
∆-Y, Y-∆, Y-Y.
When a system has a transformer with ∆-Y connection, the effect of the ∆-Y transformer phase
shift on fault currents has to be considered. The positive quantities on the high voltage side of
the transformer will be 30 degrees greater than the positive quantities on the low voltage side of
the transformer [27] - [29]. The negative sequence quantities are the reverse of the positive
sequence quantities. In other words, the negative sequence quantities on the high voltage side of
the transformer will lead the negative sequence quantities on the low voltage side of the
transformer by -30 degrees.
24
The equivalent circuit for the zero sequence impedance depends on the power transformer
winding connections. As it is known, zero sequence currents will flow only if there is a return
path through which a completed circuit is provided [26] - [29].
All possible combinations of the transformer winding connections are discussed in the following
paragraphs.
If the power transformer is connected in Y-Y and both neutrals are grounded, then there is a path
for zero sequence current to flow in both windings of the transformer. The equivalent zero
sequence circuit for this connection is depicted in Figure 2.9.
Z0
a Z a΄
g
g
Figure 2.9 – The equivalent zero sequence circuit for Y-Y connections with both neutrals
grounded
If the power transformer is connected in Y-Y with the primary neutral grounded and the
secondary neutral isolated, then the zero sequence current in the secondary winding is zero.
Consequently, the zero sequence current in the primary winding will be zero. The equivalent
zero sequence circuit for this connection is depicted in Figure 2.10.
Z0
a Z a΄
g
g
Figure 2.10 – The equivalent zero sequence circuit for Y-Y connections with the primary neutral
grounded and the secondary neutral isolated
25
If the power transformer is connected in ∆-∆, then the zero sequence current will circulate in the
∆ connected windings, but no current can leave the ∆ terminals. The equivalent zero sequence
circuit for this connection is depicted in Figure 2.11.
a Z
Z0 a΄
g
g
Figure 2.11 – The equivalent zero sequence circuit for ∆-∆connections
If the power transformer is connected in Y-∆ and neutral is grounded, then the primary currents
can flow because there is zero sequence circulating current in the secondary winding connected
in ∆ and a ground return path for the primary winding connected in Y. For these connections,
there is isolation between the primary and the secondary windings because zero sequence current
cannot leave the ∆ terminals. The equivalent zero sequence circuit for this connection is
depicted in Figure 2.12.
a Z0 a΄
Z
g
g
Figure 2.12 – The equivalent zero sequence circuit for Y-∆ connections with neutral grounded
If the power transformer is connected in Y-∆ and neutral is isolated, then the zero sequence
current cannot flow and the equivalent circuit reflects infinite impedance. The equivalent zero
sequence circuit for this connection is depicted in Figure 2.13.
26
a Z0 a΄
Z
g
g
Figure 2.13 – The equivalent zero sequence circuit for Y-∆ connections with neutral isolated
2.3 Sequence network connections and symmetrical components for various

types of common faults
Before the fault occurs, the system operates under steady-state conditions. The pre-fault system
is balanced and the zero, positive, and negative sequence networks are uncoupled [27] - [29].
The sequence components I0, I2 are equal to zero before the fault occurs. During unsymmetrical
faults, sequence networks carrying the currents I0, I1 and I2 are interconnected at the fault
location to present various unbalanced fault conditions. The presence of the negative and zero
sequence components is an indication of abnormal conditions (faults).
Since the new method is based on negative sequence currents, the negative sequence current for
different types of fault has to be observed.
Unbalanced faults such as phase-to-phase, phase-to-ground, and phase-to-phase-to-ground

produce negative sequence current. From the negative sequence current standpoint, phase-to-
phase fault is the most severe fault. The reason for this can be seen from the sequence network
diagrams.
The sequence network diagrams for three different fault conditions (phase-to-phase, phase-to-
ground, phase-to-phase-to-ground faults) are depicted in Figures 2.14 - 2.16.
27
1. Phase-to-ground fault
The negative sequence current resulting from phase-to-ground fault is limited by the positive
(Z1), negative (Z2), zero (Z0) sequence impedances and the fault impedance (Zf). In case of the
bolted fault, the fault impedance equals zero.
I0
Z0
V0
I1
3Zf
Z1
Vf
V1
I2
Z2
V2
Figure 2.14 – The sequence connection for phase-to-ground fault
The sequence components of the fault currents are:
(2.17)
Where Vf - the pre fault voltage.
The sequence fault currents can be transformed to the phase domain using equation 2.7.
2. Phase-to-phase fault
The negative sequence current which results from the phase-to-phase fault is limited only by the
positive (Z1), negative (Z2) sequence impedances, and the fault impedance (Zf).
28
I1 I2
Z1 Z2
Vf V1 V2
Figure 2.15 – The sequence connection for phase-to-phase fault
The sequence faults current are:
(2.18)
The negative sequence current resulting from the phase-to-phase fault will be higher than the
negative sequence current resulting from phase-to-ground fault.
3. Phase-to-phase-to-ground fault
For the phase-to-phase-to-ground fault, the sequence connection shows that the positive,
negative and zero sequences are connected in parallel at the fault terminal. This connection will
result in higher positive sequence current. However because the current split between the
negative and zero network sequences, the resulting negative sequence current is usually less than
that for the phase-to-phase fault.
I1
Z1 I2 3Zf I0
V1 V2 V0
Vf
Z2 Z0
Figure 2.16 – The sequence connection for phase-to-phase-to-ground fault
The positive sequence fault current is:
29
(2.19)
Using current division in Figure 2.16, the negative sequence fault current is:
(2.20)
The zero sequence fault current is:

(2.21)
2.4 Sequence network connections and symmetrical components for turn-to-

turn faults in power transformers
A turn to-turn fault is a short circuit of a few turns in the power transformer winding. When a
few turns are shorted in one phase, then the impedance of the shorted phase is not equal to the
impedances of the other two phases [25], [29]. According to reference [25], the impedance
change in one phase can be represented as a shunt unbalance as shown in Figure 2.17 for a
grounded transformer. Figure 2.18 depicts the corresponding sequence network for a turn-to-
turn fault.
Za Zb Zb
Zn
Figure 2.17 – Shunt unbalance with a shorted turn in phase A
30
Vf
Z1 Zb
1
Z2 Zb
1
Z0 Zb +3Zn
Figure 2.18 – Sequence network for a turn-to-turn fault in phase A
As it was mentioned before in Chapter 1, the traditional differential relay is not sensitive enough
to detect minor internal turn-to-turn faults in the power transformer because the changes in the
phase current are quite small.
The new protection technique for turn-to-turn fault is based on the negative sequence currents.
Negative sequence currents superimpose pure-fault quantities. Negative sequence currents have
the advantage over the zero sequence currents. It is known that removing the ground from the
unit will eliminate zero sequence components. However, negative sequence currents are
produced even when the fault does not include earth.
31
Figures 2.19 - 2.21 show phase currents (ISC1 and ISC2) and negative sequence currents (INS_P and
INS_S) on both sides of the transformer during the steady-state condition and during the fault when
5%, 3% and 1% of turns are shorted on the secondary side of the power transformer in phase C.
Figure 2.19 – Phase currents and negative sequence currents during the steady-state condition
and during the fault for 5% shorted turns
32
In the figures:
ISC1 - secondary phase current produced by CT on the primary side of TR in phase C,
ISC2 - secondary phase current produced by CT on the secondary side of TR in phase C,
INS_P - negative sequence current on the primary side of TR scaled down using CT,
INS_S - negative sequence current on the secondary side of TR scaled down using CT.
As it can be seen from Figures 2.19 - 2.21, the changes in magnitudes of the negative sequence
currents on both sides of the power transformer during the fault compared to the steady state
value of the phase currents are greater than the change in magnitude of the phase currents for a
turn-to-turn fault.
The changes in magnitudes of the phase currents and negative sequence currents during the
steady state condition and during an internal turn-to-turn fault when 5%, 3% and 1% of turns are
shorted on the secondary side of the power transformer are shown in Table 2.1 - 2.3.
33
Table 2.1 – Magnitudes of the phase currents and negative sequence currents during steady-state
condition and during an internal turn-to-turn fault for 5% shorted turns
Sequence components Steady-state, A Fault, A Change, %

ISC1 4.62 4.54 1.76
ISC2 4.62 5.29 12.6
INS_P 0.0014 0.26 99.46
INS_S 0.0014 0.57 99.75

ISC1 4.63 4.55 1.75
ISC2 4.63 5.11 9.39
INS_P 0.0014 0.24 99.41
INS_S 0.0014 0.50 99.72

ISC1 4.62 4.60 0.43
ISC21 4.62 4.85 4.75
INS_P 0.0014 0.23 99.39
INS_S 0.0014 0.42 99.66
34
2.5 Summary
This chapter introduces the concept of the symmetrical components technique. The symmetrical
components technique is very important in the analysis and design of a three-phase power
system. The symmetrical components method is a very important tool for unsymmetrical short-
circuit studies. A linear transformation from phase components (ABC) to symmetrical
components (012) is shown. The sequence networks and symmetrical components for different
types of common faults and an internal turn-to-turn fault in the power transformer are presented.
35
CHAPTER 3
SENSETIVE TURN-TO-TURN FAULT DETECTION USING
NEGATIVE SEQUENCE CURRENTS
3.1 Introduction to the new transformer protection method
The existence of negative sequence currents beyond the normal or tolerable unbalances is proof
of a fault in the power transformer windings. Negative sequence currents do not exist during the
symmetrical three phase fault. They do exist during unsymmetrical faults long enough for the
relay to make the proper decision. Negative sequence currents have an advantage compared to
zero sequence currents because they can provide coverage for phase-to-phase and turn-to-turn
faults as well as for ground faults.
The algorithm for the new turn-to-turn fault detection is based on negative sequence currents and
derived from the theory of symmetrical components explained in Chapter 2 [29].
It is known from the theory of symmetrical components that:

1. Negative sequence currents can distribute through the negative sequence network.
2. The source of negative sequence currents at the point of the fault is:
3. Negative sequence currents obey Kirchhoff‟s first law which states that at any point in an
electrical circuit, the sum of currents flowing towards that point is equal to the sum of
currents flowing away from that point.
The following sections describe flow of negative sequence currents for the power transformer
external and internal faults.
36
3.1.1 Flow of the negative sequence currents for transformer external faults
If the fault is external to the power transformer, then the source of the negative sequence fault
currents will be outside of the protected zone as shown in Figure 3.1. For the external faults in
the transformer, negative sequence current (INS_P) will enter the healthy transformer on the
faulty side and negative sequence current (INS_P ) will leave on the other side as shown in
Figure 3.1. Then, negative sequence currents on the respective sides of the power transformer
will have opposite direction. This means that the phase shift between these two phasors will be
equal to 180 º.
Wye Y/Y
Wye INSs2
ZNS_S1 INS_P
INSs1' INSs1 INS_S ZNS_S2
INS_P
ZNSs1 ZNSs2
INS_P
INSs1 E
Ens NS
INS_PINSs1
Negative sequence current

Differential Relay
relay logic
Figure 3.1 – Flow of negative sequence currents for transformer external fault
In Figures 3.1and 3.2:
ZNS_S1, ZNS_S2 - the negative sequence impedances for the equivalent sources S1 and S2,
ENS - fictitious negative sequence source,
INS_P, INS_S - the negative sequence currents on the primary and secondary side of TR,
INS_P - the negative sequence current transformed from the fault side to the
other side of TR,
INS_P, INS_S - the negative sequence currents on the primary and secondary side of TR
scaled down using CTs.
37
3.1.2 Flow of the negative sequence currents for transformer internal faults
If the fault is internal to the power transformer, then the source of the negative sequence fault
currents will be inside of the protected zone as shown in Figure 3.2. For the internal faults in the
transformer, negative sequence currents (INS_P and INS_S) will flow out of the faulty
transformer on both sides as shown in Figure 3.2. Then negative sequence currents on the
respective sides of the power transformer will have the same direction. This means that the
phase shift between these two phasors will be equal to 0 º.
Wye Y/Y
Wye
ZNS_S1 INS_P
INSs1 INS_S
INSs2
ZNSs1
ZZNSs2
NS_S2
INSs1 Ens INSs2

INS_P ENS INS_S
Negative sequence
Differential Relay
current relay logic
Figure 3.2 – Flow of negative sequence currents for transformer internal fault
3.2 The description of the new negative sequence current based protection
method
For this method to be successful, the negative sequence components on the two sides of the
protected power transformer have to be significant. The new protection technique operates
depending on the relative position of two phasors. In other words, this protection scheme will
compare the phase angle shift between the negative sequence current on the primary side of the
transformer and the negative sequence current on the secondary side.
However, the comparison between the negative sequence currents on the primary and secondary
sides of the power transformer is valid for non-zero phase displacement transformers and turns
ratio equal to 1. In other words, this comparison is valid for the ∆-∆ or Y-Y connection of the
38
transformer windings, and for the turn ratio equal to 1. If the phase displacement of the
transformer differs from zero degrees and the turn ratio differs from 1, then the phase shift and
turns ratio of the transformer have to be compensated in order to use this method. The ∆-Y
connection of three phase power transformers introduces a 30 degree phase shift between the
primary and the secondary windings. A 30 degree phase shift between the primary and the
secondary winding of the power transformer can be compensated by the correct selection of
interposing current transformers windings connections. The correct selection of interposing
current transformers ratio allows compensating for current magnitude differences on different
sides of the power transformer.
The logic of the protection method based on negative sequence currents for detecting an internal
turn-to-turn fault within the power transformer is shown in Figure 3.3.
Primary currents Secondary currents

ISA1, ISB1, ISC1 ISA2, ISB2, ISC2
FFT FFT
I I
Negative sequence current INS_P Negative sequence current INS_S

I I
I I
INS_P > Imin No Block No INS_P > Imin
Yes Yes
I No
I Phase shift = 0 I
Yes
Trip signal
I
Figure 3.3 – NegativeS sequence current based logic
39
As it can be seen from the Figure 3.3, the logic of the protection method based on negative
sequence currents has two steps:
1. Firstly it checks for the magnitude of negative sequence current components on both sides of
the power transformer.
2. The second step is to check for the phase angle shift between these two phasors.
The logical steps are explained in detail below:
1. All individual instantaneous currents on the primary side (ISA1, ISB1, ISC1) and on the
secondary side (ISA2, ISB2, ISC2) of the power transformer have to be measured.
2. Using the Fast Fourier Transform (FFT) block determines the fundamental harmonic
magnitude and phase of the input signal as a function of time.
3. Calculate negative sequence currents on the primary side (INS_P) and on the secondary side
(INS_S) of the power transformer (i.e. magnitude and phase components of the fundamental
harmonic).
4. Check the magnitudes of negative sequence currents from both sides of the transformer and
compares them with a pre-set level. The magnitudes of negative sequence currents from both
sides of the power transformer have to be above the pre-set limit in order to check the
relative position (phase angle) between these two phasors. The minimum pre-set level has to
be above values which can be measured during normal operation of the power system. The
limit value has to be set in the range within 1% to 20% of the differential protection‟s base
current. The differential protection‟s base current is the power transformer‟s high voltage
side rated current. The reason for this condition is to exclude the negative sequence currents
which can be introduced due to the pre-fault asymmetries of the power system (tolerable
imbalances).
40
If the contribution of negative sequence currents from both sides of the power transformer is
less than a pre-set level, then a directional comparison is not done to avoid the possibility of
making a wrong decision and the relaying logic will block.
But if the contribution of negative sequence currents from both sides of the transformer is
more than a pre-set level, then a directional comparison can be done and a phase angle
between two phasors of negative sequence currents can be found.
5. If the phase shift between two phasors of negative sequence currents is 0º, then an internal
fault is indicated and a trip command issued.
3.3 Summary
This chapter gives an overview on the new transformer differential protection method. The logic
of negative sequence currents based protection method is described. In order to apply this
method for detecting internal faults within the power transformer, the phase shift and turns ratio
have to be compensated. According to the new method, the differential relay will look into the
phase angle shift between negative sequence current components from different sides of the
power transformer and will make decisions depending on the relative position of two phasors.
41
CHAPTER 4
PSCAD/EMTDC MODELING OPF THE POWER
TRANSFORMER AND THE DIFFETENTIAL RELAY
4.1Tools required to model the power system and differential relays
The PSCAD/EMTDC ™ software (version 4.2.1) has been used to model the power system and
to design the traditional differential relay model and the differential relay based on negative
sequence currents models. PSCAD/EMTDC is the simulation tool for analyzing power system
transient developed by Manitoba HVDC Research Centre [31]. The PSCAD/EMTDC software
gives a complete library with the typical elements required to model the typical power system.
4.2The modeled power system
The system under study is a two machine power system connected through a three phase power
transformer. A three phase transformer bank was constructed using three single phase
transformers, each of 33.3 MVA, 23/132 kV. The simulated power system model is depicted in
Figure 4.1. To validate the model, the steady state operation and the transient behaviour during
faults were compared with theoretical calculations.
S1 CT1 CT2 S2
Relay
Figure 4.1 – Simulated power system model
42
The simulated power system model includes:
 Voltage sources (S1 and S2).
 Circuit breakers (BRK).
 Current transformers (CT1 and CT2).
 A power transformer model that can simulate internal turn-to-turn faults.
4.2.1 Voltage sources
Two sources (S1 and S2) have been used in simulation. The voltage source component model
which has been used to model a three phase voltage source is depicted in Figure 4.2.
RRL
Figure 4.2 – The voltage source component model
The source impedance type can be chosen as resistive, inductive or capacitive. The parameters
can be specified behind the source impedance or at the terminal. This component has an option
of external control of voltage and frequency. Monitoring the source can be controlled through
internal parameters or fixed parameters or variable external signals. The external inputs are:
 V - line-to-ground, peak voltage magnitude, kV,

 f - frequency, Hz.
The parameters of two sources are given in Tables A.1 - A.2 in Appendix A.
4.2.2 Circuit breakers
The circuit breaker component model is depicted in Figure 4.3. This component simulates the
operation of the three phase circuit breaker. This component has an option to specify the ON
43
(closed) and OFF (open) resistance of the breaker. The circuit breaker is controlled by the input
signal (default is BRK).
The logic of the circuit breaker is:
 OFF( Open) - 1
 ON (Closed) - 0.
BRK
Figure 4.3 – The circuit breaker component model
4.2.3 Current transformers
As it can be seen from Figure 4.1, differential relay is connected to the CTs secondary windings.
Current transformers (CTs) are used in the field to reduce the value of primary currents for
convenience of measurement to a maximum value of 5A and provide electrical isolation from the
power system. This allows personnel to work in a safer environment.
To select current transformers (CTs), rated (full-load) currents at both voltage levels have to be
calculated using the formula:
(4.1)
Where S - the maximum load,
V - the rated voltage.
The CT‟s ratios are chosen based on the values of the full-load currents. The parameters of
current transformers are given in Tables A.3 - A.4 in Appendix A. The current transformer
component model is depicted in Figure 4.4.
44
Figure 4.4 – The current transformer component model
Input parameters of the current transformer are primary and secondary turns, secondary
resistance and inductance, and burden resistance and burden inductance.
4.2.4 Power transformer model for simulating internal turn-to-turn faults
The distribution of the magnetic flux is fundamentally altered for the transformer with an internal
turn-to-turn fault. To produce the magnetic flux in the core, the exciting current is required. The
exciting current flows in the primary winding where it establishes an altering flux in the
magnetic circuit. When an internal turn-to-turn fault occurs, the situation is more complex. To
model a turn-to-turn fault, the corresponding winding is divided into three sub coils “a”, “b” and
“c” as it can be seen in Figure 4.5 [32] - [34].
Figure 4.5 depicts a single phase transformer with a turn-to-turn fault. The turn-to-turn fault
happened on the secondary winding, causing the secondary winding to be divided into three
parts:
1) Top part - sub coil a.

2) Shorted part - sub coil b.
3) The bottom part - sub coil c.
Primary winding Secondary winding
1 1
Figure 4.5 – A single phase transformer with a turn-to-turn fault on secondary winding
45
Essentially, the short-circuit impedance which is mainly the leakage inductance will totally
change for a mid-winding fault. A single phase transformer can be described by 4×4 matrix
representing [L].
(4.2)
Where L1 - self-inductance in the primary winding,

La, Lb, Lc - self-inductance of sub coils “a”, “b” and “c”,
M1a , M1b, M1c - mutual inductance between primary winding and sub coils,
Mab , Mac, Mbc - mutual inductance between sub coils.
The PSCAD/EMTDC package has a standard three phase power transformer component in the
library. But to detect and diagnose the internal turn-to-turn faults in transformers, a model of the
transformer with an internal turn-to-turn faults is required.
PSCAD/EMTDC software allows the designing of a new component. Components in

PSCAD/EMTDC are a graphic representation of the model which allows supplying input
parameters and performing pre-calculation on input data. A new component can be created in
PSCAD/EMTDC using the Component Wizard. The new component is represented by an icon in
graphical interface PSCAD/EMTDC, an internal code in FORTRAN and dialog boxes. The
function of the internal code is to process the inputs and retrieves outputs. The dialog boxes of the
PSCAD/EMTDC component allow setting values to parameters of the internal code of the new
component. The new component can interact with other components in PSCAD/EMTDC.
Figure 4.6 shows a new created component – a single phase power transformer with internal turn-
to-turn faults which has been used to model the power system and to simulate turn-to-turn faults.
46
A1 C1
A2 4 Wdg TrfC2
1 Phase
B1 D1
B2 D2
Figure 4.6 – A graphic representation of a single phase transformer with internal turn-to-turn
faults
Where A1-A2 - primary winding of the transformer,

B1-B2 - top part of the secondary winding of the transformer,
C1-C2 - shorted part of the secondary winding of the transformer,
D1-D2 - the bottom part of the secondary winding of the transformer.
Input parameters of the power transformer are:
1) System frequency.
2) Transformer MVA.
3) Nominal voltage of each sub coil.
4) Leakage reactance of each sub coil.
4.2.4.1 The calculation of the leakage reactance of the power transformer
The load current flows into the winding results in the magnetic field around the winding. This
field is the leakage flux field which exists in the spaces occupied by the windings and in the
spaces between windings. The leakage flux results in the impedance between windings. This
impedance is termed as leakage reactance.
The inductance of the winding can be defined as:
(4.3)
47
Where N - number of turns of the winding,
l - the mean length of the magnetic circuit,
μ0 - the permeability,
Ag - cross-sectional area of core.
The leakage reactance is:
(4.4)
Where f - the signal frequency,
L - the inductance.
The magnitude of the leakage reactance depends on the number of turns of the winding, the
leakage field, the current in the winding, and the geometry of the winding and core. The load
current through the leakage reactance results in voltage drop. Leakage reactance is termed as
percent leakage reactance. It is the ratio of the reactance voltage drop to the winding voltage
multiplied by 100%.
To calculate the leakage reactance, first the inductance has to be calculated. Inductances for a
single phase power transformer with the internal turn-to-turn fault were obtained using the Finite
Element Analysis [41] - [42]. The parameters of a single phase transformer are given in Table A.5
in Appendix A. The calculation of the leakage reactance of a single phase transformer for
different percentages of shorted turns on the primary and secondary windings is specified in
Tables A.6 - A.7 in Appendix A.
4.3 Modeling differential relays in PSCAD
To protect three phase power transformers against internal turn-to-turn faults, the traditional
transformer differential protection and the proposed method based on negative sequence currents
are used. Models of both protection methods are described below.
48
4.3.1 Traditional differential protection method
To detect internal turn-to-turn faults the traditional protection relay can be used. The three phase
differential relays have three single phase differential relays for each phase. Each differential
relay has a differential component which has 2 elements:
 The restraint windings. Currents in the restraint windings tend to prevent tripping.
 The operating winding.
4.3.1.1 The percentage differential slope
In the percentage differential relay, the differential current must exceed a fixed percentage of the
through current in the transformer. The through current is referred to as the restraint current and
is defined as the average of the primary and the secondary currents [9]:
(4.5)
Where K- a compensation factor, generally equals 0.5 or 1.
The relay will operate when the differential current (Id) is greater than the restraining current (Ir):
(4.6)
Where Ir - restraint current,

- the slope of the percentage differential characteristic is generally expressed as a
percent value: typically 10%, 20%, and 40%.
A percentage slope SLP of differential relay has been selected and adjusted to make the relay to
be insensitive to mismatch between CT currents and relay tap setting, differences in accuracy of
49
the CTs on either side of the transformer bank during faults, and to transformer tap changing
(TTC). The slope of the percentage differential relay determines the trip zone.
The procedure of selecting the percentage differential slope (SLP) is [25]:
1. To calculate the rated currents at both voltage levels.

2. To select current transformers. For increased sensitivity, current transformer ratios have to
be close to the rated current values as possible.
3. To calculate the current transformer secondary currents.
4. To calculate relay currents.
5. Select the relay taps that have a ratio as close as possible to the relay current ratio. Select
TL and TH.
6. To calculate the percentage of the current‟s mismatch to ensure that the relay taps selected
have an adequate safety margin.
(4.7)
Where IL, IH - relay input currents at the same kVA base for low and high voltage
sides respectively.
TL, TH - relay tap settings for low and high voltage sides respectively.
S - smaller of the two terms, or .
7. To calculate a difference in accuracy of the CT‟s on either side of the transformer bank.
To select the proper percentage slope for the relay characteristic when the information on
unequal CT errors is not available, then an appropriate assumption can be made. It is reasonable
to assume that the errors in the two CTs will not be more than 10% under all fault conditions.
Therefore, the percentage differential slope for the relay can be selected including about 5% of
safety margin.
50
4.3.2 Description of the negative sequence currents’ fault detection
component
The new differential protection method works if the protected power transformer is connected to
some load so that currents can flow on both sides of the protected power transformer. The
proposed method based on negative sequence currents‟ has the negative sequence currents fault
detection component.
The first step is to extract magnitudes of negative currents. To extract these magnitudes, the Fast
Fourier Transform (FFT) block can be used. The FFT can determine the harmonic magnitude
and phase of the input signal as a function of time. The FFT is a process of multiplying a signal
by a sinusoid in order to determine frequency components of a signal. First the input signal is
sampled before it is decomposed into harmonic constituents. The FFT provides the option of
using one, two or three inputs. In the case of three inputs, the FFT block can take a three phase
input and calculates the FFT output through a sequencer whose outputs are: positive (Mag+),
negative (Mag-), and zero (Mag0) sequence magnitude components and positive (Ph+), negative
(Ph-), and zero (Ph0) sequence phase components as shown in Figure 4.9.
1 1
INS_P 1
Mag+ Mag- Mag0
(7) (7) (7) 1
XA Ph+
Isa1 (7)
FFT
1
INS_P_ph
XB Ph-
Isb1 (7)
1
XC F = 60.0 [Hz] Ph0
Isc1 (7)
dcA dcB dcC
(a) Primary side FFT block
51
1 1
INS_S 1
Mag+ Mag- Mag0
(7) (7) (7) 1
XA Ph+
Isa2 (7)
FFT
1
INS_S_ph
XB Ph-
Isb2 (7)
1
XC F = 60.0 [Hz] Ph0
Isc2 (7)
dcA dcB dcC
(a) Secondary side FFT block

Figure 4.8 – Magnitudes and phase angles of sequence components produced by FFT blocks
Where ISA1, ISB1, ISC1 - secondary phase currents produced by CT‟s on the primary side of TR
in phase A, B, and C respectively,
ISA2, ISB2, ISC2 - secondary phase currents produced by CT‟s on the secondary side of TR
in phase A, B, and C respectively,
INS_P - magnitude of the negative sequence current on the primary side of TR
scaled down using CT,
INS_S - magnitude of the negative sequence current on the secondary side of TR
scaled down using CT,
INS_P_ph - phase angle of the negative sequence current on the primary side of TR,
INS_S_ph - phase angle of the negative sequence current on the secondary side of TR.
The second step is to check the magnitudes of the negative sequence currents from both sides of
the transformer. As it was mentioned in Chapter 3, the magnitudes of negative sequence currents
(INS_P and INS_S) on both sides of the power transformer have to be higher than a pre-set level.
The pre-set level was chosen to be 1%. The maximum current on the secondary side of the
current transformer is 5 A. Therefore, 1% is 0.05 A. Magnitudes of negative sequence currents
were extracted from the FFT blocks.
To compare magnitudes of negative sequence currents with a pre-set level, a comparator has
been used. This component compares two inputs. The output from the comparator is a level
output which is obtained when one signal is above the other signal. The comparison of negative
sequence currents with a pre-set level using the comparator is depicted in Figure 4.9.
52
A
INS_P D1
B Compar-
0.05 ator
A
INS_S D2
B Compar-
0.05 ator
Figure 4.9 – Comparison of negative sequence currents with a pre-set level
Second, if the contribution of negative sequence currents from both sides of the transformer is
less than a pre-set level, then a directional comparison is not done to avoid the possibility of
making a wrong decision.
But if the contribution of negative sequence currents from both sides of the power transformer is
more than a pre-set level, then a directional comparison can be done and a phase angle between
two phasors of negative sequence currents can be found.
If outputs (D1 and D2) from both comparators are 1, which means that both magnitudes of
negative sequence currents are greater than a pre-set level, then a phase angle between two
negative sequence currents can be checked. Phase angle comparison between two phasors of
negative sequence currents is shown in Figure 4.10.
53
A A A
INS_P_ph Ctrl = 1 D +- 0 Ctrl = 1
B F B Compar- B
1000.0 | X| 5.0 ator B1
Ctrl Ctrl
| X|
D1
Trip Signal
A
INS_S_ph Ctrl = 1
0.0 B
Ctrl
D2
Figure 4.10 – Phase angle comparison between two phasors of negative sequence currents
As it is shown in Figure 4.10, two 2 input selectors have been used in phase angle comparison.
The output from the selector will be A or B depending on the value of Ctrl. If the value of Ctrl is
1, then the output of the selector 1 will be INS_P_ph, and if the value of Ctrl of the selector 2 is 1,
then the output will be INS_S_ph. Then using the Summing/Differencing Junctions block, the
difference between these two angles can be found. If the difference is in the range between 0-5
degrees, then an internal fault is indicated and a trip command is issued.
4.4 Summary
This chapter gives a detailed description of the components of the modelled power system under
study. Also, models of the traditional differential protection and the proposed technique based
on negative sequence currents are presented in this chapter. The PSCAD/EMTDC software
(version 4.2.1) is used to model the power system and design the traditional differential relay
model and the differential relay based on negative sequence currents‟ models. The model of the
54
proposed relay based on negative sequence currents has the negative sequence currents‟ fault
detection component. The operation of the negative sequence currents‟ fault detection
component is described.
55
CHAPTER 5
TEST RESULTS
5.1 Introduction
This chapter presents the test results using the protection method based on negative sequence
currents and the traditional differential protection method. The purpose of these studies is to
detect minor internal turn-to-turn faults within the power transformer using both methods and
compare their performance for various cases.
5.2 Performance of the traditional differential protection
The purpose of this study is to investigate the performance of the traditional differential
protection for internal turn-to-turn faults. The transformer winding fault is the most difficult
fault to detect within the power transformer. The most difficult aspect of it is providing power
transformer protection when only a few turns are shorted. As it is known, the changes in
magnitude of the transformer‟s terminal currents are very small when a limited number of turns
are shorted within the power transformer. IEEE Standard indicates that at least 10% of the
transformer winding has to be shorted to cause detectable change in the terminal current.
Therefore, when fewer numbers of turns are shorted, it will result in an undetectable amount of
current [5].
The first step is to investigate how accurately the traditional differential protection can detect
minor internal turn-to-turn faults. The internal turn-to-turn fault occurred on the secondary
winding (HV side) in phase C of the power transformer as depicted in Figure 5.1. The time
taken to apply the fault is 3 seconds with duration of 0.4 seconds. In this study, the restraint
characteristic has been set to 20%.
56
CT1 CT2
S1 S2
Traditional
differential relay
Figure 5.1 – Internal turn-to-turn fault on the secondary winding of the power transformer
5.2.1 Differential current versus restraining current
The next sets of figures show the differential and restraining currents of the traditional differential
protection for an internal turn-to-turn fault for various percentages of shorted turns on the
secondary winding in phase C of the power transformer and trip signals from the relay.
Figures 5.2 - 5.4 show that the differential current (FdiffC) remains above the value of the
restraining current (FresC) which means that the traditional differential relay will operate for
internal turn-to-turn faults when a higher number of turns (for example 25%, 15%, and 10% of
turns) are shorted.
57
(a) Differential and restraining current
(b) Trip signal

Figure 5.2 – Traditional differential relay response when 25% of turns are shorted
(b) Trip signal

58
(b) Trip signal

Figures 5.5 - 5.7 show that the restraining current (FresC) remains above the value of the
differential (FdiffC) current which means that the traditional differential relay does not operate for
internal turn-to-turn faults when a small number of the turns are shorted (for example 5%, 3%,
and 1% of turns are shorted on the secondary winding of the power transformer). The restraint
characteristic has been set to 20% and a minor internal turn-to-turn fault when 5%, 3%, and 1%
of turns are shorted causes a differential current to be less than 20%. For this reason, the
traditional differential protection cannot detect a turn-to-turn fault for these three cases with
small number of turns shorted.
The results in Figures 5.5 - 5.7 are in agreement with the IEEE Standard which states that at least
10% of turns have to be shorted in order to detect internal turn-to-turn faults.
59
(b)Trip signal
(b) Trip signal

60
(b) Trip signal

The above results show that the transformer protection requires a new sensitive method to detect
minor internal turn-to-turn faults (especially when less than 10% of turns are shorted).
5.2.2 Primary and secondary phase currents during internal turn-to-turn faults
The traditional differential protection is not sensitive enough to detect minor internal turn-to-turn
faults within the power transformers because the changes in the transformer‟s terminal current are
quite small and the ratio of transformation between the whole winding and the short-circuited
turns is quite small.
The next sets of figures show phase currents on the primary side (Ia1, Ib1, Ic1) of the power
transformer and on the secondary side (Ia2, Ib2, Ic2) during the internal turn-to-turn fault on the
secondary winding in phase C for various percentages of shorted turns.
61
As it can be seen from Figures 5.8 - 5.10, the changes in the transformer terminal‟s current in
phase C due to the internal turn-to-turn fault when 25%, 15%, and 10% of turns are shorted on the
secondary winding, are sufficiently large enough. Because these changes are sufficiently large
enough, the traditional differential protection can detect the internal turn-to-turn fault for these
cases.
(a) Primary side of the power transformer
(b) Secondary side of the power transformer

Figure 5.8 – Phase currents on both sides when 25% of turns are shorted
62

Figure 5.9 – Phase currents on both sides when 15% turns are shorted
63

Figures 5.11 - 5.13 show that the changes in the transformer terminal‟s current due to the internal
turn-to-turn fault in phase C in the power transformer are quite small when less turns are shorted.
As it can be seen, the transformer terminal‟s current in phase C during the internal turn-to-turn
faults is almost the same as the current during normal operation. The results shown above
further demonstrate that it is difficult to detect minor internal turn-to-turn faults using the
traditional differential protection.
64

65

66

67
5.3 Performance of the proposed method for internal turn-to-turn faults
The purpose of these studies is to demonstrate the effectiveness of the protection technique,
which is based on negative sequence currents for detecting minor internal turn-to-turn faults in
power transformers.
In this section the results obtained from the negative sequence currents based protection method
are organized under the following subsections:
a. Different number of shorted turns of the power transformer.
b. Different values of the fault resistances.
c. Different values of the system parameters.
d. Different connections of the power transformer.
e. CT saturation.
f. Inrush current.
Performance of the proposed method based on negative sequence currents will also be compared
with the performance of the traditional differential protection method.
5.3.1 Turn-to-turn faults on the secondary winding of the power transformer

connected in Y-Y
In this study, an internal turn-to-turn fault occurred on the secondary winding (HV side) in phase
C of the power transformer connected in Y-Y. The time taken to apply the fault is 3 seconds
with duration of 0.4 seconds. The total number of the turns on the secondary winding is 866
turns.
The negative sequence currents are going to be present on both sides of the power transformer
with different magnitudes. Figures 5.14 - 5.19 show the magnitude of the negative sequence
current on the primary side (INS_P) and the magnitude of the negative sequence current on the
secondary side (INS_S) for different percentages of shorted turns. The magnitude of the negative
68
sequence current on the faulty side (INS_S) is greater than the magnitude of the negative sequence
current on the other side (INS_P) as expected.
Figure 5.14 – Negative sequence current magnitudes for 25% shorted turns on secondary (Y-Y)
69
70
Figures 5.20 - 5.25 show „zooming in‟ on the magnitudes of the two components of the total
negative sequence current for various percentages of shorted turns. Magnitudes of negative
sequence currents are compared with a pre-set level which is 1 % (0.05 A) of the differential
protection‟s base current.
Figure 5.20 – Zooming in on negative sequence current magnitudes for 25%

shorted turns on secondary (Y-Y)

71


72
Figure 5.25 – Zooming in on negative sequence currents magnitudes for 1%
Figures 5.26 - 5.31 show the relative phase angle between two phasors of negative sequence
currents during the internal turn-to-turn fault, which represent the respective contribution.
Theoretically, the phase angle between two phasors of negative sequence currents has to be 0
degrees, but in reality there are some other arbitrary phase angle shift caused by the quite high
current in the shorted turns. For the modelled power system, the phase angle shift between two
phasors of negative sequence currents is set from 0 to 5 degrees in order to ensure a very
sensitive turn-to-turn fault detection.
Figure 5.26 – Phase angle comparison between two negative sequence currents for 25%
shorted turns on the secondary (Y-Y)
73
74
Figures 5.32 - 5.37 show output signals from the proposed technique based on negative sequence
currents for different percentages of shorted turns on the secondary winding of the power
transformer. It can be seen from these figures that the proposed method can detect minor
internal turn-to-turn faults as low as 1% (8 turns) of shorted turns on the secondary winding in
phase C. A turn-to-turn fault for different percentages of shorted turns was detected within one
cycle (0.16 seconds).
75
Figure 5.32 – Trip signal from the proposed relay for 25% shorted turns on secondary (Y-Y)
Figure 5.33 –Trip signal from the proposed relay for 15% shorted turns on secondary (Y-Y)
76
secondary winding (Y-Y)
1. Effect of the fault resistance
The fault resistance is associated with the fault impedance path. The fault resistance may affect
the performance of the proposed technique based on negative sequence currents and result in
unreliable response of the relay. It should be noted that if a turn-to-turn fault involves a fault
resistance, then the sensitivity of the proposed technique is going to be reduced, but it should
also be noted that the fault resistance is going to have the same effect even on a traditional
differential protection.
77
Table 5.1 gives the values of the negative sequence currents on the primary and secondary
windings of the power transformer for a bolted turn-to-turn fault (Rf = 0Ω) which were obtained
from simulation. As it can be seen from Table 5.1, the proposed technique based on negative
sequence currents is sensitive in detecting a turn-to-turn fault for a bolted turn-to-turn fault for
all cases.
Table 5.1 – Turn-to-turn fault on the secondary winding of the power transformer connected in
Y- Y for various percentages of shored turns (Rf = 0Ω)
Shorted turns, INS_P, A INS_S, A Output signal

%
Trip action Trip time, ms

25 0.708 1.883 Trip 4
15 0.415 1.058 Trip 5.24
10 0.328 0.790 Trip 6.25
5 0.262 0.577 Trip 9
3 0.242 0.496 Trip 9.30
1 0.234 0.421 Trip 9.30
Faults are seldom solid and involve a varying amount of resistance. For the purpose of testing a
protective technique it is usually assumed that a small resistance is involved. Therefore, to make
the studies more realistic, a fault resistance Rf = 1Ω was included in the studies shown next.
Table 5.2 gives the measured values of the negative sequence currents on the primary and
secondary windings of the power transformer for the arcing turn-to-turn fault with the fault
resistance of 1 Ω. As it can be seen from Table 5.2, the sensitivity of the proposed technique is
reduced.
78
Table 5.2 – Turn-to-turn fault on the secondary winding of the power transformer connected in Y-Y
for various percentages of shored turns (Rf = 1Ω)
Output signal
Shorted turns,
INS_P, A INS_S, A
%
25 0.707 1.880 Trip 4.028

15 0.413 1.054 Trip 5.26
10 0.321 0.782 Trip 6.55
5 0.241 0.533 Trip 9.26
3 0.167 0.344 Trip 9.45
1 0.030 0.050 No Trip -
2. Effect of different system parameters
The sensitivity of the proposed technique is studied using different system parameters. A power
transformer has been installed in two different system configurations. The negative sequence
currents are going to be present on both sides of the power transformer with different magnitudes.
The magnitude of the negative sequence current on the faulty side is greater than the magnitude of
the negative sequence current on the other side. The magnitudes of the negative sequence currents
depend on the magnitudes of the negative sequence impedances of circuits on the respective sides of
the power transformer. It is known that when the magnitude of the impedance increases then the
magnitude of the current will decrease.
The results given in Table 5.3 show the performance of the proposed technique using different
system parameters on both sides of the power transformer. It can be seen from Table 5.3 that the
proposed technique is going to operate correctly even with different values of system parameters.
79
Table 5.3 – Results using different system parameter values (Y-Y transformer, turn-to-turn fault
applied on the secondary winding)
Output signal
Shorted Impedance Impedance INS_P, A INS_S, A
turns,% source 1, source 2,
Ω Ω Trip action Trip time, ms
25 1.6 17.95 0.290 2.51 Trip 6.24

3 52.9 0.558 1.99 Trip 4
15 1.6 17.95 0.177 1.365 Trip 10.25
3 52.9 0.328 1.130 Trip 6.25
10 1.6 17.95 0.146 1.007 Trip 10.25
3 52.9 0.257 0.846 Trip 9.25
5 1.6 17.95 0.131 0.728 Trip 10.26
3 52.9 0.208 0.627 Trip 9.25
3 1.6 17.95 0.125 0.631 Trip 10.27
3 52.9 0.191 0.542 Trip 9.24
1 1.6 17.95 0.125 0.541 Trip 10.27
3 52.9 0.181 0.466 Trip 9.26
5.3.2 Turn-to-turn faults on the primary winding of the power transformer

connected in Y-Y
In this study, an internal turn-to-turn fault was applied on the primary winding (LV side) in
phase C of the power transformer connected in Y-Y. There is no current transformer saturation.
The total number of turns on the primary winding is 150 turns.
The magnitude of the negative sequence current on the primary side (INS_P) and the magnitude of
the negative sequence current on the secondary side (INS_S) for different percentages of shorted
80
turns on the secondary winding in phase C of the power transformer are shown in Figures 5.38 -
5.42. Since, a turn-to-turn fault is applied on the primary winding, the magnitude of the negative
sequence current on the primary side (INS_P) is greater than the magnitude of the negative
sequence current on the secondary side (INS_S).
Figure 5.38 – Negative sequence current magnitudes for 25% shorted turns on primary (Y-Y)
81
82
„Zooming in‟ on magnitudes of the two components of the total negative sequence current for
different percentages of shorted turns is shown in Figures 5.43 - 5.47. Magnitudes of negative
sequence currents are compared with a preset level which is 1 % (0.05 A) of the differential
protection‟s base current.
Figure 5.43 – Zooming in on negative sequence current magnitudes for 25% shorted turns on
primary (Y-Y)
primary (Y-Y)
83
primary (Y-Y)
primary (Y-Y)
primary (Y-Y)
84
Figures 5.48 - 5.52 show the relative phase angle between two phasors of negative sequence
currents for various percentages of shorted turns on the primary winding.
Figure 5.48 – Phase angle comparison between two negative sequence currents for 25% shorted
turns on primary (Y-Y)
85
86
As it can be seen from Figures 5.44 - 5.48, the proposed protection technique can detect internal
turn-to-turn faults as low as 1% (2 turns) of shorted turns on the primary winding of the power
transformer.
primary winding (Y-Y)
1. Effect of the fault resistance
secondary windings of the power transformer for a bolted turn-to-turn fault on the primary
winding with the fault resistance Rf = 0Ω. As it can be seen from Table 5.4, the proposed
protection technique detects a bolted turn-to-turn fault on the primary winding for all cases.
Table 5.4 – Turn-to-turn fault on the primary winding of the power transformer connected in Y-Y
for various percentages of shored turns(Rf = 0Ω)

%
25 1.74 0.67 Trip 4.30

15 1.01 0.39 Trip 9.2
10 0.76 0.268 Trip 10.45
3 0.545 0.218 Trip 10.49
1 0.492 0.157 Trip 11.77
87
secondary windings of the power transformer for a turn-to-turn fault on the primary winding with
the fault resistance of 1 Ohm.
Table 5.5 – Turn-to-turn fault on the primary winding of the power transformer connected in Y-Y
for various percentages of shored turns (Rf = 1Ω)

%
25 0.92 0.35 Trip 8.52

15 0.368 0.143 Trip 10.54
10 0.172 0.06 Trip 17.76
3 0.01 0.0075 No Trip -
1 0.0031 0.00101 No Trip -
As it can be seen from Table 5.5, the sensitivity of the proposed protection technique is reduced.
2. Effect of different system parameters
The results given in Table 5.6 show the performance of the proposed technique for detecting a
turn-to-turn fault on the primary winding of the power transformer using two sets of different
system parameters on both sides of the power transformer. The results presented in Table 5.6
demonstrate the correct operation of the proposed technique even with two different sets of
system parameters.
88
Table 5.6 – Results using different system parameter values (Y-Y transformer, turn-to-turn fault
applied on the primary winding)
Output signal
Shorted Impedance Impedance INS_P, A INS_S, A
turns,% source 1, source 2,
Ω Ω Trip action Trip time, ms
25 1.6 17.95 0.836 0.433 Trip 9.46

3 52.9 0.716 0.524 Trip 8.51
15 1.6 17.95 0.924 0.487 Trip 10.51
3 52.9 0.779 0.577 Trip 4.25
10 1.6 17.95 0.71 0.324 Trip 10.53
3 52.9 0.60 0.398 Trip 5.49
3 1.6 17.95 0.478 0.259 Trip 10.75
3 52.9 0.411 0.308 Trip 9.50
1 1.6 17.95 0.447 0.2085 Trip 11.73
3 52.9 0.3689 0.271 Trip 9.54
The performance of the proposed method for the power transformer connected in ∆-Y is given in
Appendix B.
5.4 Performance of the proposed method for external faults
The purpose of these studies is to investigate the response of the proposed method based on
negative sequence currents to external faults. As it is know from the differential principle, the
differential relay should not operate for external faults. The proposed method based was tested
for the external phase B-to-ground fault and the external phase B-C fault.
89
5.4.1. External phase B-to-ground fault on the high voltage side of the power
transformer
The external phase B-to-ground fault was applied on the secondary (HV) side of the power
transformer. The fault is applied at 3 seconds and the duration of the fault is 0.4 seconds.
Figure 5.53 shows phase currents on the primary side (Ia1, Ib1, Ic1) and on the secondary side (Ia2,
Ib2, Ic2) of the power transformer during the external phase B-to-ground fault. As it can be seen
from Figure 5.53, the changes in the transformer terminal‟s current in phase B indicate the
existence of the fault.
90
Figure 5.53 – Phase currents on both sides of the power transformer for external phase B-to-
ground fault on HV side
Figure 5.54 shows magnitudes of the two components of the total negative sequence current.
As it can be seen from Figure 5.54, magnitudes of negative sequence currents were compared
with a preset level which is 1% of the differential protection‟s base current.
Figure 5.54 – Negative sequence current magnitudes for external phase B-to-ground fault
on HV side
91
Figure 5.55 shows the relative phase angle between two phasors of negative sequence currents
which represent the respective contribution during the external phase B-to-ground fault.
Figure 5.55 – Phase angle between two phasors of negative sequence currents for external
phase B-to-ground fault on HV side
Figure 5.56 shows that the proposed technique is going to make a correct decision (i.e. it is not
going to trip for the external phase B-to-ground fault).
Figure 5.56 – Output signal from the proposed technique for external phase B-to-ground
fault on HV side
5.4.2 External phase B-C fault on the low voltage side of the power
transformer
The results obtained for the external phase B-C fault on the primary (LV) side of the power
transformer are shown in Figures 5.57 - 5.60. The time to apply is 3 seconds and the duration of
92
the fault is 0.4 seconds. Figure 5.57 shows phase currents on both sides of the transformer
during external phase B-C fault on LV side. As it can be seen from Figure 5.57, the changes in
the transformer terminal‟s current in phase B and C indicate the presence of the fault.
(b) Primary side of the power transformer

Figure 5.57 – Phase currents on both sides of the power transformer for external phase B-C fault
on LV side
93
Figure 5.58 shows magnitudes of the two components of the total negative sequence current.
Figure 5.58 – Negative sequence current magnitudes for external phase B-C fault on LV side
which represent the respective contribution during the external phase B-C fault.
Figure 5.59 – Phase angle between two phasors of negative sequence currents for external phase
B-C fault on LV side
Figure 5.60 shows that the proposed technique is again going to operate correctly (i.e. it is not
going to trip for the external phase B-C fault).
94
Figure 5.60 – Output signal from the proposed technique for external phase B-C fault on
LV side
As it can be seen from Figure 5.54 and Figure 5.58 the negative sequence current enters the
healthy transformer on the faulty side and leaves on the other side properly transformed. In other
words, the magnitude of the negative sequence current on the faulty side of the transformer
equals the magnitude of the negative sequence current on the healthy side of the transformer.
Also, as shown in Figure 5.55 and Figure 5.59, at any point in time after the fault is applied the
phase angle between these two phasors equals 180 degrees.
As shown in Figures 5.56 and Figure 5.60 the proposed method based on negative sequence
currents shows correct operation for the simulated external faults.
5.5 Performance of the proposed method for small unbalances in the power
system
Practically, systems rarely have perfectly balanced loads, currents, voltages, or impedances in all
three phases. A small unbalance is a common occurrence in a three phase system. When a small
unbalance occurs in the three phase system, small negative sequence currents are going to be
present. Phase unbalance can be simulated by considering different impedances in the three
phases. Table 5.7 gives phase currents due to a small unbalance.
95
Table 5.7 – Phase currents due to a small unbalance
Phase Load Amps
A 4.1759
B 4.5976
C 4.3906
According to reference [35] - [36], it is safe for electrical equipment to operate if the maximum
line current of one of the three phases does not exceed 10% above the average line current of the
three phases.
The average line current is defined as:
Line current unbalance LCUR (%) ratio is defined as the maximum line current deviation from
the average line current magnitude upon the average line current magnitude. For the modelled
power system a small line current unbalance ratio equals to 4.78%.
Figure 5.61 shows the negative sequence currents on both sides of the power transformer during
the normal operation which are due to the small unbalance in the power system.
Figure 5.61 – Negative sequence current magnitudes during a small unbalance in the power
system
96
Figure 5.62 shows the relative phase angle between two phasors of the negative sequence
currents during a small unbalance in the power system.
Figure 5.62 – Phase angle between two phasors of negative sequence currents during a small
unbalance in the power system
As it can be seen from Figure 5.61, the negative sequence currents during a small unbalance are
greater than the minimum pre-set level, but the phase angle between these two phasors equals
180 degrees at any point of time during (Figure 5.62). And for this reason, the proposed
technique is going to make a correct decision and is not going to operate for a small unbalance in
the power system.
Figure 5.63 shows that the proposed technique trip signal is zero (i.e. it is not going to trip for a
small unbalance in the power system).
Figure 5.63 – Output signal from the proposed technique for a small unbalance in
the power system
97
The above verifications were important to study the reliability of the technique and showed that
the proposed relay is going to be stable for small unbalances in the power system.
5.6 Performance of proposed method during the CT saturation
The purpose of this study is to investigate the performance of the proposed method during the
CT saturation. As it is known when the current transformer is saturated, it can cause trip delays
[37] - [39].
5.6.1 CT saturation during turn-to-turn faults
The main focus of this thesis was studying minor internal turn-to-turn faults in power
transformers. It is known that current transformers can saturate while carrying fault currents.
But during minor internal turn-to-turn faults, the fault current is much less than the fault current
during internal unbalanced faults such as phase-to-phase, phase-to-ground, and phase-to-phase-
to-ground faults. So for this reason, the CT saturation might not to be a problem during minor
turn-to-turn faults (1%, 3%, 5% or 10% turn-to-turn faults).
The CT saturation might affect the performance of negative sequence currents based protection
method during other miscellaneous internal faults. The proposed negative sequence current
based protection method has significant potential for detecting other types of faults such as
phase-to-phase, phase-to-ground, and phase-to-phase-to-ground faults where CT saturation can
be a problem. So, one type of internal fault is shown here to study the impact of the CT
saturation on the negative sequence logic. Further detailed investigations can be done as part of
a future work on the thesis.
98
5.6.2 CT saturation during other miscellaneous internal faults
The impact of the current transformer saturation on the performance of the proposed method is
studied for the internal B-C fault on the secondary side of the power transformer connected in Y-
Y.
Two current transformers were selected on both sides of the power transformer to model the
power system. The current transformer on the primary side of the power transformer has a rating
of 1600 to 5 amperes and the current transformer on the secondary side of the power transformer
has a rating of 300 to 5 amperes. From the excitation curve, the knee point for the CT on the
primary side is over 320 volts and is over 60 volts for the CT on the secondary side of the power
transformer. If the secondary voltage of the CT is above the knee point then the CT will saturate.
According to the relaying current application guide, the current flowing on the secondary
winding of the current transformer is proportional to the voltage in the secondary side of the
current transformer and the burden which is connected to the secondary terminal of the current
transformer [40]. The burden impedance is typically expressed in values of less than 1Ω. The
burden impedance connected to the secondary terminals of each current transformer without
saturating their cores is 0.5 Ω. To saturate the power transformer, the burden impedance
connected to the secondary terminals of the current transformer has to be larger than the
maximum impedance. The burden impedance that was chosen to saturate the current
transformers was 7 and 10 Ω.
It is known that if the CT is saturated then the output signal from the current transformer
becomes distorted and the actual value of the current is lower than the value calculated by the
protective device. This can cause the relay to have a trip delay.
The secondary currents produced by CT‟s replicate the primary currents in the case of non-
saturated CT. When the CT is saturated then the secondary currents have a distorted waveform.
Figure 5.64 shows the secondary currents produced by CT on the primary side (Isa1, Isb1, Isc1) and
on the secondary side (Isa2, Isb2, Isc2) of the power transformer for non-saturated and saturated
CT‟s.
99
(a) Non-saturated CT
(b) Saturated CT (burden impedance of 7 Ω)
100
(c) Saturated CT (burden impedance of 10 Ω)
Figure 5.64 – Secondary currents produced by CT for non-saturated and saturated CT cases
The effect of the CT saturation on the magnitudes of the two components of the total negative
sequence current is shown in Figure 5.65.
101
(a) Saturated CT (burden impedance of 10 Ω)

Figure 5.65 – Negative sequence currents magnitudes for internal phase B-C fault for
non-saturated and saturated CT cases
for the cases of non-saturated and saturated CT.
102

Figure 5.66 – Phase angle between two phasors of negative sequence currents for phase B-
C fault for non-saturated and saturated CT cases
103
As it can be seen from Figure 5.66, the CT saturation causes the measured relative phase angle
between two phasors of negative sequence currents to be slightly different compared to the non-
saturated case. This is a factor that needs to be taken into account while determining the
tolerable phase shift between the two phasors. For the non-saturated case, the phase shift is 3
degrees whereas for the saturated case with burden impedance of 7 Ohm, the phase shift is 12
degrees and for the saturated case with burden impedance of 10 Ohm, the phase shift is 23
degrees.
Figure 5.67 shows the output signal from the technique based on negative sequence currents for
phase A-B fault for the cases of non-saturated and saturated CT.

Figure 5.67 – Output signal from the proposed technique for phase B-C fault for non-saturated
and saturated CT cases
104
As it can be seen from Figure 5.67, the CT saturation is going to result in trip delays similar to
the traditional differential relays.
5.7 Performance of the proposed method during inrush current
As it is known, the inrush current is caused by energizing the power transformer. Normally, the
magnitude of the energization current is about 1-2% of the rated current under steady state operating
conditions. When the power transformer is energized, the magnitude of the energization current can
become ten times higher than the full-rated current. This transient process causes the traditional
differential relay to trip.
The purpose of this study is to investigate the effect of the inrush current on the performance of
negative sequence currents based protection method. To see the performance of the proposed
technique during the inrush current, the inrush current phenomenon was recreated. An inrush
current can occur when sudden energization of the power transformer happens. The sudden
energization of the power transformer can happen when circuit breakers on both sides of the power
transformer are opened for a short period of time and then closed causing the energization of the
power transformer. Circuit breakers on both sides of the power transformer were opened at 2.0 sec.
and were closed at 2.05sec. Figure 5.68 shows magnitudes of the two components of the total
negative sequence current when the inrush current is present.
Figure 5.68 – Negative sequence currents magnitudes during inrush current phenomenon
105
Figure 5.69 shows the relative phase angle between two phasors of negative sequence currents,
which represent the respective contribution when the inrush current is present.
Figure 5.69 – Phase angle between two phasors of negative sequence currents during
inrush current phenomenon
As it can be seen from Figure 5.69, the phase angle between these two phasors equals 180
degrees during the inrush current. And for this reason, the proposed technique is going to
operate correctly (i.e. it is not going to operate during the inrush current phenomenon).
Figure 5.70 shows that the proposed technique is not going to trip for during the inrush current.
Figure 5.70 – Output signal from the proposed technique during the inrush current
5.8 Summary
Performance studies using the traditional differential protection method and the proposed method
are presented in this chapter. Firstly, the traditional differential protection is used to investigate
106
how accurately internal turn-to-turn faults can be detected. Also, the sensitivity of the traditional
differential protection is shown in this chapter. Secondly, the effectiveness of the negative
sequence currents based protection method was demonstrated. The performance of the proposed
method was investigated for various cases including different numbers of shorted turns of the
transformer, different value of the fault resistances, different values of the system parameters,
different connections of the power transformer, during the CT saturation and the inrush current.
The proposed method was also tested during small imbalances in the power system and for
different external faults of the power transformer.
Simulation results indicate that:
1. The traditional differential protection lacks sensitivity to detect minor internal turn-to-turn
faults which involve less than 10% of turns. The traditional differential protection method is
not sensitive to minor internal turn-to-turn faults because the changes in the transformer
terminal‟s current are quite small and the ratio of transformation between the whole winding
and the short-circuited turns is quite small.
2. The proposed method is very sensitive in detecting minor internal turn-to-turn faults and can
detect turn-to-turn faults even when 1% of turns are shorted.
3. The proposed method shows the correct operation for different system parameters and
different connections of the power transformer.
4. The proposed method can correctly distinguish between external faults, small imbalances in
the power system and internal faults.
5. The performance of the proposed method reveals that this relay can operate correctly if
current transformers are saturated.
The results show that the proposed method can provide a fast and sensitive approach for
identifying minor internal turn-to-turn faults in power transformers. Performance analysis
shows that this relay has advantages over the traditional differential relay.
107
6. SUMMARY AND CONCLUSIONS
6.1 Summary
The power transformers can sometimes fail due to different factors. Common among many of
the power transformer failure modes is a shorted turn. It is very important to identify minor
internal turn-to-turn faults before they develop into more serious and costly to repair faults. In
order to detect these types of faults, a highly sensitive and fast method is required. This thesis
described a new negative sequence current based protection method which was capable of
detecting minor internal turn-to-turn faults (even faults involving 1% of turns).
The problem associated with detecting minor internal turn-to-turn faults within the power
transformer such as low sensitivity of the existed protection methods was described in Chapter 1.
The previous methods such as methods based on electromagnetic equations of a transformer, on
increments of the flux linkage, on transient detection using DWT, on combination of DWT and
ANN for detecting internal turn-to-turn faults were discussed as well. Chapter 2 provided a
detailed description about symmetrical components and sequence networks for various types of
faults and the transformer‟s internal turn-to-turn faults. Chapter 3 gave a detailed description
about the new protection method based on negative sequence currents. Negative Sequence
currents exist during unsymmetrical faults long enough for the relay to make the proper decision.
The advantage of using negative sequence currents compared to zero sequence currents is that
they can provide coverage for phase-to-phase and turn-to-turn faults as well as for ground faults.
Chapter 4 gave the complete description of the modeled power system, the traditional differential
protection model and the detailed description of the model of negative sequence currents based
protection method. The performance of the traditional differential protection and the proposed
protection method for various case studies was presented in Chapter 5. It was shown that the
traditional differential protection lacks sensitivity whereas the proposed method is very sensitive
to minor internal turn-to-turn faults. The performance of the proposed technique was verified
using different numbers of shorted turns of the transformer, different values of the fault
108
resistances, different values of the system parameters, different connections of the power
transformer, during the CT saturation, and the inrush current.
6.2 Conclusion
This research work demonstrated that it is possible to develop a customized application for
detecting internal turn-to-turn faults using negative sequence currents. The performance of the
traditional differential protection was studied for different percentages of shorted turns. It was
found that the traditional differential protection lacks sensitivity to detect winding faults which
involve less than 10% of turns. The proposed protection scheme was evaluated with different
cases, which included different numbers of shorted turns of the transformer, different values of
the fault resistances, different values of the system parameters, different connections of the
power transformer, during the CT saturation, and the inrush current. The results illustrated that
the proposed method is able to detect minor internal turn-to-turn faults which involve only 2
turns (1% of turns) and a trip signal was activated within 1 cycle. The proposed method
discriminated accurately between external and internal faults. In addition, the proposed method
operated correctly under different system parameters, different connections of the power
transformer, during the CT saturation, and the inrush current.
6.3 Contribution
Specifically the thesis made the contributions which are described in the next paragraph.
A power transformer model for simulating internal turn-to-turn faults was developed using
PSCAD/EMTDC. The change in impedance of the total phase is difficult to calculate when the
fault involves few shorted turns. The values of the leakage reactance are not available in the
literature and are very difficult to find. Using Finite Element Analysis, the leakage reactance of
109
the power transformer for various percentages of shorted turns on the primary and secondary
windings was obtained.
The previous literature stated that the negative sequence current based method is not going to be
more sensitive than the phase current differential protection [25]. It also has been claimed that
the negative sequence current scheme is suited for careful customized application. However, it
was shown via an extensive simulation studies in the thesis that the negative sequence current
scheme is going to be much more sensitive than the traditional differential protection and can
detect even when 1% of turns are shorted. Also, the proposed scheme did not require any extra
customization more than the traditional differential protection scheme.
6.4 Future work
Results presented in this thesis showed that the proposed protection method which has higher
sensitivity for winding turn-to-turn faults may be implemented as an additional logic on top of
the traditional differential protection. In order to protect power transformers with the new
differential protection method and replace the traditional differential protection, the performance
of the proposed method has to be investigated under all factors which affect the traditional
differential protection. As future work:
 The performance of the proposed method can to be tested for transformer on-load tap-
changing operation and over excitation conditions.
 The simulation results obtained from PSCAD/EMTDC simulation package can be
validated using the Real Time Digital Simulator (RTDSTM) which will allow the closed-
loop testing of protection similar to an actual system. This will help in strengthening the
research findings.
110
REFERENCES
[1] International Standard IEC 60076-1, “Power Transformers-General”, Edition 2.1,

2000.
[2] J.H. Harlow, “Electrical Power Transformer Engineering”, CRC Press LLC, 2004.
[3] W. H. Bartley, “Analysis of Transformer Failures,” International Association of

Engineering Insurers, 36th Annual Conference, Stockholm, Sweden, 2003.
[4] P.M. Anderson, “Power System Protection”, John Wiley & Sons, Inc., 1999.
[5] A. R. Van and C. Warrington, “Protective Relays: Their Theory and Practice”,
London: Chapman and Hall, 1962.
[6] H. Ding, R. Heywood, J. Lapworth, and S. Ryder, “Why Transformers Fail”, Euro
TechCon 2009, Stretton, United Kingdom, 2009.
[7] IEEE Std C37.91-2000, “IEEE Guide for Protective Relay Applications to Power
Transformers”, 2000.
[8] W. H. Bartley, “An International Analysis of Transformer Failures, Part 1, The

Locomotive, Winter 2004, Vol. 78, No. 1, 2004.
[9] S.H. Horowitz and A.G. Phadke, “Power System Relaying”, John Wiley and Sons
Inc., 1992.
[10] IEEE Std C37.108-2002, “IEEE Guide for the Protection of Network Transformers”,
2002.
[11] G. Rockefeller, “Transformer Protection Application Guide”, Basler Electric, 1999.
111
[12] J. L. Blackburn and T. J. Domin, “Protective Relaying: Principles and Applications“,
CRC Press Taylor and Francis Group, 2007.
[13] Relay Standards Subcommittee, “Terms Used By Power System Protection

Engineers”, July 31, 1997.
[14] S. J. Chapman, “Electric Machinery Fundamentals”, Third Edition, WCB/McGraw

Hill, 1999.
[15] C. Christopoulos and A. Wright, “Electrical Power System Protection”, 2nd Edition,
Kluwer Academic Publishers, 1999.
[16] A. Guzman, S. Zocholl, G. Benmouyal, and H. J. Altuve, “Performance Analysis

of Traditional and Improved Transformer Differential Protective Relays”, SEL
Technical Papers.
[17] K. Behrendt, N. Fischer, and C. Labuschagne “Considerations for Using Harmonic

Blocking and Harmonic Restraint Techniques on Transformer Differential Relays”,
SEL Technical Papers.
[18] E. A. Klingshirn, H. R. Moore, and E. C. Wentz, “Detection of Faults in Power

Transformers,” AIEE Transactions, Vol. 76, Part III, April 1957, pp. 87 - 95.
[19] M.S. Sachdev, T.S. Sidhu, H.C. Wood, “A Digital Relaying Algorithm for Detecting
Transformer Winding Faults”, IEEE Transactions on Power Delivery, vol.4, No.3,
July 1989.
[20] Y.C. Kang, B.E. Lee, S.H. Kang, and P.A. Crossley, “Transformer Protection Based
on Increment of Flux Linkages”, IEE Proc.-Gener. Transm. Distrib., Vol.151, No.4,
July 2004.
[21] F. Jiang , Z.Q. Bo, M. A. Redfern, Z. Chen, “Power Transformer Protection Based on
Transient Detection using Wavelet Transform (DWT)”, Power Engineering Society
Winter Meeting, Vol.2-(27-31Jan.2002), 2002.
112
[22] A. Ngaopitakkul and A. Kunakorn, “Internal Fault Classification in Transformer
Windings using Combination of Discrete Wavelet Transformer and Back-propagation
Neural Networks”, International Journal of Control, Automation, and System, Vol.4,
No.3, pp.365-371, June 2006.
[23] H. KHORASHADI-ZADEH and Z. LI, “A sensitive ANN Based Differential Relay

for Transformer Protection with Security against CT Saturation and Tap Changer
operation”, Turk J Elec Engin, Vol.15, No. 3, 2007.
[24] Z. Gajic, I. Brncic, B. Hillstrom, I. Ivankovic, “Sensitive Turn-To-Turn Fault

Protection for Power Transformers”, CIGRE, Study Committee B5 Colloquium,
Calgary, Canada, 2005.
[25] W.A. Elmore, “Protective Relaying. Theory and Applications”, Second Edition,
Marcel Dekker, Inc., 2004.
[26] R.E. Ferh, “A Novel Approach for Understanding Symmetrical Components and
Sequence Networks of Three-Phase Power Systems”,
http://www.scribd.com/doc/6075396/Understanding-Negative-Sequence, accessed
on November 14, 2009.
[27] J. D. Glover and M. Sarma, “Power System Analysis and Design”, Second Edition,
PWS PUBLISHING COMPANY, 1994.
[28] H. Saadat, “Power System Analysis”, Second Edition, McGraw-Hill Companies, Inc.,
2002.
[29] C. F. Wagner and R.D. Evans, “Symmetrical Components”, Reprint Edition, Robert
E. Krieger Publishing Company, 1982.
[30] ANSI/IEEE C37.91-1985, IEEE Guide for Protective Relay Applications to Power
Transformers, 1985.
[31] Introduction to PSCAD/EMTDC V3, Manitoba HVDC Research Centre Inc.,

Canada, 2000.
113
[32] P. Bastard, P. Bertrand, and M. Meunier, “A Transformer Model for Winding Faults
Studies”, IEEE Transactions on Power Delivery, Vol.9, No 2, April 1994.
[33] H. Wang, K. Butler, "Modeling Transformers with Internal Winding Faults by

Calculating Leakage Factors," Proceedings of North American Power Symposium,
October 1999, San Luis Obispo, CA.
[34] G. Gonzalez, J. Fernandez and P. Arboleya, “Electromagnetic Model of Turn-to-Turn

Short Circuits in Transformers”, The International Journal for Computation and
Mathematics in Electrical and Electronic Engineering, Vol.23, No2. 2004.
[35] “Unbalanced Currents”, Technical Publication,

http://www.motorsanddrives.com/cowern/motorterms13.html, accessed on April15,
2010.
[36] Application Note, “Is it the drive, the motor, or the load?”,
http://www.fluke.com/download/library/1989052_w.pdf, accessed on April15, 2010.
[37] Lj. A. Kojivic, “Impact of Current Transformer Saturation on Over current Protection
Operation”, IEEE Power Engineering Society Summer Meeting (Cat.
No.02CH37376), 2002.
[38] V. Muthukrishnan, L. Sevov, I. Voloh, “Enhanced Over current Algorithm for

Performance Under CT Saturation in Industrial Applications”, 62nd Annual
Conference for Protective Relay Engineers, March 30 -April 2, 2009, Austin, TX.
[39] Westinghouse Electric Corporation, “Relaying Current Transformer Application

Guide,” Industrial and Commercial Power System Applications Series, July 1986.
[40] IEEE Std C37.110−1996, IEEE Guide for the Application of Current Transformers
Used for Protective Relaying Purposes, 1996.
[41] M.N.O. Sadiku, “A Simple Introduction to Finite Element Analysis of

Electromagnetic Problems,” IEEE Transactions on Education, vol.32, no. 2, pp. 85-
93, May1989.
[42] Volume 2 Flux 10 User’s Guide – Physical Description, Circuit Coupling, Kinematic
Coupling, Cedrat France, 2009.
114
APPENDIX A
DATA AND PARAMETERS OF THE MODELLED POWER
SYSTEM
Table A.1 – Parameters of source 1

Impedance
Base MVA Voltage, kV Phase angle, Internal phase angle, Frequency,
degrees Impedance, Ω degrees Hz
100 39.83 0 1.6 80 60
Table A.2 – Parameters of source 2

Impedance
Base MVA Voltage, kV Phase angle, Internal phase angle, Frequency,
degrees Impedance, Ω degrees Hz
100 229.99 20 52.9 80 60
Table A.3 – Parameters of CT on primary side

Primary turns Secondary Secondary Secondary Frequency, Hz
turns resistance, Ohm inductance, H
1 320 0.5 0.8e-3 60
Table A.4 – Parameters of CT on secondary side

Primary turns Secondary Secondary Secondary Frequency, Hz
turns resistance, Ohm inductance, H
1 60 0.5 0.8e-3 60
115
Table A.5 – Parameters of a single phase power transformer
Base Primary Secondary System Turns on Turns on Magnetizing
MVA voltage, kV voltage, frequency, Hz LV side HV side current, %
kV
33.3 23 132.79 60 150 866 0.1
Table A.6 – Leakage reactance of a single phase transformer for internal turn-to-turn fault on the
secondary winding of the power transformer
Leakage reactance,%
25% turns 15% turns 10% turns 5% turns 3% turns 1% turns

shorted shorted shorted shorted shorted shorted
LV-HV 15.08 15.08 15.08 15.08 15.08 15.08
LV - HVtop 121.82 156.87 178.49 204.85 218.7 232.6
LV - HVshorted 66.4 109.88 141.92 184.82 207.73 230.64
LV - HVbottom 62.74 32.97 23.71 17.68 14.82 0.1196
HVtop - HVshorted 131.9 106.58 92.09 71.13 65.15 59
HVtop - HVbottom 278.49 252.92 243.46 236.47 234.505 232.54
HVshorted - HVbottom 162.34 181.61 194.02 210.83 221.195 231.56
116
Table A.7 – Leakage reactance of a single phase transformer for internal turn-to-turn fault on the
primary winding of the power transformer
Leakage reactance,%
25% turns 15% turns 10% turns 3% turns 1% turns

shorted shorted shorted shorted shorted
HV-LV 15.08 15.08 15.08 15.08 15.08
LVtop - HV 128.48 164.42 200.36 219.09 247.69
LVshorted - HV 69.55 113.14 156.73 204.40 234.097
LVbottom - HV 71.38 36.97 25.6 16.87 7
LVtop - LVshorted 147.45 113.11 78.77 46.88 22.352
LVtop - LVbottom 308.20 277.56 246.92 244.63 217.512
LVshorted - LVbottom 178.16 196.40 214.64 222.14 237.056
117
APPENDIX B
PERFORMANCE OF THE PROPOSED METHOD FOR Δ-Y
TRANSFORMER
In this study, the internal turn-to-turn fault occurred on the secondary winding (HV side) in
phase C of the power transformer connected in Δ-Y.
Figures B.1 - B.3 show the magnitude of the negative sequence current on the primary side
(INS_P) and the magnitude of the negative sequence current on the secondary side (INS_S) for 5%,
3%, and 1% of shorted turns. Both magnitudes are compared with a preset level which is 1 %
(0.05 A) of the differential protection‟s base current.
Figure B.1 – Negative sequence current magnitudes for 5% shorted turns on secondary (Δ-Y)
118
Figures B.4 - B.6 show the relative phase angle between two phasors of negative sequence
currents during an internal turn-to-turn fault, which represent the respective contribution. As it
can be seen from Figures B4 - B6, ∆-Y connection of three phase power transformer introduces a
30º phase shift between the primary and the secondary windings.
119
Figure B.4 – Phase angle comparison between two negative sequence currents for 5% shorted
turns on secondary (Δ-Y) with 30º phase shift
Figure B.5 – Phase angle comparison between two negative sequence currents for 3%
shorted turns on secondary (Δ-Y) with 30º phase shift
120
shorted turns on secondary (Δ-Y) with 30º phase shift
However, the comparison between the negative sequence currents on the primary and secondary
sides of the power transformer is valid for non-zero phase displacement transformers. If the
phase displacement of the transformer differs from zero degrees then the phase shift has to be
compensated in order to use this protection method. A 30º phase shift between the primary and
the secondary winding of the power transformer is compensated and shown in Figures B7 - B9.
Figure B.7 – Phase angle comparison between two negative sequence currents for 5% shorted
turns on secondary (Δ-Y) without 30º phase shift
121
shorted turns on secondary (Δ-Y) without 30º phase shift
shorted turns on secondary (Δ-Y) without 30º phase shift
When a 30º phase shift between the primary and the secondary winding of the power transformer
is compensated, the trip signal from the proposed relay is activated within 1 cycle.
122

Turn-To-Turn Fault Detection in Transformers Using Negative Sequence Currents

Uploaded by

Copyright:

Available Formats

Turn-To-Turn Fault Detection in Transformers Using Negative Sequence Currents

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Turn-To-Turn Fault Detection in Transformers Using Negative Sequence Currents

Uploaded by

Copyright:

Available Formats

TURN-TO-TURN FAULT DETECTION IN

TRANSFORMERS USING NEGATIVE SEQUENCE

© Copyright Mariya Babiy, September 2010. All Rights Reserved.

Head of the Department of Electrical Engineering,

I also wish to express my sincere thanks to Dr. M. Ramamoorty, Former Director-General,

Also, I am very grateful to one of my colleagues, Mohamed El-Samahy, my one-of-a kind

LIST OF TABLES……………………………………………….……………...………... xvi

LIST OF SYMBOLS AND ABBREVITIONS …………………………………………. xviii

1.1 Introduction to the power transformer……………………………………………...... 1

1.1.1 Principles of the power transformer…………………………..…………...….... 2

1.1.2 Types of transformer failures…………………………………………………... 5

1.2 Transformer internal turn-to-turn faults………………………………………….….. 6

1.3 Literature review………….…………………………………………..…………….... 12

1.4 Objectives of the research ……………………………………………………….…... 17

1.5 Outline of the thesis………………………………………………………………...... 17

2.1 Symmetrical components……………………………...………………………….….. 19

2.2 Sequence impedances for the power transformer…………………………………...... 24

CHAPTER 3 SENSETIVE TURN-TO-TURN FAULT DETECTION USING

3.1 Introduction to the new transformer protection method…………….……………...... 36

CHAPTER 4 PSCAD/EMTDC MODELNG OF THE POWER TRANSFORMER

4.1 Tools required to model the power system and relays……………………………...... 42

4.2 The modeled power system……………………………………………….……....…. 42

4.2.1 Voltage sources…………………………………………………………..…….. 43

4.2.2 Circuit breakers ….….…………………………………………………..……... 43

4.2.3 Current transformers……………………………………………………..…….. 44

4.3 Modeling differential relays in PSCAD/EMTDC™………………………………..... 48

4.3.1Traditional differential protection method…………………………..………….. 49

4.3.1.1 The percentage differential slope……………………………..……….... 49

CHAPTER 5 TEST RESULTS …………………………………………………….……. 56

5.2 Performance of the traditional differential protection……………………………..…. 56

5.2.1 Differential current versus restraining current…………………………….....… 57

5.3 Performance of the proposed method for internal turn-to-turn faults……………...... 68

5.3.1 Turn-to-turn faults on the secondary winding of the power transformer

secondary winding (Y-Y)………………………………………..…….. 77

5.3.2 Turn-to-turn faults on the primary winding of the power transformer

primary winding (Y-Y)……………………………….…………….….. 87

5.4 Performance of the proposed method for external faults…………………………….. 89

5.6 Performance of the proposed method during the CT saturation……………………... 98

5.6.1 CT saturation during turn-to-turn faults……………………………...………... 98

5.6.2 CT saturation during other miscellaneous internal faults…………………….... 99

5.7 Performance of the proposed method during inrush current…….………………….... 105

5.8 Summary…………………………………………………………………..…...…….. 106

CHAPTER 6 SUMMARY AND CONCLUSIONS……………………………………... 108

6.1 Summary……………………………………………………………………….…….. 108

6.2 Conclusion……………………………………………………………………….….... 109

6.3 Contribution……………………………………………………………………….…. 109

6.4 Future work……………………………………………………………………….….. 110

APPENDIX A. DATA AND PARAMETERS OF THE MODELED POWER

APPENDIX B. PERFORMANCE OF THE PROPOSED METHOD FOR Δ-Y

IEEE The Institute of Electrical and Electronics Engineers

ENS Fictitious negative sequence source

RTDS ™ 2 Real Time Digital Simulator

1.1 Introduction to the power transformer

1.1.1 Principles of the power transformer

The operation of a power transformer is based on two principles:

Figure 1.1 – The ideal transformer

Where Ns - the number of turns in the secondary coil,

Where Np - the number of turns in the primary coil,

Then the induced voltage is: