Protection, Monitoring and Control of Shunt Reactors

546
Protection, Monitoring and Control

of Shunt Reactors
Working Group
B5.37
August 2013
Protection, Monitoring and Control
of Shunt Reactors
Working Group
B5.37
August 2013
Members
Stefan Roxenborg (Convenor) SE, Pouria Naisani (CA), Gheorghe Moraru (RO),
Stein Ingebrigtsen (NO), Javier Martin Herrera (ES), Simon Chano (CA)
Corresponding Members
Anita Oommen (ZA), Kevin Stephan (US), Antonio Carlos da Rocha Duarte (BR), Loi Lei Lai (UK),
Mikio Shintani (JP), Ji-Feng Wen (CN), Bapuji S Palki (IN)
Copyright © 2013
“Ownership of a CIGRE publication, whether in paper form or on electronic support only infers right
of use for personal purposes. Are prohibited, except if explicitly agreed by CIGRE, total or partial
reproduction of the publication for use other than personal and transfer to a third party; hence circu-
lation on any intranet or other company network is forbidden”.
Disclaimer notice
“CIGRE gives no warranty or assurance about the contents of this publication, nor does it accept
any responsibility, as to the accuracy or exhaustiveness of the information. All implied warranties
and conditions are excluded to the maximum extent permitted by law”.
ISBN: 978-2-85873-242-5
TABLE OF CONTENTS
1 Introduction ............................................................................................................... 12
2 Description of shunt reactors .................................................................................... 13
2.1 Design of shunt reactors.................................................................................... 13
2.2 Electrical characteristics ................................................................................... 18
2.2.1 Air-gap of the shunt reactor core .................................................................. 18
2.2.2 Inrush ............................................................................................................ 18
2.2.2.1 Switching in air core reactors ............................................................... 18
2.2.2.2 Switching in oil immersed iron core reactors ....................................... 22
2.2.3 Shunt reactor disconnection .......................................................................... 27
2.2.4 Harmonics ..................................................................................................... 28
2.2.4.1 Zero Harmonic (DC) ............................................................................. 31
2.2.4.2 2nd Harmonic ......................................................................................... 31
2.2.4.3 3rd Harmonic ......................................................................................... 32
2.2.5 Hysteresis ...................................................................................................... 33
2.2.6 Losses............................................................................................................ 33
3 Application of shunt reactors .................................................................................... 34
3.1 Connection to the Power System and Grounding Methods .............................. 34
3.1.1 Line and Bus connected reactors .................................................................. 34
3.1.2 Tertiary winding connected reactors ............................................................. 35
3.2 Effects of Shunt Reactors on Transmission Line Voltage ................................ 36
4 Shunt reactor faults and abnormal conditions ........................................................... 40
4.1 Fault types in Dry-type reactors ........................................................................ 40
4.2 Fault types in oil immersed reactors ................................................................. 40
4.3 Failure rates of shunt reactors ........................................................................... 41
4.4 Turn to turn faults ............................................................................................. 41
4.5 Bushing failure .................................................................................................. 43
5 Shunt reactor protections .......................................................................................... 44
5.1 Protection for bus connected oil immersed shunt reactors ............................... 45
5.1.1 Differential protection (87R / ∆I) ................................................................. 46
5.1.1.1 Low impedance phase differential protection – basic principles .......... 46
5.1.1.2 High impedance phase differential protection – basic principles ......... 48
5.1.1.3 Application notes differential protection .............................................. 50
5.1.2 Restricted Earth Fault protection (87N / ∆I) ................................................. 54
5.1.2.1 Low impedance Restricted Earth Fault protection- basic principles .... 54
5.1.2.2 High impedance Restricted Earth Fault protection- basic principles.... 56
5.1.2.3 Application notes REF protection ......................................................... 57
5.1.3 Phase overcurrent protection (50 and 51 / I>> and I>, t) .............................. 58
5.1.3.1 Application notes phase overcurrent protection ................................... 59
5.1.4 Earth fault overcurrent protection (51N or 51G / IE>, t) ............................... 60
5.1.4.1 Application notes – Earth fault overcurrent protection......................... 61
5.1.5 Distance protection (21 / Z<) ........................................................................ 62
5.1.5.1 Application notes distance protection ................................................... 62
5.1.6 Protection schemes dedicated to the detection of turn to turn faults ............ 66
3
5.1.6.1 Earth fault overcurrent protection controlled by directional zero
sequence relay ....................................................................................................... 66
5.1.6.2 Earth fault overcurrent controlled by directional negative sequence relay
67
5.1.6.3 Split phase protection............................................................................ 68
5.1.7 Tank protection ............................................................................................. 69
5.1.7.1 Application notes tank protection ......................................................... 70
5.1.8 Breaker failure protection (50BF) ................................................................. 70
5.1.8.1 Application notes breaker failure protection (BFP) .............................. 71
5.1.9 Pole discrepancy protection .......................................................................... 71
5.1.10 Buchholz relay (63)................................................................................... 72
5.1.11 Sudden pressure relay (63)........................................................................ 73
5.1.11.1 Application notes Buchholz and sudden pressure relay ................... 74
5.1.12 Typical shunt reactor protection schemes ................................................. 75
5.2 Protection for oil immersed line connected shunt reactors ............................... 78
5.2.1 Shunt reactor protection issues related to disconnection of transmission lines
80
5.2.1.1 Undesired tripping of ground overcurrent protection on de-energized
lines 80
5.2.1.2 Blocking of protections applied to line connected shunt reactors ........ 81
5.2.2 Protection for neutral (“fourth leg”) reactor ................................................. 82
5.3 Protection for transformer tertiary connected shunt reactors ............................ 83
5.3.1 Phase overcurrent protection (50 and 51 / I>> and I>, t) .............................. 84
5.3.1.1 Application notes - overcurrent protection for air core shunt reactors . 84
5.3.2 Negative sequence protection (46 / I2>)........................................................ 85
5.3.2.1 Application notes – negative sequence protection ................................ 85
5.3.3 Ground overvoltage protection (59N / 3U0>) ............................................... 85
5.3.3.1 Application notes – Ground overvoltage protection ............................. 87
5.3.4 Special schemes for turn to turn faults in air core shunt reactors ................. 87
5.3.4.1 Turn to turn fault detection – Split phase scheme................................. 88
5.3.4.2 Turn to turn fault detection - Voltage unbalance scheme ..................... 89
5.3.4.3 Application notes - Voltage unbalance scheme .................................... 90
5.3.5 Phase differential protection (87 / ∆I) for tertiary connected shunt reactors 90
5.3.6 Tertiary connected shunt reactors - Typical protection schemes .................. 91
6 Monitoring ................................................................................................................ 95
6.1 Thermal Overload ............................................................................................. 95
6.2 Oil temperature protection ................................................................................ 97
6.3 Winding temperature protection ....................................................................... 97
6.4 Oil level / flow monitoring ............................................................................... 98
6.4.1 Magnetic oil level gauge ............................................................................... 98
6.4.2 Bushing oil level indicator ............................................................................ 99
6.4.3 Flow indicators.............................................................................................. 99
6.5 Pressure Relief Valve ........................................................................................ 99
6.6 Fire protection ................................................................................................... 99
6.6.1 Advantages and disadvantages of pulverized water systems ...................... 102
6.6.1.1 Advantages.......................................................................................... 103
4
6.6.1.2 Disadvantages ..................................................................................... 103
6.6.2 Advantages and disadvantages of the nitrogen systems ............................. 103
6.6.2.1 Advantages.......................................................................................... 103
6.6.2.2 Disadvantages: .................................................................................... 103
6.7 Integrity of Insulating Oil ............................................................................... 103
6.7.1 Signature Analysis ...................................................................................... 104
6.7.2 Dissolved Gas Analysis .............................................................................. 104
6.7.2.1 Dissolved Gases in Oil ........................................................................ 104
6.7.2.2 Degradation of Oil-Impregnated Cellulose ......................................... 104
6.7.2.3 Degradation of Dielectric Oil.............................................................. 104
6.7.3 Early Detection on Oil-Filled shunt reactors .............................................. 104
6.7.4 Dissolved Gas Analysis .............................................................................. 105
6.7.5 Incipient Failure Condition Detection ........................................................ 105
6.7.6 Key Gases and Dissolved Gas Indices ........................................................ 105
6.8 Partial discharge measurements ...................................................................... 106
6.8.1 Acoustic method for PD detection .............................................................. 107
6.8.2 Monitoring Shunt reactor Bushings ............................................................ 107
6.9 Overvoltage measurements ............................................................................ 107
6.10 Parameters monitored ..................................................................................... 108
6.11 Monitoring systems ......................................................................................... 113
7 New solutions offered by numerical relays ............................................................ 115
7.1 Adaptive DC biasing ....................................................................................... 115
7.2 Negative-Sequence Internal-External fault discriminator .............................. 117
7.3 Open CT supervision ...................................................................................... 120
7.4 New approach to reactor turn-to-turn protection ............................................ 120
8 Protection implementation and setpoint recommendations .................................... 125
8.1 PROTECTION SCHEMES ............................................................................ 125
8.1.1 Reactor differential Protection .................................................................... 126
8.1.2 Impedance (Distance) relay ........................................................................ 126
8.1.3 Phase Overcurrent Protection ..................................................................... 127
8.1.4 Negative Sequence Overcurrent Protection ................................................ 128
8.1.5 Ground fault Protection............................................................................... 128
8.1.5.1 Restricted earth fault relay (REF) ....................................................... 128
8.1.5.2 Ground overcurrent relay .................................................................... 129
8.1.5.3 Ground overvoltage relay (Neutral Displacement) ............................. 129
8.1.6 Dedicated Turn-to-Turn Fault Protection ................................................... 130
8.1.6.1 Torque-controlled Earth Fault Overcurrent Protection ....................... 130
8.1.6.2 Spilt Winding Protection..................................................................... 131
8.1.6.3 Voltage Unbalance Scheme ................................................................ 131
8.1.7 Mechanical type fault detectors (Buchholz, Sudden Pressure, Pressure relief
devices) ................................................................................................................... 131
8.1.8 Top Oil and Winding Over-Temperature Protection .................................. 132
8.1.9 Breaker Failure/Pole Disagreement ............................................................ 133
8.2 Reactor Configurations and Protection Schemes ............................................ 133
8.2.1 Typical Main-1 protection schemes ............................................................ 134
8.2.2 Typical Main-2 and back-up protection schemes ....................................... 134
5
9 Control .................................................................................................................... 136
9.1 Manual switching ............................................................................................ 136
9.2 Automatic switching ....................................................................................... 136
9.3 Point on Wave Controller ............................................................................... 138
9.3.1 General statement........................................................................................ 138
9.3.2 Control of opening operations .................................................................... 140
9.3.3 Control of closing operations ...................................................................... 142
9.3.4 Single pole and three pole operation ........................................................... 143
9.4 New trends ...................................................................................................... 145
9.4.1 Adaptive functions ...................................................................................... 145
9.4.2 Impact of substation configuration on adaptive control arrangement ........ 146
10 Questionnaire on Existing Practices of Shunt Reactors Protection, Monitoring and
Control ............................................................................................................................ 149
10.1 Introduction ..................................................................................................... 149
10.2 Application and design of Shunt Reactor ....................................................... 150
10.2.1 Question: In what voltage levels the Shunt Reactors are being used? .... 150
10.2.2 Question: What is the type of connection to the power system? (E.g. direct
connection to the line, bus or transformer tertiary)................................................. 151
10.2.3 Question: What grounding methods are being used? (e.g. direct grounding,
reactance grounding...) ............................................................................................ 152
10.2.4 Question: How are your Shunt Reactors designed? (e.g. Oil or Dry, 3 leg
or 5 leg, 3 phase or 1 phase, one or split winding per phase, with or without auxiliary
winding,...) .............................................................................................................. 152
Tertiary connected reactors used in SA,FR,RO,NO,FN,IN,AU are normally of
dry type, single phase in construction and are air cooled. BR, SCT use 3ph, 3 leg
oil filled reactors. FN and NZ also have some 3ph, oil filled reactors. ............. 153
10.3 Protection and redundancy .............................................................................. 153
10.3.1 Question: What types of protection are being used for: e.g. Phase to Phase
fault, Phase to Ground fault, Inter-turn fault? ......................................................... 153
10.3.2 Question: What types of non-Electrical protection are being used? (e.g. Oil
Temperature, Winding Temperature, Sudden Pressure, Buchholz relay, Oil level
etc.) 155
10.3.3 Question: In what Reactor Voltage Level do you use 2 Protection Groups
(e.g. Main A and B or Main and back-up)? Describe how the protection functions
are organized in each group? (e.g. differential in group A and overcurrent in group
B) 156
10.3.4 Question: What degree of redundancy is applied? (e.g. 2 DC supplies,
different tripping coils, 2 secondary windings of CT, …) ...................................... 158
10.4 Fault types and protection limitations ............................................................. 159
10.4.1 Question: What types of fault are experienced and how often do they
occur? (e.g. percentage of the occurrence of various faults with respect to the total
number of faults per year). ...................................................................................... 159
10.4.2 Question: What percentage of each category of these faults is not cleared
correctly due to the protection functions limitation? .............................................. 161
10.4.3 Questionnaire: Are you looking at new solutions to resolve the above-
mentioned limitations? If yes, please describe. ...................................................... 162
6
10.5 Control and monitoring of shunt reactors ....................................................... 163
10.5.1 Question: What types of control are being used for switching in and out
the shunt reactors? (e.g. Manual or Automatic) If Automatic what philosophy is
used? 163
10.5.2 Questionnaire: How often the shunt reactors are being switched in and
out? 164
10.5.3 B5.3 Questionnaire: Do you use synchronized switching (point on wave
switching)? How are they applied? ......................................................................... 166
10.5.4 B5.4 Question: What parameters are monitored (e.g. I, U, Q, Oil and
winding temperature, dissolved gas, bushing residual current, fire detectors,...) and
what actions are taken? ........................................................................................... 167
A. Example of Controlled switching of a 500kV shunt reactor in japan ..................... 169
A.1 Introduction ..................................................................................................... 169
A.2 Reactor current interruption test circuit .......................................................... 169
A.3 Relation between opening phase angle and re-ignition .................................. 170
A.4 Application of controlled switching................................................................ 172
A.5 Conclusion ...................................................................................................... 174
B. Protection of 800kV 3 phase shunt reactor provided with spare single phase reactor
176
C. Protection of Shunt reactor auxiliary winding ........................................................ 180
D. Application of 500 kV Shunt Reactor with Auxiliary Winding System in Switching
Station ............................................................................................................................. 182
D.1 Introduction ..................................................................................................... 182
D.2 Solutions to the Problem ................................................................................. 182
D.3 The main structure and technical parameters of the shunt reactor.................. 182
D.4 Measures for protection .................................................................................. 184
D.5 Operating history ............................................................................................ 184
D.6 Conclusions ..................................................................................................... 184
E. Automatic Reactive Switching in UK..................................................................... 185
F. ABBREVIATIONS ................................................................................................ 190
G. REFERENCES ....................................................................................................... 195
G.1 Books .............................................................................................................. 195
G.2 Technical Papers ............................................................................................. 195
G.3 Other Documents ............................................................................................ 196
TABLE OF FIGURES
Figure 1 Three-leg shunt reactor core ............................................................................... 14
Figure 2 Five-leg shunt reactor core with three wound limbs .......................................... 14
Figure 3 Reactor construction, Core type (A), Shell type (B). ......................................... 17
Figure 4 Simple shunt reactor model ................................................................................ 19
Figure 5 Aircore reactor .................................................................................................... 21
Figure 6 Typical magnetizing characteristic of a gapped core shunt reactor ................... 22
Figure 7 Test result of inrush ............................................................................................ 25
Figure 8 Bus connected, 300kV, 150MVAr, oil immersed shunt reactor ........................ 26
7
Figure 9 Bus connected, 420kV, 200MVAr, oil immersed shunt reactor ........................ 27
Figure 10 Phase C winding currents during shunt reactor switching in and tripping out . 28
Figure 11 Shunt reactor characteristics for a gapped and air core reactor ........................ 28
Figure 12 Idealized inrush currents for an air core reactor ............................................... 29
Figure 13 Idealized inrush currents for an iron core reactor ............................................. 29
Figure 14 Example of harmonic content in an idealized inrush current for an iron core
reactor ....................................................................................................................... 30
Figure 15 Example of the relative content of the second harmonic and inrush peak current
for different degree of offset. The slope of the saturated part is 30 % of the
unsaturated slope ....................................................................................................... 30
Figure 16 Example of the relative content of the second harmonic and inrush peak current
for different degree of offset. The slope of the saturated part is 50 % of the
unsaturated slope ....................................................................................................... 31
Figure 17 Shunt reactor current for operation with over voltage...................................... 32
Figure 18 Harmonics in the reactor current in case of over voltage ................................. 32
Figure 19 Solidly grounded three phase reactor directly connected to line ...................... 35
Figure 20 Three phase and neutral reactor connected to bus or line via circuit switcher or
circuit breaker ........................................................................................................... 35
Figure 21 Shunt reactor connected to transformer tertiary winding switching via circuit
switcher or circuit breaker on supply side ................................................................ 36
Figure 22 Shunt reactor connected to transformer tertiary winding switching via circuit
switcher or circuit breaker on neutral side ................................................................ 36
Figure 23 Nominal-π circuit of a transmission line .......................................................... 37
Figure 24 Nominal-π circuit with shunt reactors added to both end of the transmission
line............................................................................................................................. 38
Figure 25 Equilibrium chart relating water vapor pressure over oil to water concentration
in insulation (kraft) paper vs. temperature. ............................................................... 42
Figure 26 Typical shunt reactor connections .................................................................... 45
Figure 27 Principle of low impedance phase differential protection ................................ 46
Figure 28 Tripping characteristic - Example .................................................................... 47
Figure 29 Principle of high impedance phase differential protection ............................... 49
Figure 30 High impedance differential protection - stabilizing principle......................... 50
Figure 31 Differential currents of a 1% turn to ground fault in phase L1 at the neutral
point of a 150 MVA, 220 kV, 50Hz shunt reactor.................................................... 51
Figure 32 Example - Tripping characteristic for a two stage numerical low impedance
differential protection applied to a shunt reactor. ..................................................... 53
Figure 33 Shunt reactors inrush current in phase C with fully DC offset – response of two
different types of digital filters. ................................................................................ 54
Figure 34 Low impedance restricted earth fault during external fault condition ............. 55
Figure 35 Low impedance restricted earth fault during internal fault condition .............. 55
Figure 36 High impedance restricted earth fault during external fault condition ............. 56
Figure 37 Restrained low impedance restricted earth fault during energizing and one
phase CT saturated .................................................................................................... 58
Figure 38 Terminal side connected phase overcurrent protection .................................... 59
Figure 39 Earth fault overcurrent protection alternatively connected to CT in the neutral
to ground connection or phase CT’s on terminal side. ............................................. 61
8
Figure 40 Terminal side connected distance protection ................................................... 62
Figure 41 Current and apparent reactance in a turn-to-turn fault affected phase for a
grounded 200MVAr, 420kV, 50Hz shunt reactor. ................................................... 65
Figure 42 Negative sequence current and residual current 3I0 in a turn-to-turn fault
affected 200MVAr, 420kV, 50Hz shunt reactor. ...................................................... 66
Figure 43 Turn to turn fault protection based on zero sequence directional control ........ 67
Figure 44 Turn to turn fault protection based on negative sequence directional control . 68
Figure 45 Three phases split winding protection .............................................................. 69
Figure 46 Tank protection ................................................................................................. 69
Figure 47 Basic breaker failure protection scheme (logic phase L2 and logic phase L3 are
similar to logic phase L1 as detailed in the figure). .................................................. 71
Figure 48 An example for CB pole discrepancy scheme .................................................. 72
Figure 49 Buchholz relay mounting arrangement ............................................................ 73
Figure 50 The sudden pressure relief device .................................................................... 74
Figure 51 Bus connected shunt reactor protection scheme number 1 .............................. 75
Figure 54 Line connected shunt reactors – typical arrangements. .................................... 78
Figure 55 Example of a reactor neutral overcurrent on a 500kV power line ................... 80
Figure 56 Breaker closed to start the line restoration ....................................................... 81
Figure 57 Trip release logic for inrush suppression and blocking when deenergized (line
connected shunt reactors) .......................................................................................... 82
Figure 58 Shunt reactors connected to tertiary winding of autotransformers. .................. 83
Figure 59 Negative sequence protection. .......................................................................... 85
Figure 60 Ground overvoltage protection connected to broken delta winding of a voltage
transformer / grounding transformer. ........................................................................ 86
Figure 61 Simple split phase protection based on current balance measurement between
the two shunt reactor neutrals. .................................................................................. 88
Figure 62 Turn to turn fault protection based on voltage unbalance scheme. .................. 89
Figure 63 Turn to turn fault protection – alternate voltage unbalance scheme. ............... 90
Figure 64 Phase differential protection for autotransformer including tertiary connected
shunt reactor in the protected zone. .......................................................................... 91
Figure 65 Transformer tertiary connected shunt reactor scheme number 1 ..................... 92
Figure 68 Temperature rise as a function of time ............................................................. 96
Figure 69 Nitrogen injections in prevention mode logic. ............................................... 100
Figure 70 Nitrogen injections in extinction mode logic. ................................................ 101
Figure 71 Transformer (and shunt reactor) fire protection system ................................. 101
Figure 72 A typical system for shunt reactor fire protection .......................................... 102
Figure 73 Voltage sensor connected to bushing ............................................................. 108
Figure 74 Typical system integrated architecture for shunt reactor monitoring ............. 114
Figure 75 Typical decentralized architecture for on-line monitoring system. ................ 114
Figure 76 Observe the DC offset of the instantaneous differential current .................... 116
Figure 77 A shunt reactor suddenly connected to the power system (inrush), with some
eventual differences in the CTs on both sides, or different loads on them, the
9
differential protection sees these currents as shown in the figure. As a result, the
differential protection calculates (false!) instantaneous differential currents as shown
in the figure. These false differential currents might cause an unwanted trip of the
reactor. .................................................................................................................... 117
Figure 78 An example of a good operate – restrain characteristic. Under heavy external
fault condition with current transformer saturation, the spurious false differential
current does not enter the operate region. ............................................................... 118
Figure 79 Internal/external fault discriminator. .............................................................. 119
Figure 80 Example of an internal fault using internal / external fault discriminator. ..... 120
Figure 81 VT and CT used for the turn-turn protection.................................................. 121
Figure 82 Zero sequence network at external phase-ground fault .................................. 121
Figure 83 Zero sequence voltage and current at external phase-ground fault ................ 122
Figure 84 Zero sequence network for internal turn-turn fault ........................................ 122
Figure 85 Zero sequence voltage and current for internal turn-turn fault....................... 122
Figure 86 External respective internal fault to ground ................................................... 123
Figure 87 To improve sensitivity for internal turn-turn faults ........................................ 124
Figure 88 Voltage waveform .......................................................................................... 124
Figure 89 Example of logical diagram for shunt reactor automation ............................. 137
Figure 90 Block diagram for principle of controlled switching ..................................... 139
Figure 91 Point -on-wave general operating principle ................................................... 139
Figure 92 Voltage across shunt reactor at de-energizing without re-ignition ................. 140
Figure 93 Voltage across shunt reactor in event of re-ignition....................................... 141
Figure 94 Target for contact separation in order to eliminate re-ignitions ..................... 141
Figure 95 Point-of-wave block diagram for opening ...................................................... 143
Figure 96 Point-of-wave block diagram for opening and closing .................................. 144
Figure 97 Example on point-of-wave selection tree. ...................................................... 145
Figure 98 Shunt reactor energizing with current start feedback loop ............................. 146
Figure 99 Circuit breaker and half scheme with CT outside reactor bays. ..................... 147
Figure 100 Shunt reactor installation in a CB and a half scheme substation. ................. 148
Figure 101 Single phase test circuit for interrupting 500kV reactor current. ................. 169
Figure 102 Oscillograms obtained in interruption test. .................................................. 170
Figure 103 Relation between probability of re-ignition and T0. ..................................... 171
Figure 104 Voltage between contacts at re-ignition. ...................................................... 171
Figure 105 Illustration of re-ignition suppression by controlled switching. ................... 172
Figure 106 Opening phase control achieved on 550kV one-break circuit breaker......... 174
Figure 107 Shunt reactor installed in the substation. ...................................................... 175
Figure 108 Single line diagram 765 kV substation ......................................................... 177
Figure 109 Three line diagram 765 kV substation.......................................................... 179
Figure 110 Grounded-wye Auxiliary Power Winding for Low-voltage Distribution .... 180
Figure 111 Ungrounded Auxiliary Power Winding for Low-voltage Distribution ........ 181
Figure 112 Automatic Reactive Switching (ARS), relationship of thresholds and dead-
bands for Coarse and Fine Voltage Control Modes of the combined HV/LVARS 187
TABLE OF TABLES
10
Table 1 Requirements met, on bus connected shunt reactor protection scheme number 1
................................................................................................................................... 75
................................................................................................................................... 76
................................................................................................................................... 77
Table 4 Requirements met, on tertiary connected shunt reactor protection scheme number
1................................................................................................................................. 92
2................................................................................................................................. 93
3................................................................................................................................. 94
Table 7 Components used in a typical system for shunt reactor fire protection ............. 102
Table 8 Key gases and fault types................................................................................... 106
Table 9 List proposed of main parameters for shunt reactors monitoring ...................... 109
Table 10 Monitored components and health condition................................................... 113
Table 11 Voltage levels .................................................................................................. 150
Table 12 Shunt reactor connection to the power system ................................................ 151
Table 13 Shunt reactor design......................................................................................... 153
Table 14 Electrical protections - directly connected reactors ......................................... 154
Table 15 Electrical protection used for tertiary connected reactors ............................... 154
Table 16 Non electrical protections- directly connected reactors ................................... 155
Table 17 Non electrical protections - tertiary connected reactors .................................. 156
Table 18 Redundancy for directly connected reactors .................................................... 157
Table 19 Redundancy for tertiary connected reactors .................................................... 158
Table 20 Types of faults on directly connected shunt reactors ....................................... 159
Table 21 Types of faults on tertiary connected shunt reactors ....................................... 160
Table 22 Protection performance- directly connected shunt reactors ............................. 161
Table 23 Protection performance- tertiary connected shunt reactors ............................. 162
Table 24 New solutions for directly connected shunt reactors ....................................... 162
Table 25 Types of control - directly connected shunt reactors ....................................... 163
Table 26 Frequency of switching - directly connected shunt reactors............................ 165
Table 27 Frequency of switching - tertiary connected shunt reactors ............................ 165
Table 28 Point on wave switching - directly connected shunt reactors .......................... 166
Table 29 Reactor parameters monitored - direct connected shunt reactors .................... 167
Table 30 Reactor parameters monitored - tertiary connected shunt reactors.................. 168
Table 31 Test conditions for single phase circuit for interrupting 500kV reactor current.
................................................................................................................................. 170
Table 32 Function descriptions with ANSI, IEC61850-5, IEC61850-7-4 and IEC
Symbols................................................................................................................... 190
Table 33 Abbreviations used .......................................................................................... 192
11
1 INTRODUCTION
Cigre Working Group B5-37 was formed to determine the present state of the art of
protection, monitoring, and control of shunt reactors. The work of this working group
includes reviewing schemes for protecting shunt reactors, especially those advancements
resulting from use of numerical relays. Recommendations for using modern protection
techniques are provided. Advancements in integrated control and monitoring of shunt
reactors are reported and recommendations for application are provided.
This report is arranged beginning with descriptions and applications of shunt reactors,
reactor fault types and abnormal conditions encountered, protection (both electrical and
mechanical) including monitoring and adaptation to instrument transformer performance,
and recommendations for using protection. Control and monitoring is then discussed
including recommendations. Control includes rate of recurrence of switching and
techniques to minimize system impact during energization/denergizations. Monitoring
includes dissolved gas and fire detection (deluge).
The working group issued a questionnaire on existing practices to the participating Cigre
countries. The survey asked for information on the voltage levels utilized, types of
connections to the power system, grounding methods, reactor design, protection and
redundancy used including use of primary and backup schemes, types of faults
experienced and protection performance, types of control (automatic and/or manual), use
of synchronized switching, and reactor parameters monitored. The results of the survey
responses, knowledge and experience of the B5-37 working group members, and industry
publications and standards were all used to develop this final report.
12
2 DESCRIPTION OF SHUNT REACTORS
General statement
Shunt reactors are inductive loads that are used to absorb reactive power to reduce the
over voltages generated by line capacitance.
An inductive load consumes reactive power versus a capacitive load generates reactive
power.
A transformer, a shunt reactor, a heavy loaded power line, and an under magnetized
synchronous machine are examples of inductive loads. Examples on a capacitive load are
a capacitor bank, an open power line and an over magnetized synchronous machine.
Although shunt reactors are inductive loads similar to transformers but they are different
than transformers in terms of construction and some electrical characteristics.
To describe the shunt reactors better, we need to look at the different designs of shunt
reactors and their electrical characteristics.
2.1 Design of shunt reactors
Design and operation

Shunt reactors are mainly used in transmission networks. Their function is to consume
the excess reactive power generated by overhead lines under low-load conditions, and
thereby stabilize the system voltage. They are quite often switched in and out on daily
basis, following the load situation in the system. Shunt reactors are normally connected to
substation busbar, but also quite often directly to the overhead lines. Alternatively, they
may also be connected to tertiary windings of power transformers. The shunt reactors
may have grounded, or reactor grounded neutral.
Shunt reactors normally have iron cores with integrated air gaps. Due to the air gaps, the
iron cores cannot be significantly saturated, and the reactors therefore will have a
reasonably linear behavior during energizing events, for example.
Three-phase shunt reactors may consist of three separate single-phase cores, or they
could be of three-leg or five-leg design (alternatively shell type), see Figure 1 and Figure
2. For transmission voltages the Five-leg core type or shell type are mainly used. They
make the three phases magnetically independent, while three-leg cores lead to magnetic
coupling between phases.
Any significant period of one or two phase excitation necessitates the provision of a
clearly defined return path for the zero-sequence flux created by the asymmetrical
excitation. Where single phase operation is likely to occur, e.g. in power systems
employing single pole auto-reclosing, there are two optional ways to achieve such a
return path for the zero-sequence flux. These are:
13
1. To use a three phase 5-limb core (or shell type core).
2. To use single phase units.
One major advantage with a five leg reactor (or shell type) compared with a three leg
reactor is that the construction to reduce vibrations and the long term use is much more
stable and stronger.
Figure 1 Three-leg shunt reactor core
Figure 2 Five-leg shunt reactor core with three wound limbs
Medium voltage reactors, connected to tertiary windings of transformers, in most cases

have air-insulated windings without iron cores.
Oil or Dry type shunt reactors

There are two types of shunt reactor groups one is the oil immersed type similar to
transformers the other type is the air core or core-less reactor.
The dry type of reactors are normally used up to 34.5 kV and often installed on the
tertiary of a transformer [ref 7].
The design is divided in gapped core and core-less reactors. The gapped core has a
subdivided limb of core steel with air gaps inside the winding and no limb at all for the
14
core less concept. The gapped core gives compact design with low losses and low total
mass, low sounds and low vibrations.
Higher energy density can be achieved in a gapped core compared to an air core.
The slope of the permeability is greater in a gapped core versus an air core reactor.
The primary advantages of dry-type air-core reactor, compared to oil-immersed types, are
lower initial and operating costs, lower weight, lower losses, and the absence of
insulating oil and its maintenance. The main disadvantages of dry-type reactors are
limitations on voltage and kVA rating and the high-intensity magnetic field. There is no
magnetizing inrush current when the reactor is energized.
Unit ratings for existing single phase or three phase shunt reactors are:
Three phase up to: 250 Mvar

Single phase up to: 130 Mvar / phase
System voltages up to: 800 kV
Single Phase reactors

Single phase reactors are used when the power is above the limits for a three phase shunt
reactor.
Three phase (3-leg or 5-leg shunt reactors)

Most three phase shunt reactors are designed with five limbs, because of a more robust
construction and to reduce vibrations over time, since shunt reactors should last 40-50
years.
The unwound side limbs results in that the zero sequence impedance is equal to the
positive sequence impedance.
In a high voltage star connected shunt reactor the zero sequence reactance is dependent
on the core arrangement.
The physics are the same as the case with a star/star connected transformer. Under
symmetrical excitation the sum of momentary flux values in the three phases is zero. But
under earth fault conditions this is not the case and the resulting flux must find a way
back external to the three phase coils. In three limb reactors, this resulting flux would go
through the air from yoke to yoke; it means that the zero sequence reactance is lower than
the normal reactance and also non-linear, leads to higher zeros sequence current.
In some applications it is a distinct advantage if the reactor has high and constant zero
sequence reactance. This is the case when single pole reclosing is either required from the
beginning, causing zero sequence flux each time a single pole is opened, or considered
for future development of the system to limit the zero sequence flux. In certain cases (e.g.
line connected shunt reactors) it is also recommended that the zero sequence reactance be
tuned to a fixed, high value by the addition of an auxiliary neutral reactor.
15
As for transformers, a high zero sequence reactance requires a low reluctance unwound
return path in the magnetic circuit, leads to smaller zero sequence current – which is
achieved with a five-limb core. In a reactor this will result in a next to absolute
decoupling of the phase limbs from each other because the wound limbs are gapped and
the outer limbs are not. This is easily verified by measuring non-induced voltage on the
other phases when one phase is energized [ref 24].
Split winding
The split winding is used when the current have exceeded the maximum of mechanical
reasons in the construction, then two windings per phase will be parallel in the reactor.
Shunt reactors equipped with auxiliary power windings

Because shunt reactors are used to control voltage at the receiving ends of long Extra
High-Voltage and Very High-Voltage transmission lines, these reactors often are located
in remote regions that may not have an extensive or reliable distribution grid. Obtaining
reliable station power for the reactor switching station can be a problem. The high-
voltage reactor application usually calls for oil-immersed reactors that look very similar
in appearance to power transformers. When designed with an air-gapped iron core, these
reactors can be equipped with a secondary core and winding such that a low-voltage can
be extracted from the high-voltage line, see [ref 8] and [ref 34].
Shunt reactor in the neutral

For line connected shunt reactors, an additional single-phase reactor (neutral reactor) is
sometimes connected between neutral and ground. The purpose of this reactor is to
increase the overall zero sequence reactance of the overhead line. In this way, the fault
current is kept small in the event of single-phase line faults cleared by single-pole
opening of the line breakers. As a result, there will be a high probability that the arc at the
fault location is extinguished and that the reclosing operation is successful.
Variable shunt reactor (VSR), with tap-changer

Shunt Reactors are used in high voltage energy transmission systems to stabilize the
voltage during load variations. A traditional shunt reactor has a fixed rating and is either
connected to the power line all the time or switched in and out depending on the load.
Recently Variable Shunt Reactors (VSR) have been developed and introduced on the
market. The rating of a VSR can be changed in steps, The maximum regulation range
typically is a factor of two, e.g. from 100-200 Mvar. The regulation speed is normally in
the order seconds per step and around a minute from max to min rating. VSRs are today
available for voltages up to 550 kV.
The variability brings several benefits compared to traditional fixed shunt reactors. The
VSR can continuously compensate reactive power as the load varies and thereby securing
voltage stability. Other important benefits are: - reduced voltage jumps resulting from
16
switching in and out of traditional fixed reactors, - flexibility for future (today unknown)
load and generation patterns, - improved interaction with other transmission equipment
and/or systems such as coarse tuning of SVC (Static Var Compensation) equipment, -
limiting the foot print of a substation if parallel fixed shunt reactors can be replaced with
one VSR, - a VSR can be used as flexible spare unit and be moved to other locations in
the power grid if needed.
Shell type core

The reactor design is said to follow either a core or shell type concept [ref 24].The
difference in the two concepts was originally attributed to the arrangement of the core. In
the core type transformers the coils appear to surround the core and in the shell type the
core appear to surround the windings, see Figure 3.
Figure 3 Reactor construction, Core type (A), Shell type (B).

This simple definition does not always hold. The way reactors are built today; a
description as follows would be more adequate:
 In a core type reactor the core limb has a shape of a cylinder around which the
coils are arranged. For normal core type power reactors the coils too are
cylindrical and arranged concentrically. Each terminal is connected to one coil or
several coils in series. Further the coils are slid down around a pre-made core
limb to which yokes are connected after the windings are in position. Most often
the core limbs and yokes are in vertical position.
 In the shell type reactor the separate coils have rectangular cross section and they
are wound in one plane. After the winding work the coils for one terminal are
stacked up on each other and connected together. The groups of coils are then in
turn, stacked together to form a winding packet for the complete circuit. The
packets for each phase are then raised to the upright and adjusted position it has in
the reactor. In and around these packets the core is now built up.
A five leg shunt reactor of core type has similar characteristics as a three leg shell type
reactor (to have a low vibration and noise level, low zero sequence current), so the reason
for a manufacturer to keep to a certain concept may today be historical.
17
2.2 Electrical characteristics
The electrical characteristics of a shunt reactor that needs to be studied are:

 Air-gap of the shunt reactor core.
 Switching in a shunt reactor
 Disconnection of the shunt reactor
 Dominating harmonics
 Hysteresis
 Losses in shunt reactors
2.2.1 Air-gap of the shunt reactor core

To avoid saturation of the iron-core of the shunt reactor, small air gaps are distributed
along the core. The distribution of the magnetic flux intensity from the iron to the air-gap
can in that way handle larger H-field (magnetic flux intensity [A/m]). The air-gap is
seldom larger than a couple of millimeter. The magnetic flux intensity is much larger in
air than in iron approximately 1000 times.
In a transformer the aim is not to have any air-gap, therefore the slope in the B-H curve is
very steep with large hysteresis and remanence compared to a shunt reactor, however for
a reactor with air-gap the B-H curve becomes flatter and the hysteresis is very small with
practically no remanence. This leads to that there are small inrush currents and long DC
constants for the shunt reactor, compared to the transformer.
Air core reactors (no iron core), have small inductance L, high losses and high current in
the windings.
Oil immersed iron core reactors with air-gap, have higher inductance, smaller losses and
less current in the windings.
By introducing iron in the winding a higher inductance can easily be achieved without
increasing massively the number of turns as in air-core.
2.2.2 Inrush
When switching in air core or iron core reactors, long DC components up to 1 second can
occur, some differences distinguishes between inrush in air respective iron core reactors.
2.2.2.1 Switching in air core reactors

B
For an air core reactor where L is practically constant (i.e. permeability    k 0 is
H
constant) a simple model with a breaker switched in can describe the electrical principles,
see Figure 4 below [ref 2]:
18
t
e  Eˆ sin(t   ) R
AC
Figure 4 Simple shunt reactor model
e  uL
 iL eq. 1
R
di
uL  L L eq. 2
dt
di
eL L eq. 3
dt  i
L
R
R e
i L  i L   0 eq. 4
L L
The instantaneous voltage value when the breaker is closed.

e  Eˆ sin(t   ) eq. 5
The derived short current obtained through the limiting values

i L  i AC  i DC  0 i  i AC eq. 6
 and  L
 t  0  t  
Eˆ   t 
R
eq. 7
i L  i AC  i DC   sin( t     )  sin(   )  e L 

Z  
Where the impedance is
Z  R 2  (L) 2 eq. 8
The first term in eq. 7 states the time function for the steady state condition which is an
AC current and the second term indicates the transient condition which is a damped DC
current.
L
The time constant of the damping is   .
R
The time constants for shunt reactors are longer than transformers and can be up to 1
second. Transformers have DC constants up to a couple of hundred milliseconds.
19
The inrush can therefore easily saturate a CT measurement of the current.
As can see from eq. 7 the size of the short circuit current is depending on the voltage
phase angle at the instantaneous moment when the breaker is closed. If    then the
elapsed curving inward oscillation disappears (transient part) and the short circuit current
only consists of the steady state part.
With regard to the instantaneous phase angle of the voltage  there is two cases of special
interests:
Case 1   0
The short circuit is here done when the voltage run-through zero. In this case the current
reaches its absolute maximal instantaneous value, which is realized analytically when the
derivative of i k with regard to  and t shall be zero in the maximum moment, i.e.
 ik   t
R
eq. 9
 2 I k cos( t     )  cos(   )  e L   0
  
 ik  R  t
R
eq. 10
 2 I k   cos( t     )  sin(   )  e L   0
t  L 
L
tan     tan( )     0,    ,   2 , eq. 11
R
The maximum inrush current ik max appears when using realistic power system
impedances, approximately at the same time when the steady state current reaches its first
 
  
peak value, i.e. t  
2
, see eq. 12 below:

    
R  
 L 2 
eq. 12
ik  2 I k 1  sin( )  e 
 
We observe that the current time derivative is zero in the short circuit moment and that Zs
Zs
AC AC
the DC component initial value is not the maximal imaginable, since in the Shunt reactor Shunt reactor
Zs
Zs
AC AC
expression of . Current
*
transforme
r
Shunt reactor
Voltage
transforme
r
Shunt reactor
Zs
Zs
AC U0
AC
fault
Voltage
*
Case 2 Current
transforme
r
I0
transforme
r
AC
Shunt reactor Shunt reactor
In this case has the steady state short circuit current the (negative) top value in the short
circuit moment and the DC component has in consequence hereby largest possible value.
Installed in eq. 7 the short circuit current can be derived:
20
  RL t  eq. 13
ik  2 I k e  cos( t )
 
Maximum instantaneous value occurs likewise approximately in the AC components first
U0
top value at time I0

and is:
Zs
Zs
AC AC
*
U0
Voltage
eq. 14
Current transforme
transforme r
r
I0
AC
Shunt reactor Shunt reactor
Because the asymmetry in a certain (some) degree here is largest it is close to expect that
this maximum value not is much less then the theoretic correct value that can be derived
in the case   0 . This is also correct for a power network characteristic data and one can
with good approximation use this simple expression when deriving the inrush current.
I0
Observe the ideal case with U0 when both cases lead to the same result. In this un-
damped case is:
Ext fault U0
I0
eq. 15
Int fault U0
Figure 5 Aircore reactor

A short circuit current with DC component is often called unsymmetrical short circuit
current, and speaks about greater or less degree of asymmetry that is dependent on the
instantaneous time of the short circuit. With maximum asymmetry it is meant a current
which DC component has the same value as the AC components un-damped top value.
Observe that the word symmetry and asymmetry here is used in another meaning then
three phase symmetry and three phase asymmetry.
21
2.2.2.2 Switching in oil immersed iron core reactors
For an iron core reactor where L is not constant, the permeability is depending on the
magnetic field intensity H [A/m] and magnetic flux density B [Vs/m2] (i.e.
B
permeability   ) a simple model with a breaker switched in can describe the
H
electrical principles see Figure 4, Figure 8 and Figure 9.
The magnetizing characteristic of a shunt reactor

The relation between voltage and current can be described with two lines one for the
unsaturated part and one for the saturated part. The intersection between is called knee
point and is usually located at 125 – 135 % of the voltage amplitude. The saturated region
has a slope of 20 – 50 % of the unsaturated region. Typical HV shunt reactor magnetizing
characteristic is shown in Figure 6 [ref 7].
Shunt Reactor Characteristic

2.5
2
Voltage [pu]
1.5
0.5
0
0 1 2 3 4
Current [pu]
Figure 6 Typical magnetizing characteristic of a gapped core shunt reactor
Description of parameters
L = inductance [H]
N = number of windings
l FE = length of flux in iron yoke and limb [m]
 0  4 10 7 permeability in air [Vs/Am]
 r = permeability constant iron
 = total permeability [Vs/Am]
B = magnetic flux density [Vs/m2]
H = magnetic field intensity [A/m]
22
 = total flux [Vs]
AFE = iron area in limb [m2]
Switching in a shunt reactor

The inductance L and permeability  of the shunt reactor can be derived [ref 1]
  A  AFE
L  N 2 0 r FE  N 2 eq. 16
l FE l FE
B
 eq. 17
H
The differential equation when a shunt reactor is switched in follows
di 1 ~
 (e  Ri ) eq. 18
dt L
By using the flux  [Vs] eq. 19 and field intensity H [A/m] in eq. 20 into the eq. 17 and
eq. 16 the inductance L can be derived see eq. 21
  B  AFE  N eq. 19
H  l FE
i eq. 20
N
d
L eq. 21
di
By introducing the inductance in eq. 21 the magnetic flux over the inductance follows in
eq. 23:
di di ~
 (e  Ri ) eq. 22
dt d
d
 (e~  Ri )  u L eq. 23
dt
To derive the magnetizing voltage and the total magnetic flux the recursive equations
(Euler equations [ref 3]) are used in eq. 24 and eq. 25:
Starting conditions:
, B1  BR , 1  BR  AFE  N
u L k   e~k   R iL k  eq. 24
k  1  k    t  u L k  eq. 25
23
The magnetic field intensity (H) can be derived from the graph in [ref 4] after calculating
of the magnetic flux density (B):
k  1
Bk  1   H k  1 eq. 26
AFE  N
Calculate the magnetizing current iL :

H k  1
i L k  1   l FE eq. 27
N
24
Simulation of Shunt Reactor inrush
E [V] 500
-500
0 0.5 1 1.5
5
Im [A]
-5
0 0.5 1 1.5
500
Um [V]
-500
0 0.5 1 1.5
5
B - [T=Vs/m2]
-5
0 0.5 1 1.5
5
H [A/m]
-5
0 0.5 1 1.5
5
Phi [Vs]
-5
0 0.5 1 1.5
Figure 7 Test result of inrush

E = Voltage over shunt reactor model [V]
Im = Magnetizing current trough shunt reactor [A]
Um = Magnetizing voltage over inductive part [V]
B = Magnetic flux density [T=Vs/m2]
H = Magnetic field intensity [A/m]
Phi = Magnetic flux over the shunt reactor [Vs]
25
During inrush the permeability can move up over the threshold knee and cause a transient
current greater than 2 2 I k (3 to 5 I k ) and after several seconds reaching steady state
current, see Figure 6 and Figure 7.
As the shunt reactor is moving from saturation region to steady state region, the
permeability increases towards the constant value (  ) and the current decreases to steady
state value. The damping time  also increases with less saturation as the permeability of
the shunt reactor moves towards steady state.
Both above statements can be seen from the following equations:
lj
iN 
0 r A eq. 28
L
 eq. 29
R
Figure 8 Bus connected, 300kV, 150MVAr, oil immersed shunt reactor
26
Figure 9 Bus connected, 420kV, 200MVAr, oil immersed shunt reactor
2.2.3 Shunt reactor disconnection

Disconnection of small reactive current was at one time regarded as a dangerous
operation because of the risk of current chopping and resulting switching overvoltage.
Modern surge arresters are fully capable of handling this condition, and besides, the
tendency of the circuit breaker to chop reactor current is not so pronounced for typical
HV shunt reactor rated current values [ref 24].
However, the primary current chopping causes a transient, an exponentially decaying dc

current component in the CT secondary circuit (see Figure 10 for a similar example).
This secondary dc current has no corresponding primary current in the power system; it is
caused by a discharge of the magnetic energy stored in the magnetic core of the current
transformer. However these discharge secondary currents are typically very small for
shunt reactors and pose no effect on the reactor protection schemes with numerical relays
[ref 7]
27
99.2MVA, 440kV, 60Hz R eactor
4
2
[pu]
2
0 2.08 4.17 6.25 8.33 10.42 12.5 14.58 16.67 18.75 20.83 22.92 25
Cycles
IC
IcN
Figure 10 Phase C winding currents during shunt reactor switching in and tripping out
2.2.4 Harmonics
The zero, 2nd and 3rd harmonics in a shunt reactor are described below.
Some figures and text under section 2.2.4 are used from the ongoing Cigre B5 WG24
“Protection requirements on transient response of voltage and current digital acquisition
chain”, to show the harmonic content during overload conditions and inrush conditions.
2,5
2,0
Voltage [pu]
1,5
Coreless
Gapped core
1,0
0,5
0,0
0,0 0,5 1,0 1,5 2,0 2,5 3,0 3,5 4,0
Current [pu]
Figure 11 Shunt reactor characteristics for a gapped and air core reactor
28
Based on the shunt reactor characteristics shown in Figure 11, the contents of harmonics
in idealized undamped inrush currents have been analyzed. A sinusoidal voltage has been
applied to the reactor at different inception angles which results in inrush currents with
different DC offset. As the damping is neglected the calculated contents of harmonics is
only valid for the initial part of the inrush.
As the reactor characteristic for an air core reactor is a straight line the inrush current will
be a sinusoidal current with different DC offset, see Figure 12. The inrush current
consists only of the fundamental frequency and in most cases a DC offset. Except for the
transient state there will not exist any harmonics. The maximum inrush peak is 2.8 times
the rated current.
3,5
3,0
2,5
2,0 100 % offset
Current [pu]
1,5 75 % offset
1,0 50 % offset
0,5 25 % offset
0,0
-0,5
-1,0
-1,5
0 10 20 30 40 50 60
Time [ms]
Figure 12 Idealized inrush currents for an air core reactor
4
100 % offset
3
Current [pu]
75 % offset
2 50 % offset
25 % offset
1
-1
-2
0 10 20 30 40 50 60
Time [ms]
Figure 13 Idealized inrush currents for an iron core reactor

Figure 13 shows idealized inrush currents for an iron core reactor with the knee point at
125 percent of the rated voltage and the slope of the saturated part is 30 % of the
unsaturated slope. The content of the harmonics in relation to the fundamental harmonic
is shown in Figure 14. The second harmonic is predominant but very dependent of the
29
degree of the DC offset. The second harmonic has been analyzed further and Figure 15
shows the relative content of the second harmonic for different offset. The peak value of
the inrush is also shown in the figure. Figure 16 shows the peak value and the second
harmonic of the inrush current if the slope of the saturated part of the reactor
characteristic is 50 % of the unsaturated slope instead of 30 %. The peak current and the
harmonic content are reduced.
25
20
40 % offset
15
an/a1 [%]
70 % offset
100 % offset
10
0
2 3 4 5 6 7 8 9 10
nth Harmonic
Figure 14 Example of harmonic content in an idealized inrush current for an iron core reactor
30 6
I100/I50
I100peak/Ipeak, I100/I50, [%]
25 5 I100p/Ipeak
Ipeak
20 4
Ipeak /Ir [pu]
15 3
10 2
5 1
0 0
0 10 20 30 40 50 60 70 80 90 100
Offset of energizing voltage [%]
Figure 15 Example of the relative content of the second harmonic and inrush peak current for different
degree of offset. The slope of the saturated part is 30 % of the unsaturated slope
30
20 5
I100/I50
I100peak/Ipeak, I100/I50, [%]
I100p/Ipeak
4
15 Ipeak
Ipeak /Ir [pu]

3
10
2
5
1
0 0
0 10 20 30 40 50 60 70 80 90 100
Offset of energizing voltage [%]
Figure 16 Example of the relative content of the second harmonic and inrush peak current for different
degree of offset. The slope of the saturated part is 50 % of the unsaturated slope
2.2.4.1 Zero Harmonic (DC)

The zero harmonic current (DC component, offset) appears at connection and
disconnection of the shunt reactor, with a time constant up to seconds for shunt reactors,
due to inherent low losses in a shunt reactor (small resistance compared to inductance).
The damping of the DC component in the inrush current is not constant but generally
slow due to the inherent low losses in shunt reactors. During the very first cycles the
damping is fairly pronounced when the core steel goes into saturation giving rise to high
current peaks. The damping during this stage might correspond to a time constant from
around 100 ms to almost 1 s. Later on when the linear part of the flux-current relation has
been reached the damping will be lower as the losses in the reactor gets low. The time to
more or less fully balanced operation around zero flux in the core may be fairly long
often in order of several seconds.
2.2.4.2 2nd Harmonic

Inrush current in a shunt reactor doesn’t appear as a differential current like that which
appears in a transformer, unless the CT saturates after some time due to long DC time
constant. Though the level of second harmonic in many cases can be relatively high there
are many cases with no or very low content of harmonics. The level of 2nd harmonic is
small in shunt reactors compared to transformers.
31
2.2.4.3 3rd Harmonic
The 3rd harmonic is the dominant harmonic in shunt reactors during normal operating
condition, due to asymmetries in the reactor windings.
The 3rd harmonic can be seen in the neutral point of the shunt reactor or as residual zero
sequence current using all phases [ref 24].
If an iron core reactor is exposed to over voltage the current will be distorted and contain
odd harmonics. Figure 17 shows the current for a gapped core with reactor characteristic
according to Figure 11 and the knee point at 125 percent of rated voltage. Figure 18
shows the content of the harmonics related to the fundamental frequency of the current.
2
130 % voltage
1
135 % voltage
Current [pu]
0 140 % voltage
-1
-2
-3
0 10 20 30 40 50 60
Time [ms]
Figure 17 Shunt reactor current for operation with over voltage
5 130 % voltage
an/a1 [%]
135 % voltage
4
140 % voltage
3
0
2 3 4 5 6 7 8 9 10
nth Harmonic
Figure 18 Harmonics in the reactor current in case of over voltage
32
2.2.5 Hysteresis
There is practically no remanence in a shunt reactor compared to a transformer. The
small air gaps along the reactor winding create a thin hysteresis in the B-H curve for the
reactor and therefore very small remanence.
2.2.6 Losses
The fundamental losses in a shunt reactor are winding resistance and magnetization
losses, eddy current losses are also present but small in comparison.
The resistance loss is proportional to the weight of the winding material and to the square
of the current density. The magnetization loss in the core steel also rises by
approximately the square of the flux density.
The total loss is generally 0.2% active power (W) of the total reactive power of the shunt
reactor distributed as follows [ref 24]:
 Resistance losses in winding, P  RI 2 60-70%
 Core steel loss 20-30%
 Eddy current losses, winding
and mechanical parts 5-15%
33
3 APPLICATION OF SHUNT REACTORS
Shunt reactors are used to compensate for large line charging capacitance of long high
voltage power transmission lines and cables.
Their major applications are:

 Preventing over voltages that occur when the line is lightly loaded (Ferranti
Effect).
 Providing voltage control.
 Compensating for line charging reactive power demand of the open-circuit line.
 Suppressing the secondary arc current for successful single pole reclosing.
3.1 Connection to the Power System and Grounding Methods
The reactors are normally connected to power system in three locations. They can be
connected to Line, Bus or Tertiary winding of the power transformer or auto-transformer.
3.1.1 Line and Bus connected reactors
The line connected reactors are normally connected at both ends of the line as each end
can be energized or de-energized independently.
The shunt reactors can be connected directly to HV lines (see Figure 19) or via circuit
switcher or circuit breaker to HV lines or buses depends on the application (see Figure
20).
The permanently connected reactors are used to prevent overvoltages appear on long
lines due to lightly loading or open circuit. The switched reactors are used for voltage
control.
These reactors are normally grounded, solidly or via a neutral reactor (see Figure 19 and
Figure 20). The neutral reactor is used where single pole auto-reclose is applied, to
suppress the secondary arc current.
34
Line
Reactor
Figure 19 Solidly grounded three phase reactor directly connected to line
Bus or Line
Switching
Equipment
Reactor
Neutral
Reactor
Figure 20 Three phase and neutral reactor connected to bus or line via circuit switcher or circuit breaker
3.1.2 Tertiary winding connected reactors

These reactors are normally ungrounded and can be switched via circuit switcher or
circuit breaker. These switching devices can be on supply side or neutral side of the
reactor (see Figure 21 and Figure 22).
Some utilities have used grounded reactors to reduce TRV (Transient Recovery Voltage)
duty of the switching breaker.
35
Tertiary
Winding
Switching
Equipment
Reactor
Figure 21 Shunt reactor connected to transformer tertiary winding switching via circuit switcher or circuit
breaker on supply side
Tertiary
Winding
Reactor
Switching
Equipment
Figure 22 Shunt reactor connected to transformer tertiary winding switching via circuit switcher or circuit
breaker on neutral side
3.2 Effects of Shunt Reactors on Transmission Line Voltage

To better understand the effects of the shunt reactors, we can use the nominal-π circuit of
a transmission line and compare the receiving-end voltage of a lightly loaded line with
and without shunt reactors.
Although the nominal-π circuit do not represent a transmission line exactly and the
discrepancy between the nominal-π and the actual line becomes larger as the length of the
line increases, it can be shown that the nominal-π may represent long lines sufficiently
well if a high degree of accuracy is not required [ref 5].
36
Is Z Ir
+ +
Vs Y/2 Y/2 Vr
_ _
Figure 23 Nominal-π circuit of a transmission line
To derive Vs from the above circuit (Figure 23), we note that the current in the
capacitance at the receiving end is VrY/2 and the current in the series arm is Ir + VrY/2,
then
Vs  (VrY / 2  Ir )Z  Vr eq. 30
Vs  (ZY / 2  1)Vr  ZIr eq. 31
Is would be the summation of the current in the shunt capacitance at the sending end
which is VsY/2, and the current in the series arm.
Is  VsY / 2  VrY / 2  Ir eq. 32
Is  VrY (1  ZY / 4)  (ZY / 2  1) Ir eq. 33
The equations eq. 31 and eq. 33 can be expressed in the following form:
Vs  AVr  BIr eq. 34

Is  CVr  DIr eq. 35
Where
A  D  ZY / 2  1 eq. 36
BZ eq. 37
C  Y (1  ZY / 4) eq. 38
A and D are dimensionless and B and C are in ohms and mhos, respectively.
Now let us look at an example of a line and using the above equations and compare the
no load receiving-end voltage before and after applying the shunt reactors.
Example: A single-circuit 215 kV, 230 mile transmission line has the following series
impedance and shunt admittance per mile:
z  0.843179.04 Ω/mi
y  5.105  10 6 90 S/mi
37
Then
Z  z  l  193.9179.04 Ω
Y  y  l  1.174  10 3 90 S
We can also derive the no load receiving-end voltage (Vr,nl) by substituting
Ir =0 in Vs equation eq. 34.
Vs  AVr , nl
Vr , nl  Vs / A
Now we need to calculate Vs and A.
To calculate Vs, we use the Vs equation eq. 34 and assume the load on the line is
125MW at 215 kV with 100% power factor.
 
Ir  125MW / 3  215kV  335.70 A
Vr  124.10 kV
A  0.891.42
B  193.9179.4 
Then
Vs  139.728.50 kV
and
Vr , nl  Vs / A  157.0 kV
Now we calculate the no load receiving-end voltage for the same transmission line when
identical shunt reactors are connected at both ends of the line (see Figure 24),
compensating for 70% of the total shunt admittance of the line.
Is Z Ir
+ +
Vs Y/2 Y/2 Vr
_ _
Figure 24 Nominal-π circuit with shunt reactors added to both end of the transmission line
Vs would be the same, but A would change since adding shunt reactors changes the value
of Y:
Z  193.9179.04 
Y  (1  0.7)  1.174  10 3 90  3.522  10 4 90 S
38
and
A  ZY / 2  1  0.9670.38
Then
Vr , nl  Vs / A  144.5 kV
This example shows that adding shunt reactors can limit the rise of the no load voltage at
the receiving end of the line from 157.0 kV to 144.5 kV.
39
4 SHUNT REACTOR FAULTS AND ABNORMAL
CONDITIONS
The modes of failure differ from air-core to oil-immersed designs and this affects their
protection requirements and schemes.
4.1 Fault types in Dry-type reactors

Three types of faults occur in dry-type reactor installations [ref 9]:
1 Phase-to-phase faults on the tertiary busbar, resulting in high magnitude phase
current.
2. Phase-to-ground faults on the tertiary busbar, resulting in a low-magnitude ground
current, dependent upon the size of the grounding transformer ground resistor.
3. Turn-to-turn faults within the reactor bank, resulting in a very small change in phase
current.
Phase-to-phase faults are a low probability fault for dry-type reactors because the reactors
are single phase units with relatively wide spacing between phases. The main cause of
these phase-to-phase faults is when arcing from a failed reactor is not detected soon
enough and the fault ionization moves up into the tertiary busbar resulting in a phase to
phase fault.
Since dry-type reactors are mounted on insulators which provide standard
clearance and insulation to ground, direct winding-to-ground faults are low probability as
well and are produced only when this neutral insulation is bridged by, for example, an
animal. Damage done by a winding to ground fault is determined by the grounding
transformer/resistor impedance.
Turn-to-turn insulation failures in dry-type reactors begin s tracking from
insulation deterioration. Once the arc is initiated, these failures, if not detected quickly,
cascade to the entire winding because of the arc's interaction with the reactor's magnetic
field. If the reactor bank is ungrounded, the current in the healthy phase will increase to 3
times normal phase current and could thermally damage the un-faulted phases of the
reactor bank.
4.2 Fault types in oil immersed reactors

The oil-immersed reactor faults are broken into three categories:
1. High current phase-to-phase and phase-to-ground faults.

2 Turn-to-turn faults within the reactor winding.
3. Miscellaneous failures such as loss of cooling or low oil.
Because of the proximity of the winding with the core and tank winding-to-ground
failures can occur. The magnitude of this fault decreases as the fault is located closer to
the neutral side of the reactor. Turn-to-turn faults start out as a small change in phase
currents but increase operating temperature internal pressure, and accumulation of gas. If
40
these are not quickly detected they will evolve into a major fault.
4.3 Failure rates of shunt reactors
The definition of failure rates yields;

Failure rate = no of failures / (total population * total unit years)
Actual failure rate data for reactors is not always kept by utilities. Failure rates of shunt
reactor may vary large from utility to utility in different countries and is affected by
design, quality and workmanship. For example a failure rate between 0.5-1.0% of shunt
reactors, may increase to several percent during large expansion of the grid. Yearly
maintenance of the shunt reactors and bushings will keep the failure rate down.
Data from Canada and India indicates the distribution of failures can for example be
approximately 30-40% bushing related, 30-40% winding related, 15-20% magnetic
circuit, 10-15% terminals, and the failure origins may be distributed as 80% dielectric,
10% thermal, 10% mechanical or others like unknown, chemical, geomagnetic induced
currents.
4.4 Turn to turn faults

Phase to Phase and Phase to Ground faults can be caused by turn-turn faults. The location
of the turn-turn fault is most likely in the windings closest to the high voltage part of the
shunt reactor, caused by for example an impulse voltage from electrostatic discharge like
lightning storms. Each winding on the shunt reactor can be seen as an inductance parallel
with a leakage capacitance and capacitance to ground. The inductive part acts stiff on
inrush currents, and the capacitive part causes an exponential distribution of voltage over
the winding, with max at the top due to high frequency.
The capacitive part consists of the insulation material e.g. paper. If the highest voltage
difference between the windings on top of the shunt reactor exceeds the capacitive
insulation level, the insulation material deteriorates and causes a turn-turn fault between
two windings. A possible way to protect for this is to design the shunt reactor with more
insulation in the top and equip the system with a surge arrester, to limit high currents.
Earlier problems with oil containing copper used in shunt reactors and transformers
caused turn-turn faults, today with improvements in oil quality, this special problem has
disappeared.
Another cause of the turn-turn fault is vibrations. Vibrations create insulation material
fatigue which in turn reduces the level of insulation and can cause a turn-turn faults.
Samples from oil and material could tell the condition of the shunt reactor insulation.
41
Turn-turn faults can also be caused by excessive water in insulation paper, which can
give raise to water vapor bubbles when temperature increases, thus creating a low
dielectric strength region leading to electric arc.
The main risk for short-time failures is the reduction in dielectric strength due to the
possible presence of gas bubbles in a region of high electrical stress, which are the
windings and leads. These bubbles are likely to occur when the hot-spot temperature
exceeds 140ºC for a reactor with winding insulation moisture content above 2%. This
critical temperature will decrease as the moisture concentration increases.
Figure 25 Equilibrium chart relating water vapor pressure over oil to water concentration in insulation
(kraft) paper vs. temperature.
The risk with excessive water in insulation paper can be mitigated by using an on-line
monitoring system with algorithms to determine water content in paper and bubbling
temperature, so as to issue warnings when the reactor is close to a dangerous condition,
before a turn-turn fault happens.
42
4.5 Bushing failure
Overvoltages due to lightning impulses or even due to the reactor switching can bring
about very high dielectric stresses to the reactor bushings. Specifically in case of
externally generated overvoltages, the bushings will be the first ones to suffer the stress.
This fact can lead to bushings insulation deterioration, which ultimately would cause a
phase-ground fault with severe damages to the reactor itself or even to neighbor devices
due to porcelain shards being thrown. This is a severe risk also to people working close to
the equipment.
43
5 SHUNT REACTOR PROTECTIONS
The main principal hazards to a shunt reactor are:

 Bushing and insulation failures
 Turn to turn faults within a winding
 Miscellaneous failures such as low oil, loss of cooling etc.)
Various types of protection functions are used to detect these failures and they include
both electrical and non electrical protection. Main focus in this section is protection for
internal electrical faults in the reactor and external faults or conditions in the power
system that may have an impact on the shunt reactor, [ref 6], [ref 8], [ref 22].
Miscellaneous fault conditions (e.g. thermal overload) are addressed in chapter 6
“Monitoring”.
General protection requirements

The following protection requirements are common for all types of shunt reactors:
 Protect for internal faults to ground or to alternate phases.
 Protect for internal partial winding or turn-to-turn faults.
 Prevent unnecessary false tripping of the reactor during switching operations and
external ground faults.
 Provide adequate breaker failure protection and pole discrepancy protection.
Additionally, for oil immersed reactors:

 Prevent rupture of reactor tank during internal fault.
Shunt reactor arrangements

The protection schemes for a shunt reactor are strongly dependent on the design and
configuration.
Three typical scenarios are thus discussed in this section. They are:
 Protection for bus connected oil immersed reactors
 Protection for line connected oil immersed reactors
 Protection for dry-type reactors connected to the tertiary winding of a transformer.
For each scenario the most common complete protection schemes are discussed.
44
Bus connected Line connected Transformer tertial
Shunt reactor Shunt reactors connected Shunt reactor
52
52 52 52
52
52
Figure 26 Typical shunt reactor connections
5.1 Protection for bus connected oil immersed shunt reactors

Oil immersed shunt reactors do not differ much from power transformers at no load. Also
the protections applied are to a large degree the same as for power transformers. There is
however some distinct differences between these two equipments which must also be
considered when selecting the protection scheme for a shunt reactor. The shunt reactor
configuration considered in this section is wye-connected with solidly grounded neutral
which is a typical configuration for bus connected oil immersed shunt reactors.
Protection devices commonly applied to oil immersed shunt reactors

According to the questionnaire issued by the WG B5.37 these protection devices are
commonly applied to bus connected oil immersed shunt reactors:
 Reactor phase differential protection (87R / ∆I)

 Restricted earth fault protection (87N / ∆I)
 Instantaneous overcurrent protection (50 / I>>)
 Time delayed overcurrent protection (51 / I>, t)
 Time delayed earth fault overcurrent protection (51N / IE >, t)
 Distance protection (21 / Z<)
 Tank leakage protection
 Breaker failure protection (51BF)
 Special schemes for turn to turn faults
 Pole discrepancy protection
 Buchholz protection / Sudden pressure relay (63)
45
These protection functions are described below with application notes.
Regarding redundancy, most utilities reported that their protection schemes are separated
in two redundant systems using separate DC supplies, CT windings and trip coils.
5.1.1 Differential protection (87R / ∆I)

Differential protection is the most commonly applied protection function for this type of
reactors. The function offers an instantaneous protection for internal phase to phase and
phase to ground faults. Internal in this context means faults located in the area between
the CTs on either side of the reactor.
The differential protection may be of the low impedance or high impedance type.
5.1.1.1 Low impedance phase differential protection – basic principles

The sum of currents to a node must, according to Kirchhoff’s current law, be zero. The
differential protection utilizes this fact and is based on comparison of the currents on all
sides of the protected object. The sum of currents flowing into an object such as a shunt
reactor is zero except when there is a fault in the object (or within the zone limited by the
CT’s). The principle of current summation is illustrated in the Figure 27 below. Please
note that primary currents flowing in direction towards the object are regarded as
positive.
Shunt Reactor Shunt Reactor

(+) ISIDE 1 (single phase) (-) ISIDE 2 (+) ISIDE 1 (single phase) (+) ISIDE 2
Iside 1 Iside 2 Iside 1 Iside 2
Restraint Restraint
Operating IDIFF = Operating IDIFF =

Iside 1 +(-Iside 2) = 0 Iside 1+Iside 2 = 2 x I
Figure 27 Principle of low impedance phase differential protection
As illustrated in the Figure 27 this principle allows a clear discrimination between

external and internal faults and thus no time delay is necessary for selective tripping.
46
In principle an instantaneous overcurrent relay connected to measure the sum of currents
(current through operating coil in the figure above) may be used as a differential
protection.
Under certain operating conditions the CT’s secondary current may not represent exactly
the current flow on the primary side. Under these conditions Kirchhoff’s current law is
not satisfied and may lead to false differential currents.
In order to not operate inadvertently under such conditions a simple differential
protection purely based on measurement of the differential current would have to be set
less sensitive.
A more sophisticated solution to this problem is the introduction of a stabilizing quantity

(normally known as biasing or restraint current) in the differential protection. The
measured stabilizing quantity is used to desensitize the relay for conditions where large
currents are flowing in or out of the protected zone. Differential protections are normally
equipped with this feature and are often referred to as percentage differential- or biased
differential protection. As indicated in the Figure 28 above the restraint and differential
currents may be measured via dedicated measuring inputs. This is true for
electromechanical and solid state relays. The name restraint current comes from the
electromechanical relay design, where this current produced a restraint torque on the
moving disc, while the differential current produced the operating torque.
In numerical low impedance diff protections the differential and restraint quantities are
calculated from the measured phase currents.
Idiff
(p.u.) Fault characteristic
4
3
Operate / Trip area
1 Restraint / Blocking area
1 2 3 4 5 6 7 8 Irestraint
(p.u.)
Figure 28 Tripping characteristic - Example
The relay manufacturers are forming the restraint quantity in various ways. Some of the
variants are:
1. I Restraint  I side1  I side2 see Figure 27 above.
2. I Restraint  I side1  I side2
3. I Restraint   I side1  I side2  / 2
47
4. I Restraint  MAX I x where I x is the highest phase current
In the first case (mainly used in analog protections for two terminal objects) the resulting
restraint current will be zero for internal faults and twice the through flowing current for
external fault. In the most of the other variants where the absolute values of the phase
currents are used as restraint an internal fault are expected to follow approximately the
fault characteristic shown in the Figure 28 above.
Common for all variants is that an external fault with large magnitude currents flowing
through the protected object the resulting diff and restraint quantity should be a point
along the restraint axis. At the same time the protection under such conditions requires a
larger differential current to reach the tripping area and thus allowing a certain amount of
CT error.
There is no uniform term for the quantity used to stabilize the differential protection. The
following terms are often used:
 Stabilizing current (Istab)
 Bias current (Ibias)
 Restraint current (I restr.)
In addition to the current restraint many differential protections also offer a 2nd harmonic
restraint feature. This is used to detect inrush currents and desensitize the differential
protection during these conditions. This feature which is especially useful in power
transformer applications is based on the fact that 2nd harmonic content in the differential
current are present during inrush conditions but normally not existing during fault
conditions. In numerical differential protection high level of 2nd harmonic normally leads
to a complete blocking of the restrained differential function.
Internal faults with high magnitude fault currents may possibly result in heavy CT
saturation and thereby produce false 2nd harmonic content in the differential current. In
worst case this might lead to an undesired blocking of the restrained differential function.
Numerical differential protection using 2nd harmonic blocking is therefore normally also
equipped with an additional high level unrestrained diff function. Purpose of this function
is to offer secure and fast tripping for such internal faults with high magnitude fault
currents.
5.1.1.2 High impedance phase differential protection – basic principles

High impedance differential protections are also in widespread use as differential
protection on shunt reactors. This is another variant of the current comparison principle.
The functional principle is that the relay, as the name indicates, are of high ohmic type
and responds to the resulting voltage across the differential junction points.
48
(+) ISIDE 1 (single phase) (-) ISIDE 2 (+) ISIDE 1 (single phase) (+) ISIDE 2
Iside 1 Iside 2
R R
MOV MOV
87 87
(∆I) (∆I)
Figure 29 Principle of high impedance phase differential protection
With reference to the Figure 29 above the principle is as follows:

For external faults the voltage across the differential junction point will be approximately
zero. For internal faults the differential currents will result in a significant voltage
increase over the junction points and the relay will operate. The relay itself is a quite
simple and robust construction and operates in high speed when a set voltage or current
level is exceeded. A voltage dependent resistor (varistor) is normally added to the circuit
in order to protect the differential relay from damaging high voltage levels during internal
faults.
The high impedance differential relays are generally regarded as stable against in
adverted operation during external faults. This is due to the fact that the relay impedance
is much higher than the impedance of a saturated CT. As illustrated in the Figure 30
below the false differential current caused by a saturated CT causes the false differential
current to mainly circulate through the secondary of the saturated CT and the relays
operating level should not be exceeded.
49
Shunt Reactor
(+) ISIDE 1 (single phase) (-) ISIDE 2
Iside 1
Iside 1 Saturated CT
Iside 1
MOV
87
(∆I)
Figure 30 High impedance differential protection - stabilizing principle
The proper operation of this scheme is to a large degree is dependent on using dedicated
CT’s and choosing the correct combination of relay impedance and setting. The
CTs are required to have low secondary leakage impedance and the same ratio.
It is also obvious that if a CT is short circuited, this will not be detected by the relay and
lead to a permanent blocking of the function.
5.1.1.3 Application notes differential protection

Applied to an oil immersed shunt reactor with directly grounded neutral the differential
protection offers fast detection and tripping in case of internal phase to phase or phase to
ground faults. Most of the possible fault localizations in the winding are expected to be
covered by this protection function. Simulation results has shown that even the currents
during a winding to ground fault 1% from the neutral could be easily detected by the
differential protection (i.e. provided adequate protection sensitivity), see figure Figure 31.
However internal winding turn-to-turn faults will however not be detected by the
differential protection.
50
Figure 31 Differential currents of a 1% turn to ground fault in phase L1 at the neutral point of a 150 MVA,
220 kV, 50Hz shunt reactor
Generally the type of differential protections devices used for shunt reactors are the same
as for used for power transformers. Compared to power transformers the application on
shunt reactors are a lot easier.
 No need for vector group correction
 No need for zero sequence current elimination
 The inrush currents are measured on both sides and should theoretically not be
seen by the protection as a differential current.
 The magnitude of the through fault currents during external faults are relatively
small (i.e. during external ground fault approximately 1.0 p.u.).
There is however some possible challenges related to phase differential protections on

shunt reactors that must be addressed.
Switching in conditions
When energizing an iron core shunt reactor the actual peak value of the current might rise
to a value between 3 and 5.5 times rated current. The inrush currents in the three phases
will experience different degree of DC-offset. Due to the long DC-Time constant of the
shunt reactor the time to steady state condition may be fairly long and in the order of
seconds. Under such conditions the operating point of the magnetic flux in the CT’s will
increase in the same direction as the DC-component. This situation persists as long as
there is a DC-component in the primary current. A result of this may be a time delayed
saturation of the CT.
Specifically for low impedance differential protection connected to closed-core CT’s, this
time delayed saturation may cause undesirable protection operation. The fundamental
51
value of the primary currents may in this situation be close to rated current of the reactor.
The differential protection then operates in the sensitive part of the tripping characteristic.
Even a small differential current may in this situation lead to an incorrect trip by the
differential protection.
Some utilities always use CT’s of the same make and with identical ratings on both sides
of the shunt reactors. The philosophy is to avoid uneven saturation and thus prevent
significant values of false differential currents. This is not a 100% fail safe practice, but
reports indicate that it has been quite successfully applied to shunt reactors. One
important reason for this is the fact that in shunt reactor applications the CT’s on the
terminal- and neutral-side both suffer under the same conditions.
Other measures to avoid unwanted tripping of low impedance differential protection
under these conditions might be:
 Use of 2nd harmonic blocking/restraint feature in combination with cross blocking

functionality
 Use of adaptive DC-biasing
The 2nd harmonic restraint feature is often available in low impedance differential
protection applied to shunt reactors. This may offer additional protection improvements
in some cases.
For numerical low impedance differential protection a setting of the 2nd harmonic
blocking function as low as 10%, may prevent the restrained differential function from
undesired operation during reactor energizing.
Additional security against undesired operation may be achieved by activating the so
called “cross blocking function”. This function blocks the differential function in all
phases for a limited time as long as the 2nd harmonic threshold is exceeded in one of the
individual reactor phases.
As these settings increase the possibility for an unwanted blocking during real internal
fault conditions, it is advisable to use it in combination with the unrestrained differential
stage which is available in most numerical low impedance differential protections.
In modern protection applications, the introductions of adaptive DC-biasing offer a

desirable operational feature that enhance security buy stabilizing the differential
protection under such conditions. This new protection feature is described in chapter 7
“New solutions offered by numerical relays”.
CT ratio versus level of short circuit current

Generally for shunt reactor installations low CT ratios are often selected due to its low
operating current during normal operating conditions. The reactors are however often
connected to busbars or lines with very high fault levels. As a result the magnitude of the
currents during internal reactor faults may reach such high levels that the CT cores suffer
severe saturation. For low impedance differential protections the resulting distorted
differential currents may contain a large degree of 2nd harmonic current and thus cause a
blocking of the restrained differential function. Selecting a higher CT ratio may improve
the situation in some instances. For numerical differential protection the high level
52
unrestrained diff function, as described earlier, should also be used in order to secure a
fast fault clearance (pick up).
This high level stage works totally independent of the 2nd harmonic blocking or other
restraining quantities. Generally it also has quite low requirements to the CT
dimensioning. Dependable protection operation should in most cases be guaranteed as
long as the CT’s steady state saturation current exceeds the protection setting level. The
actual protections CT requirements must nevertheless be checked against the relay
manufacturer’s recommendations.
Fault characteristic
Idiff
(p.u.)
3 Operate area Operate area

High level Restrained & Unrestrained Only unrestrained
unrestr.
2
Differential
Operate area
Low level 1 Only restrained
Restraint / Blocking area
restrained
differential
1 2 3 4 5 6 7 8 9 10 11 Irestraint
(p.u.)
Figure 32 Example - Tripping characteristic for a two stage numerical low impedance differential
protection applied to a shunt reactor.
For a shunt reactor application it should be possible to choose a setting of the high level
unrestrained diff function as low as 200% of rated reactor current [ref 7]. This relatively
sensitive setting is possible due to the fact that:
 There will be no high level through fault currents for external faults.
 The possible false differential currents caused by temporary CT saturation during
inrush conditions will be decisive for the setting. These inrush currents are of
relatively moderate level.
 The numerical differential protection normally utilizes DFT filter technique of the
input current which effectively suppresses the dc component.
53
Figure 33 Shunt reactors inrush current in phase C with fully DC offset – response of two different types of
digital filters.
5.1.2 Restricted Earth Fault protection (87N / ∆I)

Restricted earth fault protections are commonly used on grounded shunt reactors to
successfully detect internal phase to ground faults. This protection is also based on the
current comparison principle. In contrast to the phase differential protection this scheme
exclusively compares the residual currents on the terminal side with the current in the
neutral to ground connection. The principle should be able to clearly identify internal
phase to ground faults from external faults. The protection function is also named ground
differential or restricted ground fault. In this section the term Restricted Earth Fault
(REF) is used.
5.1.2.1 Low impedance Restricted Earth Fault protection- basic

principles
A zero sequence overcurrent relay connected as illustrated in the Figure 34 below may be
used as a simple form of restricted earth fault protection. During normal operating
conditions or during external faults practically no zero sequence current will flow though
the relay.
In the low impedance principle the CT’s may be of different ratio and the secondary
circuits may be shared with other protections. In case of different CT ratio auxiliary CTs
are used if the protection is of analog type. In numerical protection this ratio matching is
performed by the software.
54
Io (A) Io (A)
Io (B) I0 (a)
Io (B)
I0 (b)
Io (C) Io (C)
I0 (c)
3I0 Phase relays
3I0'
3I0'
87N (∆I)
Figure 34 Low impedance restricted earth fault during external fault condition
I0 (A) I0 (A)
I0 (B) I0 (a)
I0 (B)
I0 (b)
I0 (C) I0 (A+B) + ext.
3I0 Phase relays
3I0'
87N (∆I)
Figure 35 Low impedance restricted earth fault during internal fault condition
As described earlier there is a possibility for saturation of the phase current CTs during
reactor switch in conditions due to the long DC-time constant of the reactor. In this
55
situation the REF protection will see a residual current from the terminal side which is
not present on the neutral side. A REF scheme based on a simple zero sequence over
current protection as described in Figure 35 above may in this situation operate
unnecessarily.
A solution to this problem is to use dedicated restricted earth fault protection with
restraint feature. Various methods are used to form the restraint (biasing) quantity in
dedicated low impedance REF protection. Some of the common methods are:
 Using the residual current
 Using the maximum phase current.
5.1.2.2 High impedance Restricted Earth Fault protection- basic

principles
High impedance REF protections are also in widespread use on shunt reactors, see Figure
36. As for low impedance REF protection the high impedance REF compares the residual
currents from the CT’s on the terminal side with the currents from the neutral side CT.
For this type of protection to work correctly, CTs used in all connected windings must be
of the same ratio. Dedicated CTs must be used as this type of relay has a high burden.
Io (A) Io (A)
Io (B) I0 (a)
Io (B)
I0 (b)
Io (C) Io (C)
3I0 I0 (c)
3I0'
3I0'
87N (∆I)
Figure 36 High impedance restricted earth fault during external fault condition
The high impedance REF protection detects the voltage that develops across the
differential junction points. During normal load conditions or external faults there will be
56
approximately no voltage applied to the REF protection. If there however is an internal
fault the differential currents will result in a significant voltage increase over the junction
points and the relay will operate. The relay itself is a quite robust construction and
operates in high speed when a set voltage or current threshold is exceeded. A voltage
dependent resistor (varistor) may be added to the circuit in order to protect the differential
relay from damaging high voltage levels during internal faults, pay attention for the
selection of varistor recommended by the manufacturer.
As described for the high impedance phase differential protection this principle will be is
quite stable against CT saturation during external phase to ground faults because the
resulting false differential current mainly will flow through the low impedance of the
saturated CT core.
5.1.2.3 Application notes REF protection

Applied to an oil immersed shunt reactor with directly grounded neutral the REF
protection offers sensitive and fast detection and tripping in case of internal phase to
ground faults. Most of the possible fault localizations in the winding are expected to be
covered by this protection function.
There is however some possible challenges related to REF protections on shunt reactors
that must be addressed.
During energizing of the shunt reactor there is a possibility of saturation of CTs on the
terminal side see Figure 37. In this situation the REF protection will see a residual current
from the terminal side which is not present on the neutral side. Especially low impedance
REF protection may operate unnecessary under such conditions. When using low
impedance REF it is therefore strongly recommended to use a protection with restraint
feature. The restraint quantity should be formed from the residual current from the
terminal side CTs or from phase current.
57
Saturated CT
IA IA
IB IB
Ib
IC IC
Ic
Ib + Ic
87N (∆I)
Operate Ib + Ic Restr.
Figure 37 Restrained low impedance restricted earth fault during energizing and one phase CT saturated
5.1.3 Phase overcurrent protection (50 and 51 / I>> and I>, t)

Phase overcurrent protection (Figure 38) is an inexpensive, simple and reliable device for
detection of phase to phase short circuits on the terminal side of the shunt reactor. On bus
connected oil immersed shunt reactors it is commonly used to supplement the differential
scheme for high fault current levels depending on the sizing of the shunt reactor.
As the readers of this document most probably are familiar with the overcurrent
protection, the basic principle is not described here. The focus in this chapter is the
application of overcurrent protection on shunt reactors.
58
52
I >> I >, t
50 51
Shunt
reactor
Figure 38 Terminal side connected phase overcurrent protection
Overcurrent protections applied to shunt reactors are facing two different considerations:
 The protection must allow a certain degree of current above rated current. An
overload of the shunt reactor can only be caused by increased system voltage. It is
exactly under these conditions the system needs to have the shunt reactors
energized.
 The fault currents measured by the overcurrent protection on terminal-side will be

reduced to values close to nominal current for fault locations close to the neutral.
The purpose is to protect as much of the winding without risk for unwanted trip.
The phase overcurrent scheme for the shunt reactor may consist of a 51 type (time
delayed) protection or a combination of 51 and 50 (instantaneous overcurrent).
5.1.3.1 Application notes phase overcurrent protection
Instantaneous overcurrent (50)

The overcurrent threshold for the instantaneous overcurrent function (50) must set above
the shunt reactor inrush currents. A typical conservative setting for analog instantaneous
overcurrent protection is 6 times rated current of the shunt reactor. For modern numerical
protections using Discrete Fourier Filter (DFF) and thereby only extracts the RMS value
of the fundamental component a more sensitive setting is possible (relay manufacturers
have to be consulted). This filter effectively suppresses the DC component and higher
harmonic components in the input quantity. For instance and according to some relay
manufacturers, an adequate setting of 2.5 times rated current with a time delay of 0.1s for
this function could be applied to the (50) function of shunt reactors.
Time delayed overcurrent (51 / I>, t)
59
The overcurrent threshold for the time delayed overcurrent function (51) should account
for normal system operation. In this regard, the current threshold must be set above the
maximum reactor current as a result of temporary system overvoltage or the voltage rise
in healthy phases during a single phase to earth fault in the connected grid. This voltage
will rarely exceed 1.3 per unit in systems with solidly earthed neutral. A setting of 1.5
times rated reactor current should normally be sufficient to override these temporary
overvoltages.
5.1.4 Earth fault overcurrent protection (51N or 51G / IE>, t)

According to the result of the questionnaire, time delayed ground overcurrent protections
are quite commonly applied to oil immersed shunt reactors, mainly as backup protection
for phase to ground faults.
During a turn to turn fault in a grounded reactor there will also be zero sequence currents
present. The earth fault overcurrent protection therefore also offers some degree of
protection for this type of fault.
The earth fault overcurrent protections can be provided according to Figure 39 a) and b)
below. The connection method applied has significant impact on the sensitivity of the
protection and is described further in the application section.
Earth fault overcurrent protections are generally immune to off-nominal frequency

currents. 3rd harmonic order currents and its multiples are effectively filtered by the relay
device.
52 52
IE >, t
51N
Shunt Shunt
reactor reactor
IE >, t
51G
a) b)
60
Figure 39 Earth fault overcurrent protection alternatively connected to CT in the neutral
to ground connection or phase CT’s on terminal side.
5.1.4.1 Application notes – Earth fault overcurrent protection
Connection of earth fault overcurrent protection to reactor neutral- or terminal-side

During a phase to ground fault located close to the neutral point, the currents on the
terminal side will be practically the same as before the fault. However due to the so
called transformer effect the current in the common neutral point (and in neutral side of
the affected phase) will increase to a large magnitude.
This phase to earth fault would therefore be easily detected by an earth fault protection
located on the neutral point side but not by the terminal side earth fault protection.
During a phase to earth fault on the terminal side of the reactor (i.e. on bushing), the fault
current measured on the terminal side will have a large magnitude while the earth fault
current in reactor neutral point will have a moderate magnitude. In this situation, one of
the reactor windings is practically short-circuited. The resulting unbalance current that
appears in the neutral point will typically have a value of approximately 1 per unit.
This phase to earth fault would therefore be easily detected both by an earth fault
protection located on the terminal side and also by the earth fault protection located on
the neutral side. If both protections are initiated, then the location of the fault will be
easier to determine.
Conclusion: Earth fault overcurrent protection located on the neutral side is the preferred
method for the following reasons:
 It provides adequate and sensitive ground fault detection for faults near the neutral
side of the reactor.
 It provides adequate back up coverage in case of failure of primary protection
(differential)
 The use of 51N function on both terminal side and the neutral side of the reactor
will enhance the protection application for full fault coverage of the reactor
winding and fault location.
The earth fault overcurrent protection is affected by shunt reactor inrush currents. If the
relay is set too sensitive, it might yield to undesired tripping under such conditions. Thus
the relay pickup current and associated time delay element must be accounted for by
setting engineers. In addition, earth fault overcurrent protection connected to the terminal
side may experience CT saturation during shunt reactor energization and adequate
settings should be applied to avoid false tripping. This issue does not apply to the Earth
fault overcurrent protection connected to the common neutral side of the reactor.
Nowadays, modern numerical protection offer unprecedented techniques and solutions
with various algorithms to detect 2nd harmonic currents and offer the possibility to
program logical equations to block and unblock various protection devices according to a
specific philosophy. For example, the 51G on the neutral side winding of the reactor may
61
be set to detect 2nd harmonic energization current and block the 51N provided on the
terminal side of the reactor. These new applications however have to be validated through
proper relay simulation tests in the laboratory.
General notes
Because the time delayed earth fault protection can be initiated for external phase to earth
faults on the system, it is important to use sufficient time delays to override the external
faults.
5.1.5 Distance protection (21 / Z<)
Applied to oil immersed shunt reactors, the distance protection is mainly used in Asian
countries and the US as a backup protection. They normally consist of a single-zone
underimpedance relay looking into the reactor from the terminal side as shown in the
Figure 40 below. The distance relay may provide instantaneous protection for phase-
phase, phase-earth and to some extent also turn-to-turn faults. Similar to the overcurrent
protection the distance protection will generally not detect short circuits close to the
neutral point. It is important to note that the correct setting of the distance protection is of
concern to avoid operation during switching conditions.
52
VT
Z<
CT 21
Shunt
reactor
Figure 40 Terminal side connected distance protection
5.1.5.1 Application notes distance protection

The settings of the underimpedance-zone should account for:
 Possible system dynamic temporary overvoltage
 Shunt reactor inrush current (reduced reactor impedance)
 Above the turn to turn shorted reactance.
62
Setting of impedance zone reach below reactance during overvoltage conditions
Temporary overvoltage in the system may cause the reactor to operate in the non linear
area, see Figure 6 of the magnetizing characteristic (saturated part). The current in the
non linear area of the B-H curve will increase and will necessarily affect the apparent
reactor reactance as measured by the distance protection. This reactance will naturally be
lower than the rated reactor reactance XR (Ohm/phase primary) given by eq. 39:
U2 eq. 39
XR 
Q
Where U is the line-line voltage and Q is the 3phase VAR rating of the reactor.
An approach to determine the saturated reactor reactance will be to first select the
maximum expected short time overvoltage in the system which may be impacted by the
distance protection (time response of the distance protection). The actual reactor current
at this voltage level may then be determined from the reactor magnetizing characteristic.
The reactor reactance XMIN (Ohm/phase primary) under this condition is then
approximately given by eq. 40:
U  XR eq. 40
X MIN   MAX  
 UR  I MAX ( pu )
Where UMAX is maximum expected system overvoltage, UR is rated system voltage and
IMAX is the per unit reactor current at UMAX.
E.g. Introduction of an additional safety margin leaves a resulting setting of the reactive
zone reach to (0.85-0.9) * XMIN.
Setting considerations accounting inrush conditions of reactors

During extreme inrush conditions with fully DC-offset, the shunt reactor might saturate,
resulting in significant reduction of the apparent reactor impedance. The zone reach of
the distance protection must be chosen to a value below this impedance with an
additional safety margin. The saturated impedance will have to be calculated or
determined by tests. Generally the impedance measured by the distance protection may
be as low as 70% of shunt reactor rated impedance during switching.
In addition to the possible saturation of the reactor also the CT’s might suffer saturation
during the inrush conditions due to the slowly decaying DC-component.
From the working group survey [ref 23], some countries that apply distance protection
have indicated that:
63
 Distance protections for shunt reactors are specified with a maximum impedance
setting to cover at least 60% of rated reactor impedance.
 Undesired operation during switching conditions has been observed but the
tendency seems to be reduced by numerical distance protection due to modern
filtering algorithms.
Setting of impedance above the turn to turn shorted reactance

A short circuit between winding turns of the same phase (turn-to-turn fault) will result in
decreased apparent phase impedance. The distance protection may detect such conditions.
However the sensitivity may not be sufficient to detect all turn-to-turn faults. As
discussed above a zone reach setting of approximately 60% of rated reactor impedance
may be necessary in order to be stable against inrush conditions. A turn-to-turn fault that
only affects one or a few turns will then not be detected by zone 1 distance protection
function.
If time delayed zone 2 element is applied to the under impedance protection scheme, the
sensitivity of the protection for turn-turn faults will be enhanced. However, careful
measures must be applied to secure the protection during shunt reactor energization. This
zone 2 element should be blocked during inrush conditions or time delayed sufficiently to
override the inrush phenomena.
According to eq.17 in section 2 it is seen that the inductance is proportional to the square
to the number of turns of the winding. It is then appropriate to conclude that a short
circuit involving only 5% of the total turns would reduce the apparent impedance of the
reactor by approximately 10% of its rated value. For example, the apparent impedance
for a 5% turn-turn fault is approximately (0.95)2*XL and by neglecting the magnetizing
impedance and the apparent leakage impedance of the reactor. Similarly a 20% turn-turn
short circuit within the reactor will yield an apparent impedance of 0.64*XL.
This relationship is illustrated in the Figure 41 below. The figure also visualizes the
strong relationship between the current in the faulted phase and the number of shorted
turns.
64
Figure 41 Current and apparent reactance in a turn-to-turn fault affected phase for a grounded 200MVAr,
420kV, 50Hz shunt reactor.
As the Figure 41 above shows, the distance protection scheme does not offer a very
sensitive detection of turn to turn faults. Its advantage compared to other current based
schemes is the ability to trip instantaneously for faults within zone 1. However, when set
to typically 60% of the rated shunt reactor impedance, more than 20% of the total turns
have to be shorted in order to achieve this fast tripping.
For comparison, also the amount of earth fault current 3I0 and zero sequence current
during turn to turn faults are shown in the Figure 42 below. The fact that the 3I0 current
increases rapidly when turns are shorted are utilized in some special turn to turn
protection schemes.
65
Figure 42 Negative sequence current and residual current 3I0 in a turn-to-turn fault affected 200MVAr,
420kV, 50Hz shunt reactor.
5.1.6 Protection schemes dedicated to the detection of turn to

turn faults
Turns to turn faults are not only the most common failures in shunt reactors but it is the
most challenging fault type to detect. In oil immersed shunt reactors, the
Buchholz/Pressure relief protection (63) represents a quite dependable protection for this
type of fault. Many utilities consider the 63 protection element as a main protection
function due to the limitation from electrical protection schemes.
New methods for detecting turn to turn faults that benefit from the possibilities offered by
numerical protection have been developed in recent years. Some of these methods are
presented in chapter 7.
5.1.6.1 Earth fault overcurrent protection controlled by directional zero

sequence relay
Use of directional earth fault protection controlled by a directional relay is applied as turn
to turn fault protection in North America. The scheme uses a zero sequence directional
relay to control (release) an earth fault overcurrent protection in the neutral. The blue
arrow in the figure below indicates the forward operating direction. With the indicated
direction it is apparent that the scheme might not operate for phase to earth faults. This is
of no concern as this scheme is exclusively provided for detection of internal winding
turn to turn faults.
66
52
VT
Shunt
reactor
CT 50G 67N
Figure 43 Turn to turn fault protection based on zero sequence directional control
As the Figure 43 illustrates, both protections are connected to the CT in the neutral to
earth connection and not to a residual connection on the terminal side. This minimizes the
risk of undesirable operation of the earth fault protection or the directional relay due to
unequal saturation of phase CT’s during inrush conditions.
Due to low zero sequence voltages resulting from turn-turn reactor faults, the 67N
function may have difficulty achieving adequate protection due to inadequate polarizing
voltage. This problem can be overcome by using a directional relay that has the
polarizing voltage reinforced by using the zero sequence current passing through an
impedance. The impedance value used should be less than the impedance of the shunt
reactor so as to prevent incorrect directional operation for unbalanced faults external to
the reactor. In addition, a small time delay in the 67N function (a few hundred
milliseconds), increases security of the scheme.
5.1.6.2 Earth fault overcurrent controlled by directional negative

sequence relay
This is a variant of the scheme above and also exclusively used for the detection of
winding turn to turn faults. The scheme is using a negative-sequence directional relay to
control (release) an earth fault overcurrent protection in the neutral as seen in Figure 44.
The blue arrow indicates the forward operating direction. As with the zero-sequence
scheme above, adequate sensitivity can be achieved by reinforcement of the negative
sequence polarizing voltage. Again this additional polarizing voltage is produced by
passing some relay operating current through an impedance of a value less than the
impedance of the shunt reactor.
67
52
VT
67Q
Shunt
reactor
CT 50G
Figure 44 Turn to turn fault protection based on negative sequence directional control
During energization there is a possibility for apparent high levels of negative sequence
currents due to unequal saturation of the phase CT’s. It is therefore necessary to block
this turn to turn protection until after the DC component of the phase currents have
completely decayed.
5.1.6.3 Split phase protection

Some shunt reactors are designed with two parallel windings per phase, see Figure 45.
With this design it is possible to apply a simple, sensitive and secure turn to turn fault
protection, the so called split phase protection.
In this design the two neutral ends per phase are brought together in opposite direction
through a CT. The protection itself is a simple and sensitive 3 phase overcurrent
protection. A sensitive setting is possible due to the fact that only unbalances between the
two parallel winding causes a current flow on the CT’s secondary side. A typical pick up
threshold for the overcurrent protection is approximately 2.5% of reactor rated current
[ref 8].
68
52 52 52
Shunt
reactor
I>, t I>, t I>, t

51 51 51
Figure 45 Three phases split winding protection
5.1.7 Tank protection

The tank protection is widely used in France and some African countries to detect
insulation breakdown between a winding or terminal and the tank, see Figure 46.
It is based on establishing just one single connection path between the reactor tank and
earth. This single connection runs though a CT and current flow is detected by an
instantaneous overcurrent protection.
Figure 46 Tank protection
A precondition for correct operation is that all common earth paths from tank and
69
connected metallic parts (cable shields, oil conservator, pipes for fire protection etc.) are
insulated in such a way that the connection through the CT is the only path towards earth.
This measuring principle provides fast unit protection based on a simple arrangement.
5.1.7.1 Application notes tank protection

In case of an external phase to ground fault the tank protection should not operate.
Certain amounts of undesired measuring current may however apply under such
circumstances due to the potential rise towards earth caused by the earth fault current
from the neutral (through Z0 and earth contact resistance). If the insulation level of the
tank system is not sufficient (leakage resistance caused by snow coverage etc.) there will
be an undesired amount of current flowing through the measuring CT.
The ratio of earth contact resistance to tank leakage resistance must therefore be kept low
regardless of atmospheric conditions. This ratio determines the possible sensitivity of the
tank protection scheme.
5.1.8 Breaker failure protection (50BF)

The circuit-breaker is the last link in the chain of protection; usually it is not duplicated
because of the high cost. If the circuit-breaker fails to operate, a back-up system is
needed which is able to clear the fault by tripping the surrounding circuit-breakers on the
same busbar. The protection scheme which performs this function is called breaker
failure protection (BFP), see Figure 47.
A typical BFP scheme monitors the time which expires between the generation of a
tripping signal by shunt reactor protection and the discontinuation of the fault current.
The BF current threshold must be set below the rated shunt reactor current.
70
Retrip timer
RTPU CB
Start BFP ph L1 Retrip
0
BF timer
BFPU
L1 current
+ 0 Trip
measured & ≥1 BFP
BF Current -
Threshold
Logic ph L2
Logic ph L3
Figure 47 Basic breaker failure protection scheme (logic phase L2 and logic phase L3 are similar to logic
phase L1 as detailed in the figure).
5.1.8.1 Application notes breaker failure protection (BFP)

The fault detector may use several criteria to detect breaker failure. The most obvious
criterion is that the phase current is higher than a predetermined value (current threshold
in Figure 47). However the BFP function may not be initiated if the current initiating the
BFP is too low. The operation of a Buchholz or other technological protection can occur
when current may be far below the operate value of the fault detector. The fault detector
may combine the current condition with information from auxiliary switches of the
circuit breaker to detect breaker failure accompanied with low currents. BFP scheme
using only auxiliary switches may create reliability concerns over a scheme that uses both
detection principles (current and CB contacts), see also [ref 9].
Usually breaker failure protection have two steps: first step is to retrip its own circuit
breaker, while the second step is provided with a longer time delay to trip the same and
adjacent circuit breakers.
If BFP is provided, it would minimize the damage in case of internal shunt reactor faults
in case of a BFP.
5.1.9 Pole discrepancy protection

Breaker pole discrepancy is a protection against un-logical breaker position that remains
after an opening or closing command. If an unsymmetrical condition appears, a time
delay relay will trip the circuit breaker. It is a protection for circuit breaker, see Figure
48.
Note: this is relevant for breakers with separate operating mechanism per phase.
71
A
B L+ (1) L+ (2)
C
52-A
52-B
52-C
Shunt
reactor
L- (1) L- (2)
3 phase trip 3 phase trip

(coil 1) (coil 2)
Figure 48 An example for CB pole discrepancy scheme
5.1.10 Buchholz relay (63)

For oil immersed shunt reactors with conservator tank the gas-operated Buchholz relay is
commonly applied. It is used to detect internal short circuits within the reactor tank
included turn to turn faults in the phase windings. The ability to detect turn to turn faults
is a benefit to minimize damage within the shunt reactor as it is difficult for the electrical
protection devices to detect small turn-turn faults.
Operating principle
A Buchholz relay is a standard protection fitted to oil-immersed shunt reactors which
detect all insulation breakdowns inside the shunt reactor tank, causing either the
formation of gas or surges of oil flow from the tank to the expansion vessel see Figure 49.
This applies to all phase and ground faults on the windings and to inter-turn faults. The
relay also detects losses of oil caused by leaks as well as defects such as broken
conductors and corroded or otherwise bad connections.
The operating principle of the Buchholz relay is based on the fact that firstly the pressure
of gas in the upper part of the shunt reactor tank increases due to the chemical
decomposition of the oil and/ or the combustion of solid insulating materials and
secondly that massive gas development gives to a surge of oil towards the expansion
vessel.
72
A Buchholz relay is installed in the pipe connecting the shunt reactor tank and the
expansion vessel. It generally comprises two floats one above the other. The upper one
signals the slow collection of gas and the lower float which in the event of a surge of oil
operates a mercury contact to trip the shunt reactor.
REACTOR TANK
Figure 49 Buchholz relay mounting arrangement
5.1.11 Sudden pressure relay (63)

The sudden pressure relay is commonly applied to oil immersed shunt reactors without
conservator tank, but can be used with any type of reactor, see Figure 50. As for the
Buchholz relay it responds to internal faults within the reactor tank and is used to detect
internal short circuits within the reactor tank including turn to turn faults.
Operating principle
The simplest form of pressure relief device is the widely used fragile disc that is normally
located at the end of an oil relief pipe protruding from the top of the shunt reactor tank.
The surge of oil caused by a serious fault bursts the disc, to allow the oil to discharge
rapidly. Relieving and limiting the pressure rise avoids explosive rupture of the tank and
consequent fire risk. Outdoor oil-immersed shunt reactors are usually mounted in a
73
catchments pit to collect and contain spilt oil thereby minimizing the possibility of
pollution.
A drawback of the fragile disc is that the oil remaining in the tank is left exposed to the
atmosphere after rupture. This is avoided in a more effective device, the sudden pressure
relief valve, which opens to allow discharge of oil if the pressure exceeds a set level, but
closes below this level. If the abnormal pressure is relatively high, the valve can operate a
few suitable contacts for tripping and to extinguish the fire.
PRESSURE-RELIEF DEVICE
Figure 50 The sudden pressure relief device
5.1.11.1 Application notes Buchholz and sudden pressure relay

Sudden pressure relays provide an excellent example of the classical protection dilemma
of dependability versus security. They are dependable for internal reactor faults but have
shown a tendency to operate undesired if exposed to vibrations. Design improvements
have reduced their tendency to misoperate but have not totally eliminated the security
problem. If there is repeatedly problems with misoperation of the sudden pressure relay,
installation of a newer style relay (immunity to vibrations) may be solution to the
problem.
The Buchholz relay has a significant higher security rating than the sudden pressure
relays and this type should be the preferred solution for conservator type shunt reactors.
74
5.1.12 Typical shunt reactor protection schemes
Bus connected shunt reactor protection

In this section some different protection schemes applied to bus connected oil immersed
shunt reactors are presented. Each complete scheme reflects some utility practices as
illustrated in below, see Figure 51, Figure 52 and Figure 53.
Main 2 / Backup 52 Main 1
I>> I>, t BFP IE>, t

50 51 51BF 51N
IE>, t BFP I>, t I>>
51N 51BF 51 50
ΔI> BUS
87BB
ΔI>
Buchh./
Sudden 87R
pressure
Shunt
reactor 63 REF
87N
Figure 51 Bus connected shunt reactor protection scheme number 1
Requirements Main 1 Main 2 / Backup
Protect for internal fault ph-ph X [87R], (X) [87G, 50, 51] (X) [50, 51]
Protect for internal fault ph-earth X [87R, 87N], (X) [51N] (X) [51N]
Protect for internal winding turn to turn fault X [63]
Provide for breaker failure X [51BF] X [51BF]
X = Full protection coverage
75
(X) = Partial protection coverage
ΔI> BUS
87BB
BFP I>, t I>>
51BF 51 50
Buchh./
Sudden
pressure ΔI>
Shunt
reactor 63 87R
HI IMP
REF
87N
IE>, t
51G
Protect for internal fault ph-ph X [87R] (X) [87N, 50, 51]
Protect for internal fault ph-earth X [87R] X [87N, 51G] (X) [50,51]
Protect for internal winding turn to turn fault X [63] (X) [51G]
Provide for breaker failure X [51BF]
76
Main 2 / Backup Main 1
52
BFP ΔI> BUS
51BF 87BB
Z<
21
IE>, t IE>>, t I>, t I>>
51N 50N 51 50
Buchh./
Sudden HI IMP
pressure ΔI> REF
Shunt
reactor 63 87R 87N
Protect for internal fault ph-ph X [87R], (X) [87N] (X) [21, 50, 51]
Protect for internal fault ph-earth X [87N], (X) [87R] (X) [21, 50N, 51N]
Protect for internal winding turn to turn fault X [63] (X) [21]
Provide for breaker failure X [51BF]
77
5.2 Protection for oil immersed line connected shunt reactors
Shunt reactors are often connected to both ends of long transmission lines. The majority
of these reactors are of oil immersed type.
The reactor construction and the applied protection schemes are basically the same as
already discussed in section 5.1
Compared to the bus connected shunt reactors there are however some differences that
affect the protection schemes:
 Teleprotection becomes a vital part (remote end communication) of the protection
scheme depending on breaker arrangement or type.
 Some protections may be affected by oscillations after disconnection of the
transmission line.
 An additional reactor (fourth leg) is sometimes installed between the neutral and
ground.
A) Switched line reactors (Circuit breaker) B) Switched line reactors (Circuit switcher)
52 52 52 52
Line CB Line CB Line CB Line CB

52 52 52 52
Reactor Reactor Reactor Reactor
CB CB circuit circuit
switcher switcher
C) Permanently connected line reactors D) Switched line reactors (CB) with neutral reactor
52 52 52 52
Line CB Line CB Line CB Line CB

52 52
Reactor Reactor
CB CB
Shunt Shunt
reactor reactor
Bypass Neutral Bypass Neutral

device reactor device reactor
Figure 54 Line connected shunt reactors – typical arrangements.
78
The breaker arrangements have a large impact on the tripping schemes.
Switched line reactors with reactor circuit breakers (case A in Figure 54): The shunt
reactor protection will normally only trip its own circuit breaker. The line may still be in
service. Trip from line protection will normally not trip the reactor circuit breakers. Trip
from reactor breaker failure protection must however trip the local line CB directly and
initiate direct transfer tripping at the remote end.
Switched line reactors with reactor circuit switchers (case B in Figure 54): Depending on
the interrupting capability of the circuit switcher some of the shunt reactor protection
may be used to disconnect the circuit switcher while others like short circuit protections
must trip both line CB’s. In the last case, the communication channel becomes a vital part
of the protection scheme. As this communication channel is a vulnerable part of the
scheme it is often duplicated. In some schemes auto reclose (AR) function is used. The
correct time coordination between the breakers is then of importance. The trip-reclosing
cycle may be in the order as described below:
 A short circuit in a reactor bay trips both line CB’s
 The circuit switcher of the shunt reactor opens during the AR dead time.
 The line CB’s are successfully reclosed.
Permanently connected line reactors (case C in Figure 54):

All reactor protections must trip both line CB’s. Additionally reactor protection trip
blocks the AR function in the line bays. The communication channel for transfer tripping
is normally duplicated.
Switched line reactors with reactor circuit breakers and neutral reactor (case D in Figure
54):
The tripping scheme is practically identical to the scheme described for case A.
The (fourth) neutral to ground connected reactor is used for successful suppression of
arcs during single pole AR. The inductance and voltage rating of neutral reactor are based
on the shunt capacitance of the actual transmission line.
One disadvantage with this scheme is the fact that the neutral reactor is only energized
during phase to ground faults, and it is very difficult for an electrical protection to detect
any faults within the neutral reactor before it is called upon, this issue could be regarded
as a major challenge for protection engineers.
The bypassing device is used to remove the neutral reactor from service by shorting out
the reactor and thus removing the voltage from it. This device can be triggered by the
reactor protection.
79
5.2.1 Shunt reactor protection issues related to disconnection
of transmission lines
When disconnecting a transmission line, the line capacitances can form a parallel-
resonant circuit with the connected shunt reactors. This may cause zero- and negative
sequence oscillation currents flowing in the shunt reactors. This might lead to undesired
pick up and tripping by some of the shunt reactor protections. For example, the
combination of earth fault protection and directional relays (see 5.1.3.1 and 5.1.3.2) used
for turn to turn fault detections could be impacted.
In order to avoid undesired tripping, different types of blocking logic have been applied
to protection schemes for line connected shunt reactors. Some applied blocking logics are
described below. In addition, mutual coupling between parallel lines may cause undesired
operation by shunt reactor protection on de-energized lines.
5.2.1.1 Undesired tripping of ground overcurrent protection on

de-energized lines
Care must be taken into account for applications of ground overcurrent protection in
parallel lines, due to the mutual coupling between them.
Example of undesired ground overcurrent tripping:

Figure 55, below shows a situation where a 500 kV line reactor back-up neutral
overcurrent relay tripped when its breaker was closed in order to start the line restoration,
see Figure 56. Despite the fact that the line end was still open; the induced current due to
the mutual coupling between the parallel lines provoked the relay’s 51G to trip.
Figure 55 Example of a reactor neutral overcurrent on a 500kV power line
80
Figure 56 Breaker closed to start the line restoration
5.2.1.2 Blocking of protections applied to line connected shunt

reactors
The shunt capacitance of the transmission line forms a parallel resonant circuit with the
reactance of shunt reactors. In some cases the resonance frequency after de-energization
may be close to the fundamental frequency (50/60 Hz).
Protection functions, such as ground overcurrent protection, negative sequence protection
and distance protection can misoperate when exposed to such conditions.
Blocking of some shunt reactor protection functions may therefore be appropriate.

Blocking may be achieved using an undervoltage (dead line) scheme. When applying
such a scheme, the undervoltage settings should be chosen such that other reactor
protection (e.g. instantaneous phase overcurrent) is in place in case of a severe turn-to-
turn fault that drops the voltage enough to block.
Another blocking logic, used by some utilities for line connected shunt reactors, is shown
Figure 57. This logic is used to block certain protection functions instantaneously on de-
energization of the reactor/line and will retain the blocking function for a few seconds
after energization.
81
Trip release logic
AND
I L1 (50A)
I L2 (50B) &
I L3 (50C)
AND
Shunt reactor energized ON delay Tripping AND
AND
240 cyc released Trip local line CB
One or both ends closed
&
Local line end Open &
Remote line end Open &
Transfer trip
to remote end
Trip from 50N/51N (IE>) etc.
Figure 57 Trip release logic for inrush suppression and blocking when deenergized (line
connected shunt reactors)
5.2.2 Protection for neutral (“fourth leg”) reactor

For shunt reactors connected to lines where single phase tripping and auto-reclosing is
applied, a neutral reactor is commonly used. This reactor may be necessary to
successfully extinguish the fault arc during the single phase dead time. The shunt
capacitance of the transmission line is decisive for the proper impedance rating of the
neutral reactor.
The neutral reactor is connected between the neutral of the line phase reactor and ground.
Hence, during normal service operation there will be practically no voltage across the
reactor. This lack of voltage during normal operation causes a significant difference in
the fault conditions encountered by the neutral reactor as compared to the phase reactors.
The difference represents a challenge regarding protection functions as some fault
conditions may not be detected before unbalanced voltages appear and the neutral reactor
is called upon. Especially for electrical protection, it is almost impossible to detect faults
in the neutral reactor before the unbalanced operating conditions appear.
For oil immersed neutral reactors a limited way of early failure detection may be
achieved using low oil detection, sudden pressure, and pressure relief device. However,
for the sudden pressure or the pressure relief device, some unbalance in the system is
needed to supply the energy to generate the gas. The sudden pressure device is critical for
detecting turn to turn faults in the neutral reactor [ref 8].
A single phase differential protection is commonly applied to neutral reactors. This

protection will instantaneously detect winding to ground faults. As the service time of
the shunt reactor is very short (e.g. 0.5-2 s), the differential protection is possibly the only
electrical device that is fast and sensitive enough to detect this type of fault.
82
5.3 Protection for transformer tertiary connected shunt
reactors
For economical reasons, shunt reactors are sometimes connected to the tertiary winding
of transmission power transformer (mainly autotransformers).
52 52 52 52
52
Circuit breaker or
circuit switcher on
reactor supply side
52
Circuit breaker or
circuit switcher on
reactor neutral side
Figure 58 Shunt reactors connected to tertiary winding of autotransformers.
Figure 58 illustrates an ungrounded dry type shunt reactor from a tertiary winding of a
transformer, the following considerations apply:
 The shunt reactor is of dry-type (air core).
 The shunt reactor is connected in ungrounded wye.
 The transformer tertiary is delta connected.
 The medium voltage system connected to the transformer tertiary is ungrounded
or high resistance grounded.
Protection devices commonly applied to dry type air core shunt reactors
Typical protection devices applied to tertiary connected dry-type reactors are:
 Instantaneous overcurrent protection (50 / I>>)
 Time delayed overcurrent protection (51 / I>, t)
 Negative sequence protection (46 / I2>)
 Ground overvoltage protection (59N)
83
 Special schemes for turn to turn faults
Additionally the following protection functions may also apply:

 Phase differential protection. This can be the transformer differential (87T / ∆I)
when the shunt reactor is included in the power transformer or a separate reactor
phase differential protection.
 Breaker failure protection (51BF)
5.3.1 Phase overcurrent protection (50 and 51 / I>> and I>, t)

On dry-type tertiary connected shunt reactors, phase overcurrent relays are commonly
used as main short circuit protection.
Dry-type reactors are normally arranged as single phase units with adequate separation
between the phases. Thus phase-phase faults within the reactor are not of frequent type.
However instances have been reported where arcing from a faulted reactor has developed
into phase-phase faults.
The phase overcurrent protection is normally measuring the currents on the supply side of
the reactor but also measurement on the neutral-side is used in some applications.
5.3.1.1 Application notes - overcurrent protection for air core

shunt reactors
In the dry type shunt reactors there are no iron core that might suffer saturation during
energizing. Hence there will be no real inrush currents under these circumstances.
It is therefore possible to apply a relatively sensitive setting of the overcurrent protection
for dry type reactors.
Depending on the breaker closing instant, the initial current may however have a quite
large DC-offset. The first peak of the current with full offset will be 2√2 times rated
current.
If the DC-current is not sufficiently suppressed by the overcurrent protection, this current
must also be considered when setting the overcurrent threshold in order to avoid transient
overreach.
Modern numerical protections that effectively suppress the DC and higher harmonic
components in the input quantity are principally unaffected during energizing of these
types of air core shunt reactors. For such protection devices, the instantaneous
overcurrent function (50/ I>>) may be set to approximately 2 times rated reactor current.
The time delayed overcurrent (51/ I>,t) should be set as sensitive as possible but must
allow a certain safety margin to account for temporary overvoltages in the system.
A setting between 1.4-1.5 times rated reactor current should normally be sufficient to
override these conditions.
In addition to the short circuit currents caused by phase-phase faults, the time delayed
overcurrent function may also operate for severe turn-to-turn faults. This is due to the
increased current in the two healthy phases. If the turn-to-turn fault develops to involve
84
the entire phase winding a maximum of approximately √3 times rated current will occur
in the healthy phases.
5.3.2 Negative sequence protection (46 / I2>)

Negative sequence protection is also applied to some dry type shunt reactor installations
to detect unbalanced currents. Unbalanced currents may be caused by phase-phase faults
and to some degree reactor turn-to-turn faults. Unbalanced currents caused by open
circuit conditions in one of the phases will also be detected by the negative sequence
protection.
52
I2>
CT 46
Shunt
reactor
Figure 59 Negative sequence protection.
5.3.2.1 Application notes – negative sequence protection

The negative sequence current threshold should be set above the maximum level of
natural system voltage unbalance and shunt reactor manufacturing tolerances.
The time delay should be set to coordinate with other protection devices that operate
during unbalanced faults external to the reactor.
5.3.3 Ground overvoltage protection (59N / 3U0>)

Ground overvoltage protection is an inexpensive, simple and reliable device for detection
of phase to ground faults in the entire tertiary system. However the protection cannot
locate the ground fault and thus will not differentiate between ground faults on the shunt
reactor or other parts of the tertiary system.
85
Transformer tertiary
Tertiary Bus
L1
L2
L3
Grounding
transformer
52
Grounding
resistor
3Uo>
Shunt
reactor 59N
Ground overvoltage
protection
Figure 60 Ground overvoltage protection connected to broken delta winding of a voltage

transformer / grounding transformer.
Because the tertiary system is ungrounded or high resistance grounded, the resulting
ground fault current will be of low magnitude. In some countries it is therefore an
accepted practice to alarm and not to trip for this fault condition.
Grounding transformers are used on tertiary systems to limit the potential rise against
ground faults. In order to limit the phase to ground current to acceptable recommended
levels, resistors connected on the secondary side of grounding transformers are used.
During phase to ground faults, sufficient 3U0 quantities across the resistor will serve the
59N function.
Generally, it is recommended that the primary ground fault current caused by the resistor
should be equal to or higher than the capacitive ground fault currents in the system. This
will reduce the transient overvoltages that may occur after re-strikes in the fault arc to
safe levels [ref 14].
Also zigzag transformers with grounding resistor can be applied as an alternative way to
achieve high resistance grounding of the system.
86
If more switched reactors are connected to the tertiary bus, the active fault current
contribution from the grounding transformer (Figure 60) can be utilized by directional
zero sequence protections to selectively trip the faulty shunt reactor.
5.3.3.1 Application notes – Ground overvoltage protection

A ground fault in the reactor neutral will not cause any voltage development over the
broken delta winding and thus not be detected by the ground overvoltage protection.
On the other hand, a phase to ground fault on the terminal side of the reactor causes the
voltage across the broken delta winding to reach typically √3*ph-ph voltage.
The ground overvoltage protection should be set as sensitive as possible to cover most of
the phase winding of the reactor.
In order to achieve a sensitive setting it is imperative to use an overvoltage protection that
is tuned to the fundamental frequency. Normally, 3rd harmonic components (and its
multiples) are suppressed.
5.3.4 Special schemes for turn to turn faults in air core shunt
reactors
A turn to turn fault (fault between turns within the phase winding) is a damaging
condition to shunt reactors. Once the arc between some turns has been initiated, the fault
may develop to flashover the entire winding. If not detected, this resulting phase to
neutral fault may then also result in thermal damage of the healthy phases (which in this
situation is connected to a voltage √3 * nominal voltage).
Even one shorted turn will cause a high magnitude circulating current in the fault
location. However, seen from the outside, the phase currents and the voltages can be of
the same order of magnitude as could be expected during normal service.
The ultimate aim of the turn to turn protection is to be able to detect a fault as a result of
one shorted turn in the winding.
This is a real challenge and calls for dedicated protection schemes for turn to turn faults.
Commonly applied schemes for turn to turn fault protection in dry type shunt reactors
are:
 Split phase protection scheme (also described for oil immersed shunt reactors -
section 5.1.3.3) which should be used when the reactor are arranged using
multiple parallel circuits per coil.
 Voltage unbalance scheme which is described below.
87
5.3.4.1 Turn to turn fault detection – Split phase scheme
If the dry type shunt reactors are arranged in two groups as shown in the figure below, it
is easy to use the so called split phase differential principle to achieve a sensitive and fast
turn to turn fault protection.
The split phase protection is an overcurrent protection monitoring the single current path
between the two neutral points. During normal healthy operation this current is very
small, mainly given by the fixed reactor manufacturing tolerances.
Regarding sensitivity and speed, this protection is sensitive to operate under minimum
fault conditions and remain stable under maximum load and external fault conditions. A
distinct rise in the current between the neutral points can only be caused by an unbalance
in the shunt reactor itself.
Therefore this method provides very fast and sensitive protection of the reactor windings,
especially for inter-turn fault.
Tertiary Bus
L1
L2
L3
52
Circuit breaker or
circuit switcher
Shunt
reactor
I>, t
51
Figure 61 Simple split phase protection based on current balance measurement between
the two shunt reactor neutrals.
88
5.3.4.2 Turn to turn fault detection - Voltage unbalance scheme
This protection scheme (Figure 62) which in fact is a voltage differential scheme utilizes
the following facts:
 A phase to ground fault in the tertiary system (1) affects both the reactor bank
neutral-to-ground voltage and the grounding transformer open delta winding to
the same degree. The same applies for other system voltage unbalances such as
phase-phase faults.
 A reactor turn to turn fault (2) however, will mainly cause a rise in the reactor
bank neutral-to-ground voltage.
Thus it is possible to use the measured voltage from the open delta winding to cancel the
measured reactor neutral-to-ground voltage in case of phase to ground faults. If correctly
connected and tuned, the overvoltage protection in the voltage unbalance scheme will
only see an overvoltage in case of reactor turn to turn faults.
Tertiary Bus
L1
L2
L3
1
52
Grounding
transformer
3Uo>
59N
Shunt 2 Ground
reactor overvoltage Voltage unbalance protection
protection
Summation Overvoltage
protection
59
Neutral
VT
Figure 62 Turn to turn fault protection based on voltage unbalance scheme.
An alternate method where the summation is done directly in the VT secondary circuits is
shown in the Figure 63 below.
89
Tertiary Bus
L1
L2
L3
1
52
Grounding
transformer
3Uo>
59N
Shunt 2 Ground
reactor overvoltage
protection
Eventual
matching VT Voltage differential
protection
Neutral 59
VT
Figure 63 Turn to turn fault protection – alternate voltage unbalance scheme.
5.3.4.3 Application notes - Voltage unbalance scheme

Due to manufacturing tolerances there will be a fixed neutral to ground error voltage
during normal operation. This error may increase further if the single phase units in a dry
type shunt reactor are not arranged in an equilateral triangle configuration.
The sensitivity of the setting is then mainly limited by the tolerance of the voltage
transformers.
5.3.5 Phase differential protection (87 / ∆I) for tertiary

connected shunt reactors
It is a common practice in some countries to include the shunt reactor in the tertiary zone
of the phase differential protection applied for autotransformers (Figure 64).
Because the setting of this differential protection must be based on the actual rating of the
power transformer and depending on the MVAr rating of the shunt reactor, sensitivity
consideration should be accounted for.
90
1000 MVA
50 MVA
ΔI>
87T
Figure 64 Phase differential protection for autotransformer including tertiary connected

shunt reactor in the protected zone.
5.3.6 Tertiary connected shunt reactors - Typical protection

schemes
In this section some typical protection schemes applied to transformer tertiary connected
air core shunt reactors are presented. Each complete scheme reflects some utility
practices as illustrated in below, see Figure 65, Figure 66 and Figure 67.
91
Tertiary Bus
VT
I>> I>, t
CT 50 51 59N 3Uo>
Shunt
reactor
Figure 65 Transformer tertiary connected shunt reactor scheme number 1
Table 4 Requirements met, on tertiary connected shunt reactor protection scheme number 1
Protect for internal fault ph-ph X [50, 51]
Protect for internal fault ph-earth X [59N]
Protect for internal winding turn to turn fault (X) [51]
Provide for breaker failure

92
Tertiary Bus
Main 2 / 52 Main 1
Backup VT
ΔI> I>> I>, t I2>

87T CT 50 51 46 59N 3Uo>
Voltage unbalance
protection
Shunt
reactor 59
Neutral
VT
Protect for internal fault ph-ph X [50, 51] , (X) [46] (X) [87T]
Protect for internal fault ph-earth X [59N], (X) [46]
Protect for internal winding turn to turn fault X [59 unbalance], (X) [46]
Provide for breaker failure

93
Tertiary Bus
Main 2 / 52 Main 1
Backup VT
I>> I>, t I2>

CT 50 51 51BF 59N 3Uo>
Shunt
reactor
Split phase
protection
51
Protect for internal fault ph-ph X [50, 51]
Protect for internal fault ph-earth X [59N]
Protect for internal winding turn to turn fault X [51 split phase]
Provide for breaker failure X [51 BF]

94
6 MONITORING
The following monitoring benefits are common for all types of shunt reactors:
 To prevent damage that could end shunt reactor life.
 Provide condition-based shunt reactor maintenance.
 Increases reliability by detecting a problem before the shunt reactor fails.
 Decreases repair costs.
Measurable indicators for oil immersed shunt reactor serviceability include:

 top oil temperature
 winding hottest-spot temperature
 ambient temperature
 fault history
 dissolved gas analysis
Measurable indicators for air core shunt reactor serviceability include:

 winding hottest-spot temperature
 ambient temperature
 faults history
6.1 Thermal Overload
The purpose of the thermal overload protection is to prevent overheating of the shunt
reactor due to increased system voltages as a result of harmonic distortions. A single
overcurrent relay can perform this function in case of small shunt reactors, with the
disadvantage that it does not take the load history prior to the overload into account.
Using an overcurrent relay is thus only an approximation and not a true overload
protection.
A principle is therefore employed for larger shunt reactors based on a thermal image of
the reactor. Such relays usually detect several defined temperatures and give alarm in
stages to increase the forced cooling and initiate a reduction of load. The reactor is only
tripped, if the overload persists for a long period or the temperature rise approaches its
maximum permissible limit.
6.1.1 Thermal process

Heat transfer takes place in three main components of the shunt reactor: core, windings
and oil.
The extent of heating of these components is according to their thermal time constants.
Considering a component as a homogenous body, the thermal time constant can be
defined by the following equation.
95
mc
 eq. 41
 S
m = the weight of the element being considered [kg]

c = the specific heat [Ws/kg o C]
α = the coefficient of heat discharge by convection [W/m2 o C]
S = the area through which the heat is discharged [m2]
The denominator of eq. 41 can be replaced by the amount of heat discharged in a second
per degree of temperature difference of the heat dissipating surface and its surroundings.
Making this substitution, eq. 41 becomes
m  c 
 eq. 42
Qc
QC = the amount of heat discharged per second [W]

θ = the temperature difference of the body and its surroundings [oC]
Figure 68 presents an exponential curve that represents the heating process that takes
place in the transformer. The temperature increases to 63% of the final value after a time
equal to the time constant, τ. Because the increase of the temperature is proportional to
the square of the current, a current equal to 1.26 times the rated value would cause the
temperature to rise to a final value in time T (1.262²x 0.63 = 1) and the final temperature
would be 1.6 times the rated value.
Figure 68 Temperature rise as a function of time
When the amount of heat generated in the winding/core is equal to the heat carried
through the oil to the surface of the tank, the over-temperature is given by the following
exponential.
96
 t   in  1  e t /   eq. 43
θt = the over-temperature at time t

Θin = the initial temperature that is stabilized by the initial load of the reactor
τ = the time constant
The heating process of the different reactor components takes place by the same law of
variation, but is different at different moments in time due to their different thermal time
constants [ref 15]. The time constant of oil is of the same order of magnitude as the time
constant of the core but the time constant of the winding is much lower. Due to this
difference, winding temperature is the weaker element in comparison to the oil and core.
6.2 Oil temperature protection

Because of the heating and cooling requirements of an oil immersed transmission reactor,
some specialized temperature protection is required to provide protection over the full
range of operating limits of the reactor. The reactors are temperature limited by the
ambient temperature, the cooling system condition, and the excitation voltage.
The oil temperature protection is intended to keep the insulation oil from reaching levels
where there would be deterioration of the liquid properties or even a risk of fire if its
flash point was reached.
Modern oil temperature protection is generally comprised of one or two temperature

sensors located at the reactor cover, so as to measure the top oil temperature. Two sensors
are generally used when a redundant temperature measurement is wanted, so that the
temperature protection remains functional in case of a sensor failure. This can be of great
advantage, considering that a sensor replacement is generally possible only with the
reactor de-energized.
The sensors, commonly of type Pt100Ω at 0ºC, are connected to an electronic unit that
indicates the temperature locally and remotely through analogue outputs (e.g. 4-20mA) or
through a communication protocol (e.g. DNP3.0 or IEC61850). The electronic unit has
programmable settings for high temperature alarm at a first stage and then very high
temperature alarm and reactor trip at a second stage.
Since oil temperature is a very slow changing variable, some utilities have the trip signal
intentionally time delayed for several minutes, so as to give operators time for remedial
actions. These oil temperature devices do not meet any of the other requirements but are
again the only devices which meet the temperature limit requirements.
6.3 Winding temperature protection
97
The winding temperature protection is intended to keep the solid insulation, typically oil-
impregnated paper, from reaching levels where there would be very fast deterioration of
its mechanical properties (commonly referred as insulation aging) or even a risk of turn-
turn fault due to the release of water vapor bubbles if bubbling temperature is reached.
To simulate the winding temperature, a resistor sized to approximate the heating in the
reactor winding at full load is supplied by a current transformer from one of the phase
currents. The resistor heating is added to the top oil temperature but circulating the top oil
into a well with the resistor. This combined heating temperature is used to simulate the
winding temperature. The winding temperature is usually limited to 140-180 degree C.
Modern winding temperature protection is usually integrated to the same electronic unit
for oil temperature protection. Since a regular temperature sensor cannot be installed at
the winding due to the insulation needed, winding temperature is commonly calculated
based on top oil temperature and load current, besides the reactor particular
characteristics programmed by user according to manufacturer data. For this purpose, a
CT secondary is connected to the winding temperature relay for reactor’s load
measurement.
Cooling fans or oil circulation pumps, if any, are normally controlled by the winding
temperature relay based on winding temperature and/or load current.
These winding temperature protection do not meet any of the other requirements but are
again the only protection which meets the temperature limit requirements.
6.4 Oil level / flow monitoring
Levels of oil in shunt reactors should be monitored to ensure that their operation is not
jeopardized. Devices used for this purpose are described in this section.
6.4.1 Magnetic oil level gauge

Many shunt reactors are provided with an expansion vessel, called conservator, to
take care of expansion in the oil volume due to rise in temperature when the load on the
transformer increases or the ambient temperature increases. It is essential to maintain the
oil level in the conservator above a pre-determined minimum level. All shunt reactors
with conservator tanks are, therefore, fitted with magnetic oil level gauges that also
incorporate mercury switches; these switches close and actuate audible alarms in the
control room in the event of oil level dropping to near empty position in the conservator.
Normally prismatic/ toughened glass type gauges are used for oil level indication.
New method for oil level measurement is carried out by a sensor (e.g. ultrasonic sensor)
which is mounted onto the top of conservator. Ultrasonic pulses send from the ultrasonic
sensor are metering the oil or air border and reflected back. The sensor calculates the oil
level from the time that has elapsed between the transmitting of the ultrasonic pulse of
the echo. The measured distance is converted into a distant- proportional current signal
and evaluated by the monitoring system.
98
6.4.2 Bushing oil level indicator
Prismatic/toughened glass type gauges are also used for oil level indication in
shunt reactors bushings.
6.4.3 Flow indicators

Flow indicators are provided in oil forced air forced (OFAF) shunt reactor. These
indicators are safety devices that provide an electrical signal on failure of forced
circulation of oil in the cooling circuits. These flow indicators show specified rate of
full flow in specified direction in known size of pipe and operate one or two micro-
switches when rate of flow drops to approximately 70% of the specified full flow rate.
These switches can be used to initiate precautionary system or safety devices.
6.5 Pressure Relief Valve

The pressure in a shunt reactor tank can rise due to a fault inside the tank. One or more
devices are provided on the shunt reactor tank to relieve the pressure in the tank.
The pressure reliving equipment is spring-operated, self resealing device that
releases volume of oil that is just sufficient to relieve the excess pressure before
resealing the tank. The oil pressured relief valve is often monitored and connected to the
SCADA system.
The PRD is finding widespread preference over the explosion vent or bursting diaphragm
type equipment due to its superior features as it re-seals itself.
6.6 Fire protection
Fire extinguishing systems are provided on many shunt reactors and for preventing
serious damages if a fire takes place in or around the shunt reactor. The fire extinguishing
equipment use either pulverized water or nitrogen.
Shunt reactors and power transformers are among the most expensive items of equipment
located in power plant and substations. They generally contain a large quantity of
combustible substance, which can, if ignited, propagate fire to nearby installations.
Special attention should therefore be given to their protection.
The shunt reactor fire protection acts when the tank pressure increases after an internal
insulation breakdown. The depressurization process ensures that the shunt reactor
pressure is returned to normal after several milliseconds. During the depressurization
process the oil gas mixture are directed towards the oil gas separation tank.
The nitrogen injection logic requires two independent signals to trigger. One of the
signals will come from the shunt reactor rupture disks or fire detection system and the
other from the shunt reactor electrical protection, see Figure 69.
99
Figure 69 Nitrogen injections in prevention mode logic.
For the system to initiate the nitrogen injection in extinction mode, two signals are
required: fire detection system signal and electrical protection signal, see Figure 70.
100
Figure 70 Nitrogen injections in extinction mode logic.
A block diagram and a description of the shunt reactor fire protection are shown in Figure
71 and Figure 72. The components of the protection system are presented in Table 7.
Figure 71 Transformer (and shunt reactor) fire protection system
101
Figure 72 A typical system for shunt reactor fire protection
Table 7 Components used in a typical system for shunt reactor fire protection
Item Description Item Description
number number
1 Transformer (shunt reactor) 15 Nitrogen Cylinder
2 On load tap changer (OLTC) 16 Electrical actuator
3 Conservator 17 Explosive gas elimination pipe to
transformer
4 Buchholz relay 18 Explosive gas elimination pipe to
OLTC
5 Bushing 19 Explosive gas elimination pipe to
OGST
6 Isolation valve 20 Cabinet
7 Absorber 21 Explosive gas elimination valves
on transformer (shunt reactor)
8 Rupture disk 22 Oil gas separation tank (OGST)
9 Decompression chamber 23 Air isolation shutter
10 Depressurization set support 24 Explosive gas evacuation pipe
11 Oil drain pipe 25 Oil drain pipe from other
equipment
12 OLTC rupture disk 26 Conservator shutter
13 OLTC decompression chamber 27 Explosive gases burn in safe area
14 OLTC oil drain pipe
6.6.1 Advantages and disadvantages of pulverized water

systems
The fire extinguishing systems utilizing pulverized water have the following advantages
and disadvantages.
102
6.6.1.1 Advantages
 Water spray covers all parts of the shunt reactor.
 No foreign material is injected into the shunt reactor.
6.6.1.2 Disadvantages
 Require large facility for storing water.
 Require an auxiliary building.
 Require high pressure compressors.
 Require large space around the shunt reactor.
 Have nozzles that may get clogged.
 Oil that may accidentally come into contact with water; the oil would have to be
replaced.
6.6.2 Advantages and disadvantages of the nitrogen systems

The nitrogen system has the following advantages and disadvantages.
6.6.2.1 Advantages
 Simple to construct.
 Requires little space.
 Can be quickly easily installed, commissioned and utilized.
6.6.2.2 Disadvantages:
 Introduction of nitrogen into the shunt reactor tank and evacuating oil are
necessary.
 Risk of unwanted operation of the nitrogen system, if the fire detection system
operates unnecessarily.
 Special oil draining system is required.
 Does not cover fire in the bushings.
6.7 Integrity of Insulating Oil
The oil used in shunt reactor performs three essential functions; these are thermal transfer
of heat, dielectric insulation and transport information about the health of the shunt
reactors. A shunt reactor may be considered as a chemical reactor to understand the
process of physics and chemistry for assessing the health of the shunt reactor.
The oil is the messenger that holds vital information on the physical condition of the
shunt reactor.
Normal practice consists of obtaining oil samples from the shunt reactor tank, if it is
used, for testing. Standard tests on the oil samples include determining the absolute
water contact in oil, particle count in the oil, IFT, color, and most importantly perform
dissolved gas analysis.
103
6.7.1 Signature Analysis
This topic of signature analysis or interpretation of results from oil tests is addressed in
many papers, as well as in the guide lines IEC [ref 16], [ref 17] and IEEE [ref 18], [ref 19].
6.7.2 Dissolved Gas Analysis

Many gradually evolving incipient faults in reactors have detectable symptoms that
indicate problems. One of these symptoms is the production of dissolved gases in oil.
6.7.2.1 Dissolved Gases in Oil

Dielectric oil and cellulose dielectric insulation (paper) break down under thermal and
electrical stresses in the reactor. This process produces gases of varying concentrations
depending on the stresses applied to these materials. The gases dissolve in the oil. The oil
is sampled and analyzed; the composition of the gases and their concentrations that are
indicative of the nature and severity of the fault in the reactor are determined. The
changes in the production of each gas and its rate of production are important factors in
determining the fault(s) and their evolution.
Some specific gases are recognized as being indicative of certain types of faults.
6.7.2.2 Degradation of Oil-Impregnated Cellulose

The thermal degradation of oil-impregnated cellulose produces carbon monoxide and
carbon dioxide. Hot spots in the windings, on insulated leads, and in areas where
pressboard and cellulose components and spacers are used produce these gases.
6.7.2.3 Degradation of Dielectric Oil

The degradation of oil through abnormal dissipation of energy in the reactor can be
detected by analyzing the produced gases. The energy released through a fault such as
overheating, partial discharge, corona and arcing results in production of gases as the oil
degrades. The detection of these gases allows for not only the identification of the fault
process, but also for its monitoring.
These degradation by-products, known as fault gases, include hydrogen, hydrocarbon
gases, methane, ethane, ethylene and acetylene. It is important to note that each of these
gases has a characteristic energy required for its formation. As a result, each individual
gas can be related to a specific fault process.
6.7.3 Early Detection on Oil-Filled shunt reactors

Regularly scheduled and periodic use of Dissolved Gas Analysis (DGA) on a population
of transformers and shunt reactors usually reveals that 90 % of the sampled units are
behaving in a satisfactory manner.
The balance of the units may be considered suspect and, therefore, should be closely
watched.
104
The behavior of a shunt reactor is satisfactory when the reactor has not deviated from its
previously established baseline, equilibrium point or fingerprint. A normal and constant
gas level for one shunt reactor may be very high for another. Each reactor has its own
unique normal dissolved gas pattern. It is the change in gas levels and, equally important,
the rate of change of the gas levels that cause a problem unit to stand out from the others.
6.7.4 Dissolved Gas Analysis

A DGA represents only a five-minute data window or snapshot in time about the
condition of a shunt reactor. It cannot and does not guarantee that a good report means
status quo until the next DGA is performed.
There are markedly long periods of time during which fault gases of the shunt reactor are
not monitored if a DGA is applied on a six or twelve month schedule. A serious problem
could easily start and go undetected for days, weeks, or even months and fully evolve
into a catastrophic failure with no warning. All of this could occur after a good DGA, and
before the next scheduled DGA.
In order for a DGA program to be truly effective, one of the following two changes
should be made:
 Either DGA needs to be performed on a much more regular basis, approaching the
unrealistic schedule of once per day.
 A cost-effective and reliable real-time gas-trending trigger or early warning signal
should be used to effectively bridge the time gap between regularly scheduled
DGAs.
6.7.5 Incipient Failure Condition Detection

The number of different technologies that can be used in various ways to detect on-line
symptoms of shunt reactor failures has steadily increased during the previous twenty-five
years. Since the root cause of failure of a reactor is the breakdown of the insulating
system (oil and paper), the techniques detect both key fault gases (hydrogen and carbon
monoxide) to provide early warning, alarming and trending.
These technologies have developed to the point that thousands of on-line dissolved gas
monitors are installed on critical reactors. Many published papers describe cases in which
these monitors successfully detected failure conditions before other protection systems
operated.
Unfortunately, on-line dissolved gas-in-oil monitors do not detect all failure modes.
Some failure modes are very fast and no amount of on-line monitoring can detect these
fast evolving faults.
6.7.6 Key Gases and Dissolved Gas Indices

The key gasses usually looked for are Hydrogen, Acetylene, Methane, Ethylene, Ethane,
Carbon monoxide, and Carbon dioxide.
The breakdown of the key gasses and the associated fault types are listed below in Table
8.
105
Table 8 Key gases and fault types
Key Gas Associated Fault Type
Hydrogen (H2) Arcing, Corona, Overheated oil
Acetylene (C2H2) Arcing
Methane (CH4) Corona, Overheated oil, Cellulose breakdown
Ethylene (C2H4) Arcing, Corona, Overheated oil, Cellulose breakdown
Ethane (C2H6) Corona, Overheated oil
Carbone monoxide (CO) Cellulose breakdown
Carbone dioxide (CO2) Cellulose breakdown
The gasses listed above are typically grouped together, except for CO2, and are identified
as Total Dissolved Combustible Gas (TDCG). [ref 18] provides recommendations for
determining what course of action should be taken depending on the level of the key
gases and the TDCG. The courses of action are designated as “conditions” and are
divided into four categories with associated recommendations.
Condition 1: TDCG levels are normal and indicate the reactor is operating properly. Any
individual combustible gas exceeding specified levels should prompt additional
investigation.
Condition 2: TDCG levels within this range indicate greater than normal combustible
gas level. Exercise caution, analyze monthly.
Condition 3: TDCG levels within this range indicate a high level of decomposition.
Exercise caution, analyze weekly, consider planned outage, notify manufacturer.
Condition 4: TDCG levels within this range indicate excessive decomposition.

Continued operation could result in reactor failure. Analyze daily, consider removal from
service, notify manufacturer.
Another approach to determine the types of faults is to calculate the ratios of the key
gasses found dissolved in the reactor oil. Depending on the level of each key gas and the
ratio of specific combinations of gasses, the type of fault can be determined. The ratios
are designated as Doernenburg and Rogers ratios, [ref 18] for further information.
6.8 Partial discharge measurements

Partial discharge (PD) measuring technique represents an efficient method to assess the
faults of Shunt reactors insulation. This technique was previously used in fully screened
laboratories but it is now being used on site as well. This is possible due to the use of
digital techniques in PD Signal acquisition process and processing the acquired
information with special software filtering the useful components from external
disturbances.
PD measurement requires taking the shunt reactors out of service for a short
period of time that leads to a disturbance of the power system it is part of. That is why
106
operative monitoring of the operational state of its insulating parts was made by the
analysis of oil dissolved gasses content as well as by PD measurement using acoustic
method. If the results of the two measurements show that there is a fault PD electric
measurements are performed to determine its nature and location.
6.8.1 Acoustic method for PD detection

To detect acoustic emission transducers with piezoelectric sensors that have maximum
sensitivity for frequencies of 25, 65 and 125 kHz are usually used.
6.8.2 Monitoring Shunt reactor Bushings

The analysis of failures of large shunt reactor units in operation has led to the conclusion
that the majority of events are caused by the failure of auxiliary parts in general and by
the failure of bushings in particular. There were instances when the deterioration of
bushings resulted in the shunt reactor unit being taken out of service and/or a fire leading
to important economic losses.
An important countermeasure in this respect is to monitor the bushings. Traditionally, the
Bushings have been monitored off-line and on-line.
Most systems that are available at this time are based on the principle of measuring on-
line the variation of power factor of the bushing capacitive current or tan δ.
Measuring the bushing capacitive current variation with time has yielded good results for
more than 20 years and has pointed out the evolving faults consisting of the deterioration
of parts of the insulation.
On-line monitoring by tan δ measurement is relatively new at the international level and
has just begun to be utilized on power transformer units and reactors. In order to expand
its application it is necessary to test its measuring sensitivity and accuracy by means of
off-line comparative measurements.
Both methods have the alarm and switch-off functions of the reactor units in case of
failures that lead to deterioration of the bushings.
6.9 Overvoltage measurements

Knowledge of overvoltages which affected the shunt reactor, in connection with the
quantity of gases which are dissolved in the oil, allows determination of possible
damages of the insulation of the energized part.
Capacitive voltage sensors are used for the detection of the overvoltages on the shunt
reactor, see Figure 73, [ref 18].
107
Figure 73 Voltage sensor connected to bushing
6.10 Parameters monitored
What can be monitored?
 Temperatures
 Currents, system voltages
 Tap-changer
 Oil levels
 Oil -DGA, moisture, breakdown
 Bushing taps, Leakage current, RF, pressure
 Radio Frequency Signals
 Acoustic signals
 Neutral current
 Magnetic circuit
 Cooler operation
A typical monitoring system for an oil immersed shunt reactor is capable of monitoring
the conditions of various shunt reactor components as shown in Table 9. By the use of
software to store and perform trend analysis of the measured data, the operator can be
presented with information on the state of health of the shunt reactor and alarms raised
108
when measured values exceed appropriate limits. This will normally provide the operator
with early warnings of degradation within one or more components of the shunt reactor,
enabling maintenance to be scheduled to correct the problem prior to failure occurring.
The maintenance can obviously be planned to suit system conditions, provided the rate of
degradation is not excessive.
Proposed list of main parameters for shunt reactors monitoring is presented in Table 9.
Table 9 List proposed of main parameters for shunt reactors monitoring
Nr Monitoring subject Parameter for Actions

monitoring
1 Oil status Temperature Alarm/ trip
Gas Monitoring Alarm/ trip*
Hydrogen (H2)
Carbon monoxide
(CO) *see comments
Acetilene (C2H2) below
Etylene (C2H4)
Ethane (C2H6)
Water in oil Alarm
2 Insulation Status Paper humidity Alarm
Hot-spot Alarm/trip
temperature
Operating time Alarm
Lifetime Alarm
Acoustic signals:
-vibrations Alarm
-partial discharges Alarm
3 Bushing status Environment Alarm
temperature
Insulation Alarm/ trip
Capacitance
Insulation Power Alarm
Factor (or Tangent
Delta)
4 Gas presence and gas Gas presence Alarm/ trip
pressure detected by
Buchholz relay
Overpressure valve Trip
5 Cooling system Pump function Signal
Cooler function Signal
Status of cooler
Signal
switches
Cooling system Alarm
efficiency (normal /
109
low efficiency)
6 Oil in tank Oil level Alarm
Status of rubber Alarm
bag/membrane in
conservator tank
(normal / leaking).
7 Electrical Primary currents Displayed
parameters Phase voltages Displayed
Active power
Reactive power Displayed
Copper losses Displayed
Iron losses Displayed
Connected/ Signal
8 Shunt reactor status
disconnected
1. Oil status:
 Oil temperature: knowledge of the top oil temperature allows the following
statements to be made: thermal monitoring (e.g. triggering of the alarm in the
event of excessive temperatures), hot-spot temperature calculation and overload
capacity.
 Gas dissolved: Trip by monitored Gas level is not a common practice, since this a
slow changing variable (hydrogen takes time to dissolve in oil), so that it is up to
the operation and/or maintenance people to decide for an immediate or scheduled
trip of the reactor after receiving an alarm.
2. Insulation status:
 Partial discharges and thermal overload lead to gaseous decomposition products
of the oil. The quantity of the harmful gases in the oil is therefore a sign for
oncoming problems within the shunt reactor insulation. The hydrogen in the oil
can be detected with a special sensor. The sensors measures dissolved hydrogen
continuously and accurately. The quantity of hydrogen is a reliable indicator of a
recent or existing fault, as it is produced by all fault types. Small concentrations of
Hydrogen are thus providing an early warning for the presence of a fault. The
sensor measures the cumulative gas quantity on the basis of a fuel cell in relation
to a proportional distribution formula. The increase of the quantity of gas can be
used as an initiative for a conventional gas analysis in order to draw conclusions
about the type of the fault.
 Moisture of oil: The knowledge of the water content in the oil allows conclusions
to be made on the absorption of humidity caused by the breathing of the shunt
reactor and development of humidity caused by disintegration of the paper
insulation. The humidity content is a measure for the electrical strength of the
insulating oil.
110
 Low frequency vibrations, occurring from the fundamental frequency causing
mechanical vibrations with double frequency (2*fundamental frequency) can be
measured with microphones attached magnetically on the reactor tank. By
studying the amplitude of over tones up to approximately 1 kHz and further, tells
if the fundamental frequency signal changes shape and gets sharper because of
changes in the mechanical construction due to vibrations. High frequency
vibrations, Partial Discharge (PD) measurements (acoustic- and electrical-signals)
in the frequency range 100 kHz can be measured and studied, if they occur, they
need to be located where they arise from PD origin i.e. if there is a arising
problem with insulation in the shunt reactor or no PD like core vibrations or other
places, see [ref 10], [ref 11], and [ref 37].
 Hot-spot temperature: this is the parameter used by the Winding Temperature

Relay for alarm and trip.
3. Bushing status:
 The external insulations consist of the bushing and an oil-filling for insulation. A
failure of the bushing can happen due to leakages. If the bushings have separate
oil volume it is possible to install a pressure sensor. The pressure of the bushings
is acquired and compared with the other phases.
 Ambient temperature: a resistance thermometer with 3-conductor connection is

usually used for the detection of the ambient temperature.
4. Gas presence and gas pressure:

 Detection of gas amount in the Buchholz relay: the condition of the isolation-
system of oil immersed shunt reactors is monitored mainly by the Buchholz relay.
The purpose of this equipment is to detect the gases, which leave the shunt reactor
tank during operation. The main disadvantage of the Buchholz relay is its
cumulative characteristic, the history of gas generation is unknown. The
difference between long duration low energy faults like partial discharge and high
energy faults like local overheating cannot be determined by a Buchholz relay.
The low energy faults lead to a continuous gas production of a small amount of
gases over a long period of time and high energy faults lead to high gas rates in a
short time. To assess these kinds of errors efficiently, it make sense to determinate
the rate of gas generation. This is the task of the Buchholz gas sensor. The
protective function of the Buchholz relay is not influenced by the Buchholz gas
sensor.
5. Cooling systems:
 Temperature in/out of the cooler system: knowledge of the oil temperatures at the
inlets and outlets of the cooler allows the determination of different characteristic
values. The decrease in the oil temperature via cooler and the oil flow may be
used to establish the cooling effect of the individual cooler.
111
 The shunt reactors are generally designed for natural cooling with the radiators
mounted directly on the tank. However sometimes it is required to have some
action in the cooling circuit, such as operating pumps and fans, depending on the
status of the shunt reactor circuit breaker. The control action can be initiated by
the circuit breaker auxiliary contact or by operation of an overcurrent relay. By
using the overcurrent relay a secure control action is obtained when the reactor is
energized independent of the circuit breaker auxiliary contact status.
 The operating condition of pumps and fans: by recording the status of the pumps
and fans it is possible to establish the operating time of the individual elements,
also the operating states are necessary for determining the thermal time constants
of the shunt reactor.
6. Oil in tank:
 To detect oil leakage, the oil levels in the conservator tank or main tank are
measured.
7. Electrical parameters:
 Load current: the knowledge of the load current allows the calculation of the
following values: power, hot spot temperature, ageing rate, thermal modeling and
actual possible overload capacity. E.g. the hot spot temperature can be
determinate by load current and top oil temperature. This enables the calculation
of the ageing rate and the actual overload capacity, see IEC 60354.
 Voltage: Knowledge of over-voltages in connection with the quantity of dissolved

gases helps determine possible damages of the insulation of the energized parts.
8. Shunt reactor status

 Circuit breaker normal open auxiliary contact is used for shunt reactor breaker
status. It is provided to monitoring/protection equipment as a binary input signal.
112
Table 10 Monitored components and health condition
6.11 Monitoring systems
The monitoring system architecture can typically be integrated – all sensors are
connected to a centralized unit, which then is connected to a monitoring computer – or
decentralized – intelligent sensors (IED) are connected directly to a monitoring computer
through a data communication network. Typical system integrated architecture for monitoring
is presented in Figure 74. A typical monitoring system using decentralized architecture is
shown in Figure 75.
113
Figure 74 Typical system integrated architecture for shunt reactor monitoring
IED Sensors Monitoring

Computer
Conservator tank monitoring

Communication
Network
Bushing Monitoring
Temperature Monitoring
Intranet /
LTC Monitoring Internet
Gas and Water Monitoring
Other parameters
Maintenance Remote Operation
Warning Center
Figure 75 Typical decentralized architecture for on-line monitoring system.
114
7 NEW SOLUTIONS OFFERED BY NUMERICAL RELAYS
7.1 Adaptive DC biasing

Switching in a shunt reactor gives rise to inrush currents – specific for shunt reactors is
that shunt reactors show relatively small inrush currents (reactor core keeps practically no
remanence because of the air gaps), but the duration of the off-set is fairly long (long DC
time constant, up to seconds, due to the inherent low losses in a shunt reactor, very small
resistance). The slow decay of the DC components in the reactor currents makes it rather
difficult for current transformers, which tend to saturate after some time, often after
several fundamental frequency cycles up to several seconds. Saturated current
transformers mean in general unreliable information on the true currents, and unwanted
events, such as unwanted trips of differential protection for inrush or external faults,
could take place. It is therefore necessary to keep this phenomenon in mind when
designing the relay protection system of shunt reactors.
To avoid unwanted operation of the differential protection during switching in a shunt

reactor, an adaptive DC biasing can for example be used to change the trip level
(sensitivity of the differential protection) due to false differential currents, [ref 39].
Example:
The DC components (DC offsets) are continuously extracted from the three instantaneous
differential currents. The highest DC component of all three is then taken as a kind of a
DC bias in the sense that the highest effective, temporary sensitivity of the protection
(normally this is the sensitivity in Section 1 of the operate–bias characteristic) is
temporarily decreased as a function of this DC component. The DC bias current is not
allowed to decay (from its highest value) faster than with a time constant i.e. T = 1
second, see Figure 76.
115
Instantaneous differential current and its DC component. DC bias current.
80
Instantaneous
differential This line exponentially decays from
current the maximum DC offset of the inst.
60
diff. current. The line represents the
DC bias current. Decays
exponentially with T = 1 second
40
20
Fault occurs
DC offset of
the instantaneous
-20
differential current Mean value of a secondary
CT saturated current is zero for a
fully saturated CT
-40
0 50 100 150 200 250
Time in milliseconds 
Figure 76 Observe the DC offset of the instantaneous differential current

As long as the current transformers on both sides of a reactor transform the primary
currents in exactly the same way (whether saturated or not) there will be no differential
currents, see for example Figure 77, where the instantaneous differential currents IDL1
and IDL2 under the first 200 ms are practically zero. On the other hand, if current
transformers on one side saturate (first), false differential currents will appear (see IDL3)
that might cause the shunt reactor to be switched off the power system. This can take
place after such a long time as one second or more. To counteract these unpleasant
phenomena, which may result in an unwanted trip command, the sensitivity of the
differential protection is temporarily decreased, based on the DC offset in the
instantaneous differential currents as described above. This DC desensitization is not
active, if a disturbance has been detected and characterized as internal fault by the
internal/external fault discriminator described in section 7.2.
116
Reactor neutral side currents (measured input currents)
1
Currents in kA iL3 iL1 (neutral side)
CB iL2 (neutral side)
0.5
closes iL3 (neutral side)
-0.5
0 50 100 150 200 250 300 350 400 450
Reactor terminal side currents (measured input currents)
1
iL1 (terminal side)
iL3
Currents in kA
CB iL2 (terminal side)

0.5
closes iL3 (terminal side)
-0.5
0 50 100 150 200 250 300 350 400 450
Instantaneous differential currents (calculated by the differential protection)
0.4 IDL1
Currents in kA
CB IDL3
IDL2
0.2 closes
IDL3
0
-0.2 IDL2
50 100 150 200 250 300 350 400
Time in milliseconds 
10-Jan-2007 14:16:28
Figure 77 A shunt reactor suddenly connected to the power system (inrush), with some eventual
differences in the CTs on both sides, or different loads on them, the differential protection sees these
currents as shown in the figure. As a result, the differential protection calculates (false!) instantaneous
differential currents as shown in the figure. These false differential currents might cause an unwanted trip
of the reactor.
7.2 Negative-Sequence Internal-External fault discriminator
To enhance the differential protection a negative-sequence-current-based internal-

external fault discriminator can be used.
The internal-external fault discriminator not only positively detects a fault but in a very
great majority of cases correctly discriminates between internal and external faults, and is
thus a powerful tool. Short operate times for internal faults are obtained by implementing
this new feature.
The operate–restrain characteristic should be constructed so that in a certain protection

application, it can be expected that:
117
 For internal faults, the operate currents (differential) are always safely, i.e.
with a good margin, above the operate–restrain characteristic, see Figure 78.
 For external faults, the false (spurious) operate currents are safely, i.e. with a
good margin, below the operate–restrain characteristic, see Figure 78.
Differential currents versus bias currents for heavy internal and external faults
2
IDL1MAG, internal fault
1.8 IDL1MAG, external fault
"steady-state" operate-restrain characteristic
1.6 for internal fault
1.4 SlopeSection3 = 80%

Differential current in kA
1.2
OPERATE
1 REGION
0.8
SlopeSection2 = 40 %
0.6
RESTRAIN
0.4 IdMin REGION
0.2
0
0 0.5 1 1.5 2 2.5 3 3.5
Bias current in kA
Figure 78 An example of a good operate – restrain characteristic. Under heavy external fault condition
with current transformer saturation, the spurious false differential current does not enter the operate region.
The internal-external fault discriminator feature can enhance the stability of the
differential protection by making this protection to operate faster for internal faults and
blocking this protection for external faults, see Figure 79 and Figure 80.
The internal-external fault discriminator responds to the relative phase angles of the
negative-sequence fault currents at both ends of the shunt reactor winding:
 If the two negative sequence currents flow in the same direction, the fault is
internal.
 If the two negative sequence currents flow in opposite directions, the fault is
external.
118
If a fault is classified as internal, then any eventual block signals by the harmonic
criterion are overridden, and the differential protection can operate very quickly without
any further delay.
If a fault (disturbance) is classified as external, then generally, but not unconditionally, a

trip command is prevented.
One of the advantages of using the negative-sequence currents compared to the zero-
sequence currents is that they provide coverage to phase-to-phase faults as well, not only
faults involving earth.
The negative sequence internal-external fault discriminator works satisfactorily even in
case of three-phase faults. Because of the decaying DC offset of the fault currents, the
system is not quite symmetrical immediately after the fault. Thanks to the transient
existence of the negative sequence system, faults can be distinguished as internal or
external, even for symmetrical three-phase faults.
90 deg
120 deg
Angle could not

be measured. Relay Operate Angle
One or both
currents too Internal
small … fault
The directional
region
characteristic
is defined by 180 deg 0 deg
the settings:
******************
Minimum negative
1. Minimum negative sequence current
sequence current External
fault
2. Relay operate angle region Internal / external
fault boundary.
Default ± 60 deg
270 deg
Figure 79 Internal/external fault discriminator.
119
Decision by the internal - external fault discriminator: Internal Fault
90 Internal / external
When directional fault boundary
120 60 with the angle
test is impossible,
the points are placed Negative Sequence
on this line ... Relay Operate Angle
150 30 (ROA)
External
fault Internal
region fault
region
180 0
Internal
5 kA fault declared
here 8 ms
210 330 after fault
10 kA
240 300
270
Negative sequence differential current phasor (in kA)
Directional limit (within ± NegSeq ROA degrees is internal fault)
Figure 80 Example of an internal fault using internal / external fault discriminator.

The Figure 80 shows a case where the trajectory of the negative-sequence phasor remains
within the internal fault region all the time in spite of current transformer saturation. The
fault must definitely be internal. In the beginning, when one or the other of the two
negative sequence phasors (i.e. two contributions to the total negative sequence
differential current) is too small for a directional measurement, the magnitude of the total
differential current phasor is mapped on the 120-degrees axis. A long as this condition
persists neither external-, nor internal fault (disturbance) is declared. The figure
corresponds to an internal L1– L2 short circuit on terminals of a generator.
7.3 Open CT supervision
If open CT supervision is provided in the protection devices, the protection engineer

should apply this feature according to the utility philosophy and directives.
7.4 New approach to reactor turn-to-turn protection

Reactor turn-to-turn fault is a common internal fault. Research on theorem related to
reactor turn-to-turn fault is less seen at present. Various protection equipment functions
such as transformer protection are combined to provide such protection.
120
The following analyzes electrical characteristics of reactor internal turn-to-turn fault and
external grounding fault using the equivalent circuit of Figure 81.
The current and the voltage for the calculation of reactor turn-to-turn fault protection are
from the bushing CT at reactor terminal and from the line VT respectively, and the
polarity of the CT is at the line side shown in Figure 81.
AC AC
ZS ZS
CT VT
ZR ZR
Figure 81 VT and CT used for the turn-turn protection
1. External ground fault on the line:

The fault point is the zero sequence source of the system, see equivalent zero sequence
network in Figure 82.
External
ZS fault ZS
CT VT
I0
E0
ZR + ZR
Figure 82 Zero sequence network at external phase-ground fault

As this is external grounding fault, at the point where reactor protection is installed, zero
sequence current flows from grounding point into the terminal side of the reactor, and
then into the ground. Therefore, at the point where the protection is installed, U0 and I0
satisfy the following relation, see eq. 44:
U0  I0  Zr eq. 44
Zr = impedance of the reactor
121
The vector relation is, see Figure 83:
U0
I0
Figure 83 Zero sequence voltage and current at external phase-ground fault
2) Internal turn-to-turn fault:

Since the fault point is inside the reactor, equivalent zero sequence network is, see Figure
84.
ZS ZS
CT VT
I0
Internal
fault
ZR ZR
+E
0
Figure 84 Zero sequence network for internal turn-turn fault

Internal turn-to-turn fault or turn-to-ground fault is the fault source of zero sequence
network. At the point where the reactor protection is installed, zero sequence current
flows from inside of the reactor, and into the ground via equivalent system impedance. In
this way, at the point where the protection is installed, U0 and I0 satisfy the following
relation, see eq. 45:
U 0   I 0  Z S _ tot eq. 45
ZS_tot = total system impedance
Vector relation is, see Figure 85:
I0
U0
Figure 85 Zero sequence voltage and current for internal turn-turn fault
122
By comparing electrical quantity characteristics of external grounding fault and internal
turn-to-turn or turn-to-ground fault, the following conclusions can be obtained:
1) Opposite zero sequence power directions.
In case of the external grounding fault, U0 leads I0 by 90 degrees; while in case of the
internal turn-to-turn or turn-to-ground fault, I0 leads U0 by 90 degrees. It can be seen
that reactor internal turn-to-turn fault can be differentiated by means of relative phase
relation of I0 and U0.
2) Difference in measured amplitude of zero sequence impedance.
In case of the external fault, measured zero sequence impedance (Z0) is the Z0 of the
reactor itself; while in case of the internal fault, measured Z0 is the equivalent Z0 of
the system. Since reactor Z0 is far larger than system equivalent Z0, reactor internal
turn-to-turn fault can be differentiated by amplitude of measured Z0.
Using an optimized algorithm, a reactor turn-to-turn fault of about 1% can be detected

correctly.
Example:
From above vector analysis, it can be seen that the vector relation between U0 and I0
during internal faults is totally opposite to that during external faults. At the same time,
because the reactor impedance is far greater than the system impedance, the U0 amplitude
is very small which can be seen in the following figure showing two vector relations (it is
supposed that I0 values during the two conditions are equal).
Ext fault U0
I0
Int fault U0
Figure 86 External respective internal fault to ground

If U0 is changed, the U0 magnitude during internal faults can be raised to improve the
protection sensitivity.
U 0 (calc )  U 0  U eq. 46
After the transformation, the vector relation is changed to, see Figure 87:
123
External fault U0
External fault U 0 + U
I0
Internal fault U 0
U
Internal fault U 0 + U
Figure 87 To improve sensitivity for internal turn-turn faults
Voltage waveforms are shown below, which are recorded during of a simulation test of
2% turn-to-turn fault of reactor in the dynamic simulation lab, see Figure 88.
0 0 3 > 3 5 _ A -U A (V )
0 0 3 > 3 5 _ A -U A ( V ) 0 0 3 > 3 6 _ B -U B ( V ) 0 0 3 > 3 7 _ C -U C ( V ) 0 0 3 > 3 8 _ N -3 U 0 ( V )
100
50
M a g n it u d e ( M a g )
-5 0
-1 0 0
0 50 100 150 200 250 300
T im e ( m s)
E le c tr o te k C o n ce p ts ? T O P , T h e O u tp u t P r o ce sso r ?
Figure 88 Voltage waveform
124
8 PROTECTION IMPLEMENTATION AND SETPOINT
RECOMMENDATIONS
This chapter is based on the settings recommendations and concepts for further
consideration borne out of Sections 5, 6, & 7. Reactor protection depends on the type of
reactor and its connection to the power system as discussed in depth in Section 5.
Looking at the example Figure 51, Figure 52 and Figure 53 for bus-connected oil-
immersed reactors, it is obvious there are multiple protection configurations possible.
Some protections are not possible such as sudden pressure or Buchholz relaying on an
air-core reactor. Whether or not to use redundant or backup protection is a matter of user
choice and depends upon factors such as
1. The importance of the particular reactor installation to the electric grid system.
2. The cost of the reactor equipment.
3. Any special design likely to be more susceptible to certain faults.
4. The need to disable the Main 1 (Primary) protection system and maintain full
protection on a second protection system while the reactor remains in service.
Most oil-immersed reactors are probably worthy of redundant (Main 1 and Main 2) or at
least limited backup protection in addition to a complement of full protection. Most
utilities reported in the survey that when two redundant systems are used, there are
separate DC supplies, CT windings, and trip coils. Lower-cost, lower-voltage air-core
reactors may not warrant more than one set of full protection.
8.1 PROTECTION SCHEMES
The main types of reactor protection schemes are listed below.
 Differential relay for phase-to-phase and phase-to-ground faults

 Impedance relay for turn-to-turn faults and back-up protection for phase and
ground overcurrent relays
 Phase Overcurrent Protection
 Negative-sequence Overcurrent protection
 Ground fault protection by Restricted Earth Fault, or ground overcurrent, or
Neutral voltage displacement (shift) relay
 Mechanical type fault detectors (Buchholz, Oil level indicators, Pressure relief
devices)
 Oil temperature and winding temperature relays
 Acoustic Amplitude of Vibration and Tank Monitoring
 Dedicated turn-to-turn protection by sequence controlled ground overcurrent,
split phase, or voltage unbalance.
125
8.1.1 Reactor differential Protection
Differential protection is the most commonly applied protection for bus-connected oil-
immersed shunt reactors. Generally, the differential relays provide quick clearance of
phase-to-phase and ground faults. When using low-impedance differential protection
(section 5.1.1.1), recommendations include:
 It is imperative to verify that the CTs and wiring fulfill the requirements
necessary for proper functioning of the protective relay scheme.
 Use a relay with biasing or restraint of the fundamental frequency to guard against
false operation due to unequal CT response to large external fault currents.
 Use a relay with 2nd harmonic restraint or blocking to guard against false
operation due to unequal CT response to DC offset with long time-constant.
Utilizing such a device that also cross-blocks all of the phases during this
condition will provide even more security against false trips. Another method is
adaptive DC biasing which desensitizes the differential relay if DC offset is
detected.
 Use a relay that includes an unrestrained differential element with a setting chosen
below the steady-state saturation current of the connected CTs (a typical setting is
200% of the rated reactor current).
 Improved internal/external fault discrimination using negative-sequence algorithm

is another refinement available in certain numerical differential relays (See
Section 7.2).
High-impedance differential protection (Section 5.1.1.2) minimizes the above concerns

regarding CT response causing false operation but requires dedicated CTs of the same
type. These relays are set to operate correctly assuming poor CT response occurs.
 It is imperative to verify that the CTs and wiring fulfill the requirements
necessary for proper functioning of the protective relay scheme.
8.1.2 Impedance (Distance) relay
Impedance relays are used as primary protection or as back-up protection for the reactor.
It is also used for detecting turn-to-turn faults within the reactor. Such relays detect the
faults inside the reactor for a good percentage of the winding. Turn-to-turn faults may not
change the through-current of the reactor but the impedance values change drastically up
to at least 60% of impedance of the reactor. Distance protection consists of a single or
126
preferably a two-zone impedance relay on the high side of the reactor looking into the
reactor.
The two-zone impedance relay has the benefit of providing high speed tripping in the
Zone-1 protection and slower speed tripping in Zone-2. It must not be set such that it
operates for inrush characteristics during reactor energization or de-energization. The
setting of the relay may need to be coordinated with other protection while also taking
into account the effects of energizing and de-energizing transients. Other protection
might be line distance relaying when a reactor is tied to a line and has its own breaker.
Setting recommendations include:
 Consider the apparent impedance of the reactor during overvoltage conditions

(see Section 5.1.5).
 Consider the conditions of inrush with full DC offset.
 The WG B5-37 survey concludes fast-operating distance protection is generally

limited to 60% of reactor impedance to ensure security.
 Consider a second zone of distance protection set to cover more reactor

impedance but with time delay to allow for inrush and overvoltage to be reduced
before this second zone operates. This may also improve the sensitivity of turn-
to-turn fault protection furnished by the impedance relay.
 Using microprocessor-based impedance protection can provide better filtering for

inrush and DC offset concerns.
Additional considerations on distance protection of shunt reactors:
 Consider 2nd harmonic restraint for impedance protection (either implemented in

microprocessor-based relay or added in series with the distance tripping element).
8.1.3 Phase Overcurrent Protection
Phase overcurrent protection is a very inexpensive, simple, and reliable scheme for fault
detection and is used for some reactor protection applications as a back-up protection.
The setting must be high enough to prevent inrush currents from causing unwanted
operation. It should have both instantaneous and time-delayed elements. The
instantaneous elements help in providing high-speed clearing of heavy current faults
which may threaten system stability.
Settings recommendations include (see Section 5.1.3):
 A typical conservative setting for analog instantaneous protection is 6 times rated
127
reactor current.
 Some relay manufacturers suggest an adequate setting for microprocessor

instantaneous protection can be 2.5 times rated reactor current with a 0.1 second
delay if the relay uses Discrete Fourier Filtering (DFF) and looks at only the RMS
value of the fundamental. Consult the relay manufacturer to confirm this feature.
 A typical setting for the time-overcurrent pickup is 1.5 times the rated reactor
current with intent to minimize false operation during temporary overvoltages.
8.1.4 Negative Sequence Overcurrent Protection
Settings recommendations include:
 Negative sequence overcurrent (usually used for air-core reactor designs) should
be set above the maximum level of natural system unbalance and manufacturing
tolerances.
 The time delay should be set to coordinate with other protections that operate for
unbalanced faults external to the reactor.
8.1.5 Ground fault Protection
Protection for a ground fault within a grounded wye-connected shunt reactor is best
provided by a simple phase differential or conventional Restricted Earth Fault (REF)
relay selected and set on the same philosophy as that for transformer REF protection.
Sometimes a ground overcurrent relay is used as a back-up protection when phase
overcurrent protection is provided.
8.1.5.1 Restricted earth fault relay (REF)
Zero-sequence differential relays (Restricted earth fault relay) are one way of providing
protection against phase-to-ground faults in shunt reactors supplied from solidly
grounded systems. Generally, this protection is also provided for EHV shunt reactors
with a Neutral Grounding reactor connected between the star point of the shunt reactor
and ground.
Implementation recommendations include (see Section 5.1.2):
 For low-impedance REF, use a relay with biasing or restraint of the fundamental
frequency to guard against false operation due to false residual current caused by
unequal CT response.
128
8.1.5.2 Ground overcurrent relay
The ground overcurrent protection is a very inexpensive, simple, and reliable scheme for
fault detection and is used for grounded wye-connected reactor protection applications as
a back-up protection for phase-to-ground faults. It is usually used in conjunction with
phase overcurrent relay. It can, as in phase overcurrent, have both instantaneous and
time-delayed elements. The ANSI designation for a ground overcurrent relay connected
in the residual of the reactor terminal CTs is 51N. A relay connected to a single CT in
the reactor neutral is a 51G.
Implementation/setting recommendations include:
 Confirm that the relay used is immune to off-nominal frequency currents. 3rd
harmonic and its multiples must be filtered out by the relay.
 The sensitivity to the harmonic and inrush currents is one of the main problems
with back-up ground over current relays. Settings must be able to allow inrush,
which usually means desensitizing the back-up relay. Numerical relays are
recommended as they offer the best characteristic since the digital filters remove
harmonics and DC offset currents from the inrush.
 If only one ground overcurrent function is used, it is recommended to be

measured from a single CT in the reactor neutral because 1) it will perform well
for ground faults throughout the reactor, 2) it offers some degree of protection for
turn-to-turn faults, 3) it is more immune to CT saturation.
 This protection may need to be coordinated with protections for external ground
faults.
Further considerations:
 Using a 51N function from the residual CT connection on the terminal end, plus a
51G function measured from a single CT in the reactor neutral will provide
improved ground fault coverage and provide better fault location.
 Modern numerical relays detect 2nd harmonic currents and offer the possibility to
program logical equations to block and unblock various protection devices
according to a specific philosophy. For example, the 51G on the neutral side
winding of the reactor may be set to detect 2nd harmonic energization current and
block the 51N provided on the terminal side of the reactor.
8.1.5.3 Ground overvoltage relay (Neutral Displacement)
Considerations include:
129
 For air-core reactors connected to a transformer delta tertiary winding, ground
fault detection can be done using zero-sequence overvoltage (neutral
displacement voltage). This protection will not detect ground faults at or near the
reactor neutral but should be set as sensitive as possible to cover most of the
reactor winding.
 A voltage relay that is tuned to the fundamental frequency is required. 3rd

harmonic components (and its multiples) are present because of the delta and their
effects are undesirable to this overvoltage protection.
 This protection cannot determine the location of the ground fault. The fault can
be anywhere on the delta tertiary system.
8.1.6 Dedicated Turn-to-Turn Fault Protection
Several protection schemes in this report are effective at detecting some level of turn-to-
turn faults. These schemes in general would be more sensitive than the ordinary distance
(21) function.
8.1.6.1 Torque-controlled Earth Fault Overcurrent Protection
Considerations for these devices
 The instantaneous ground overcurrent relay of this scheme could be susceptible to

false operation if the residual from three phase CTs is used. Using a single CT in
the reactor neutral to minimize this possibility is recommended.
 To provide adequate operating quantities the directional relay used has voltage
polarization reinforced by passing appropriate sequence fault current through a
“replica” impedance. This impedance represents the reactor impedance but must
actually be a value less than the impedance of the reactor. Otherwise the scheme
may misoperate for earth faults external to the reactor.
 If a source of zero-sequence voltage is not present to be reinforced or zero-

sequence directional control is inadequate, negative sequence directional relaying
may be used. However, a negative sequence current relay is susceptible to
unbalances such as may occur during unequal CT response during energization or
denergization.
 For line connected shunt reactors, blocking these schemes with an undervoltage
relay and time delay may be appropriate to prevent false operation during
temporary unbalances from other causes including line switching.
130
8.1.6.2 Spilt Winding Protection
Detection of turn-to-turn faults can be better assured by using a sensitive unbalance

scheme similar to that applied to detecting faults within capacitor units (see Section
5.1.6.3). Split-phase overcurrent protection can be set for a typical pickup threshold of
2.5% of reactor rated current.
Thoughts for future turn-to-turn protection:
 Specify split-winding reactors in the future to get good coverage for turn-turn
faults.
8.1.6.3 Voltage Unbalance Scheme
 This scheme applies to delta tertiary (ungrounded) operation of reactors and

operates on the voltage difference between 3U0 measured using a broken-delta
VT configuration at the reactor terminals compared to the voltage measured to
ground at the reactor neutral. It can be set very sensitively but is limited by a
fixed neutral-to-ground error voltage from reactor manufacturing tolerances,
reactor physical arrangement, and voltage transformer tolerances.
Thoughts for future improvement:
 Given microprocessor technology, this unbalance scheme could benefit from

adaptive techniques to compensate for the natural unbalances of the system
while taking into account the manufacturing tolerances all in real-time,
resulting in even more sensitive protection.
8.1.7 Mechanical type fault detectors (Buchholz, Sudden

Pressure, Pressure relief devices)
These protective devices when provided on shunt reactors of medium to high capacity are
quite similar to those provided on power transformers. Although these devices will
detect most faults within a reactor tank, they are often the only protection for certain turn-
to-turn faults as it is difficult for the electrical protection devices to detect this type of
fault.
Considerations for these devices:
131
 The Buchholz relay is significantly more secure in operation than the sudden
pressure relay. The Buchholz is the preferred device for conservator-type shunt
reactors.
 According to the WG B5-37 questionnaire, some utilities are considering

duplication of these devices. Also for higher reliability, duplicating the initiating
contacts is sometimes done and may be considered on case-by-case basis
depending upon the location and size of shunt reactors and voltage class.
 To increase reliability and to allow optional gas collection from different places in
the tank, one of the surveyed utilities provides two Buchholz relays in each tank,
located in separate pipes from different areas of the tank to the conservator.
Further there are two pressure relief devices on each tank. This is because they
feel that using only one of these protection functions for sensitive internal fault
detection does not adequately cover the entire tank area.
Additional thoughts for utilizing mechanical protection devices:
 Compare the sensitivity of the electrical to the mechanical gas pressure protection
to determine which should be the main protection. Consider the following:
1. To alarm for small turn-turn faults, before gas appears
2. To coordinate time delayed tripping of shunt reactors with the sudden pressure
(63) function
3. To provide instantaneous tripping
8.1.8 Top Oil and Winding Over-Temperature Protection
These protections provided on shunt reactors are similar to those provided on power
transformers and criteria for selection and setting should therefore be the same.
Considerations for these devices include:
 For higher reliability, duplicating of the initiating contacts is sometimes done and
may be considered on case-by-case basis depending upon then location and size
of shunt reactors and voltage class.
Thoughts for future improvement to thermal overload protection:
 Develop adequate shunt reactor overload protection taking into account the
surrounding external ambient temperatures and time constants of the reactor. The
protection scheme should measure the true RMS (RMS up to the 50th harmonic,
Nyquist frequency of 2.5 kHz and sampling frequency of 5 kHz), determine the
overcurrent in the reactor winding, and start the reactor cooling.
132
8.1.9 Breaker Failure/Pole Disagreement
Although the topic is mentioned in this B5 report, all of the variations of breaker failure
schemes could constitute a separate report. For shunt reactors, it is important to consider
the breaker failure initiate (BFI) logic that may need to include trip initiations from non-
electrical protections such as sudden pressure and Buchholz relays.
Considerations:
 Some relay manufacturers combine breaker failure and pole discrepancy in the
breaker failure scheme, the scheme of Figure 47 is very desirable if you use two
trip coils and list the functions associated with the pole discrepancy.
8.2 Reactor Configurations and Protection Schemes
Protection systems for reactors should provide full coverage during fault conditions. The
first system of protective functions to achieve this is referred to as Main-1 (Primary) in
this report. In addition, Main-2 or back-up protection should be provided when a likely
potential for failure to trip exists. Such a condition might be when routine maintenance is
performed on the Main 1 protection system while the reactor remains in service. Full
redundancy (Main-2) or partial (Backup) protection should be considered; otherwise the
reactor should be de-energized during such maintenance. In the future, full coverage
Main-2 protection should be the goal. This would be easy to accomplish with numerical
relays.
The protective functions applicable to a reactor depend on the type (oil-immersed or air
core) and connection configuration of the reactor. For example, for line-connected
reactors equipped with a 4th neutral reactor, the neutral reactor is not stressed under
normal conditions and is normally protected in the Main-1 protection system by only the
Buchholz relay and the single phase differential, which is the only electrical protection
fast enough. Oil temperature is usually the only protection in the Main 2 system for the
neutral reactor. Line-connected reactors usually need a communications channel to the
remote line end to transfer trip for reactor faults if the reactor does not have its own
breaker or transfer trip for reactor breaker failure if the reactor has its own breaker.
Some reactor protective functions may need to be blocked for certain conditions such as
reactor application to a series-compensated line, predicted switching transients, or ill-
effects to ground relays from mutual inductance during ground faults. Example schemes
for inrush and deenergization blocking for line connected reactors are discussed in
section 5.2. The trip release logic scheme shown in Figure 57, provides blocking for
inrush and deenergization.
133
8.2.1 Typical Main-1 protection schemes
Typical oil-immersed line/bus-connected Main-1 systems consist of these schemes:
 Differential protection for internal three phase-to-ground, phase-to-phase and

phase-to-ground faults.
 Buchholz gas-accumulator relay or sudden pressure relay for low current turn-
to-turn faults.
 Top oil or winding over temperature protection for temperature protection.
 REF or neutral overcurrent protection as ground fault protection.
 Breaker Failure and pole disagreement protection.
Typical air-core tertiary-connected Main-1 schemes consist of:
 Overcurrent or impedance relay for phase fault protection.
 Neutral overvoltage protection for ground fault protection.
 May have split phase, distance, negative sequence, or voltage unbalance

protection for turn-to-turn faults.
Additionally for air core reactors, phase differential and breaker failure protection may be
considered. The phase differential used can be the transformer differential if it is
acceptable to clear the entire transformer for loss of a reactor.
8.2.2 Typical Main-2 and back-up protection schemes
A true Main-2 protection scheme may not be provided. As operating requirements change
and single contingency limits are reached, use of Main-2 protection may need to be re-
evaluated in future.
For line/bus connected oil-immersed reactors, typical Backup protection consists of:
 Overcurrent or impedance relays for internal three phase-to-ground, phase-to-

phase and phase-to-ground faults.
 Instantaneous and time-delayed overcurrent relays for phase and ground fault.
134
 Directionally-controlled ground overcurrent relays or to some degree
impedance relay for detection of turn-to-turn faults.
Typical air-core tertiary-connected Main-2/Backup schemes consist of:
 Typically not provided due to lower cost and less significance of air-core
reactors.
Most utilities reported in the B5-37 survey that when two redundant systems are used,
there are separate DC supplies, CT windings, and trip coils.
135
9 CONTROL
To control the power system voltage, shunt reactors are switched on/off either manually
or automatically. Standard [ref 20] provides guidelines for switching requirements of
shunt reactors. For example, one of the most important aspects of switching reactors is
current chopping by forcing reactor current to zero. This will result in high voltage across
breaker poles. Solution to this problem may include surge arrestors applied to the reactor
and transformer tertiary terminals to avoid excess voltage during switching. This section
discusses switching actions.
9.1 Manual switching

The shunt reactors are normally switched in/out by manual mode. In Table 25, Table 26,
Table 27 and Table 28 illustrates the utilities using manual or automatic control for
switching shunt reactors in or out. 67% of responding utilities connect shunt reactors
manually.
9.2 Automatic switching
In order to improve power system performance, utilities may choose to perform

automatic shunt reactor switching by monitoring the busbar voltage level. This
functionality is quite easy to integrate into multifunctional numerical relays.
However the user must carefully check the relay performance regarding the following
points:
 Over/ under voltage relay or voltage function with high precision and pickup drop
off ratio is required for such an application.
 Typically more than one over/ under voltage level with independently settable
time delays is required within the relay.
 Over/ under voltage relay shall be capable to operate only when all three voltages
are above/ below set operate level or relay must be capable to measure and
operate on the value of the positive sequence voltage.
The automatic closing and opening commands are issued on over voltage and under
voltage detection respectively after a certain time delay.
The commands will be a pulse with adjustable duration.

The automatic closing operation will be blocked:
 against successive commands
 by operation of shunt reactor protection
 by operation of busbar protection
 by operation of transformer protection (when the shunt reactor is connected at the
tertiary of the transformer).
136
Unblocking will be made only by the operating personnel. The automatic opening
operation will be blocked in case of VT open circuit.
U>
t1 tC
AND T ON T ON CLOSE
CMD
taux2
AUTOMATIC ON OR T ON
AND
AUTOMATIC OFF AND

taux3 taux1
T ON T ON
OR
PROT Block
t2 tO
AND T ON T ON OPEN
CMD
U<
Figure 89 Example of logical diagram for shunt reactor automation
Where:
U> - overvoltage function
U< - undervoltage function
tO – open pulse duration
tC – close pulse duration
taux1 – time after the scheme is ready to block actions
taux2 – blocking time with automatic recovery after command (close/ open)
taux3 – monitoring and blocking time for repeated commands
t1 – time required to measure overvoltage level
t2 – time required to measure undervoltage level.
137
9.3 Point on Wave Controller
9.3.1 General statement
All types of shunt reactors, independent of magnetic and electric circuit, can be switched
in a controlled manner. The strategy for controlled opening is to select arcing times long
enough to avoid re-ignitions at de-energizing. The strategy may vary depending on the
size of the shunt reactor.
The strategy for controlled closing is to energize at instants resulting in flux symmetry
(current symmetry) thereby minimizing the inrush current and the risk for nuisance
tripping and rotor vibrations in nearby generators due to zero sequence current.
An alternative strategy for controlled energizing is to energize the shunt reactor such that
the transient voltage will be minimized. This will require energizing close to voltage zero
and there is no possibility to find compromise targets to reach both low inrush currents
and low transient voltage. It should also be noted that the zero sequence current
protection needs to be disabled for a certain time or needs to be made less sensitive.
When a single-pole operated circuit breaker is controlled, a separate output command is
given to each pole. In that case the non-simultaneity is achieved by electrical means and
the staggering is named electrical staggering. See Figure 95 and Figure 96 for typical
block diagrams.
When three-pole operated circuit breakers are used a mechanical phase shift must be built
in. This mechanical phase shift, staggering, ensures that contact touch and/ or contact
separation for -all poles occur at intended instants. This mechanical staggering is
achieved by special design of the mechanical linkages in the bottom mechanism housing.
For three-pole operated circuit breaker only one pole, the master pole, is controlled while
the other two operate in slave mode.
It is not always possible to combine controlled closing and controlled opening with three-
pole operated circuit breakers.
Suppression of switching transients
There are several important circuit breaker applications where random closing or opening
instants may lead to severe voltage and current switching transients. These transients
occur in the main circuits, but may also induce transients in control and auxiliary circuits,
as well as in adjacent low voltage systems. The switching transients are associated with a
variety of dielectric and mechanical stresses on the high-voltage equipment and may lead
to a variety of disturbances (e.g. in substation control and protection systems, computers
and processors, or telecommunications).
Normal energizing of shunt reactors may cause severe transients: high over-voltages,
under-voltages, or high inrush currents. Upon, de-energizing of shunt reactors, re-
ignitions will occur, resulting in steep voltage surges. The magnitude of the transients
depends on the point of wave where closing or opening of the circuit breaker contacts
occur. In a situation without controlled switching, sooner or later the switching instant
will occur at the worst possible phase angle.
138
Even though a modern breaker will have very low re-strike probability at switching of
reactive loads, for statistical reasons a few occasional re-strikes may occur during the
course of a large number of switching operations. This risk of occasional re-strikes may
be eliminated by means of controlled opening operations.
Conventional countermeasures such as pre-insertion resistors, damping reactors or
resistors, or arresters are used to limit the magnitude and effect of the switching
transients, after they have occurred. In addition, system and equipment insulation may be
upgraded to withstand the stresses. These methods, however, may be inefficient,
unreliable or expensive, and do not treat the root of the problem.
Principle of controlled switching
Controlled switching is a method for eliminating harmful transients via time controlled
switching operations. Closing or opening commands to the circuit breaker are delayed in
such a way that making or contact separation will occur at the optimum time instant
related to the phase angle. See Figure 90.
Figure 90 Block diagram for principle of controlled switching

The general operating principle of point-on-wave controller is presented in Figure 91.
BB
VT Reference
Voltage
Output Input
Command Command
CB POW
SR
Figure 91 Point -on-wave general operating principle
139
If controller is used for controlled opening, it is important that all protection trip
commands are by-passing the controller. When the shunt reactor is to be energized, an
input command is given to the point-on-wave controller. Following the command the
controller will determinate a reference time instant, related to the phase angle of the
busbar voltage. When this has been done, and after an internally created waiting time, the
controller will then give an output closing command to the circuit breaker. The time
instant for the output closing command is determined by the make time of the circuit
breaker and the target point for making. Both the predictable make time and target point
has been pre-programmed into the controller. The circuit breaker will then close at the
correct time instant and minimize the switching transients.
9.3.2 Control of opening operations
Problems related to chopping over voltages and re-ignition
In addition to the inductance of the winding, a shunt reactor always has some stray
capacitance in the windings, the bushing and in the connecting leads. When a reactor is
de-energized, the voltage across it will oscillate with the natural frequency determined by
the inductance and stray-capacitance. The oscillation frequency is typically a few kHz.
Due to chopping (premature interruption) of the current slightly before the current zero
crossing, the oscillating reactor voltage will have higher amplitude than the supply
voltage.
For example, typical magnitudes of this “chopping over voltage” for modern SF6 circuit
breakers are 1.2 to 2 pu with the highest values occurring for small reactors. The
chopping over voltage, with its limited amplitude and frequency, is normally quite
harmless both for the reactor itself and for the surrounding system see Figure 92.
Figure 92 Voltage across shunt reactor at de-energizing without re-ignition
Due to the oscillating reactor voltage, there will be a high voltage stress across the circuit
breaker. If the contact gap is short, the circuit breaker probably will reignite, see Figure
93. A re-ignition will generate high-frequency transients (typically hundreds of kHz) in
both reactor voltage and current. Following a re-ignition, the reactor current will be
interrupted again either at high-frequency zero crossing of the current, or most probably,
140
at the subsequent power frequency zero crossing. The very steep voltage transients
caused by re-ignitions will be unevenly distributed across the reactor winding, with the
highest stress on the initial turns. There is a risk that the voltage stress will lead to
puncture of the winding insulation in the reactor, which in the long run may lead to
complete breakdown. Insulation of nearby equipment may also be damaged. Surge
arresters will only protect to a limited extent, since the severity of the voltage stress is
related both to the rate-of-change and to the amplitude.
Figure 93 Voltage across shunt reactor in event of re-ignition
Advantage of using POW Controller
Point-on-wave controllers for shunt reactor circuit breakers are normally used for control
of opening operations. Uncontrolled de-energizing will, in a typical case, cause re-
ignition in at least one circuit breaker pole. By controlling the contact separation in such a
manner that short arcing times will occur, re-ignition will be eliminated. See Figure 94.
The remaining voltage transient is a harmless chopping overvoltage with relatively low
frequency.
Figure 94 Target for contact separation in order to eliminate re-ignitions
141
9.3.3 Control of closing operations
Problem related to inrush current
Energizing of shunt reactor may cause inrush currents with high asymmetry and long
time constants. The actual magnitude of the inrush current is quite dependent on the
range of linearity of the reactor core. Due to the air gaps utilized in shunt reactor cores
there are no severe saturation effects. In spite of their limited amplitude, reactor inrush
currents may still have adverse effects. These inrush currents may lead to zero sequence
current, cause saturation of CT cores, with resulting nuisance tripping of relays or cause
other network disturbances.
Advantage of using POW Controller
Controlled closing of shunt reactor circuit breakers is utilized in several cases, and
normally as a complement to controlled opening.
The making target that gives the lowest reactor inrush current is the peak of the power
frequency voltage across the circuit breaker, and this target is normally utilized. Making
of the current in this case creates a transient voltage stress equal to that which occurs if
the circuit breaker reignites at 1 p.u. voltage during a de-energizing operation. This
voltage stress is normally acceptable but if such a voltage stress is considered
unacceptable, an alternative procedure is to make the current at voltage zero cross
contacts. This will in principle lead to maximum inrush current. Zero sequence relays
may then be set with time delay (or be set less sensitive), in order to avoid nuisance
tripping. It is not possible to minimize both the inrush current and the transient voltage
stresses at the same time.
Suitable Circuit Breakers for controlled switching
The circuit breaker used for controlled switching should have a stable opening and
closing times and contacts with high dielectric withstand capability.
The spring operated circuit breakers have stable operating times that are not affected by
ambient temperature and control voltage extensively.
These circuit breakers also have a high and stable dynamic dielectric withstand capability
between the contacts, both upon making and breaking operations.
These properties, together with the stable operating times, make these circuit breakers
well suited for controlled switching.
Experience has shown [ref 21] that with synchronous switching of circuit breaker with
independent poles, a phase inrush current can be limited to less than 1,5 p.u. with the
zero-sequence current lower than 0,5 p.u. Thus mechanical stresses of the shunt reactor
and adjacent transformers are reduced and false operation of zero-sequence current
protection prevented.
142
9.3.4 Single pole and three pole operation
When a single-pole operated circuit breaker is controlled a separate output command is

given to each pole by controller.
Three-pole operated circuit breakers can be arranged with mechanical staggering, making
them suitable for controlled switching. In these cases there will be a built-in mechanical
phase shift between the poles that is appropriate for optimized switching of the load at the
specified frequency. The mechanical phase shift is arranged to reach proper switching
conditions with minimum staggering.
Figure 95 Point-of-wave block diagram for opening
143
Figure 96 Point-of-wave block diagram for opening and closing
144
Figure 97 Example on point-of-wave selection tree.
9.4 New trends
9.4.1 Adaptive functions
Some controllers are equipped with special adaptive functions to control the result of a
controlled switching operation. The adaptive control can be arranged in different ways
for both controlled closing and controlled opening. The principle of the adaptive control
is that a detected error from the target will be compensated for in the next controlled
operation.
Adaptive control on closing operations

The ultimate target for controlled closing is the intended energizing instant. The optimum
way of supervising the energizing instant is to note the phase angle at which the CB starts
to pre-strike with respect to the selected reference. This kind of control can easily be
arranged by receiving the current start signal “current detection” from a current
145
transformer. As an alternative the voltage onset instant can be detected by means of a
voltage transformer behind the CB “voltage detection”. A typical arrangement for
detection of current start is shown in Figure 98.
Figure 98 Shunt reactor energizing with current start feedback loop

If no instrument transformers are available, it is still possible to supervise the results of a
controlled closing operation by detecting the contact touch instant of the CB. In this type
of applications, the “voltage detection” function will be used and the signal will be given
by an auxiliary contact. For single pole operated CB the adaptation control can be
arranged for each pole individually.
Adaptive control for opening operations

In most cases of controlled opening, the target consists of a wide range of arcing times
and not a single instant. The need of adaptive control is of less importance compared to
that for controlled closing having high demands on precision to make the current at the
correct moment.
There are two options depending on the type of opening operating mode and controller:
 To detect the contact separation instant with respect to the selected reference. This
must be arranged by use of special auxiliary contacts. This method is not
reflecting the result of the interruption.
 To use the signal from the current transformers in the load branch. The intended
interruption instants are known by the controller by its inputs.
9.4.2 Impact of substation configuration on adaptive control

arrangement
146
Different substation arrangements may call for special solutions for overall total
functionality. Double circuit breaker schemes and ‘one and a half” circuit breaker
schemes may require special arrangements to attaining proper function of adaptive
control.
In Figure 99 it shows a CB and a half scheme with controlled reactor circuit breakers. All
CT’s are installed outside the reactor bays.
Figure 99 Circuit breaker and half scheme with CT outside reactor bays.
Generally every circuit breaker will be controlled with its own POW controller. However,
in some substation configurations one circuit breaker may switch two different loads,
which may call for more than one controller per circuit breaker. In most cases a specific
load is switched by its own circuit breaker. In some substation configurations, one load
may be alternatively be switched by different circuit breakers.
The following two conditions require special care:
 One load switched by two circuit breakers.
 Two different loads alternatively switched by the same circuit breaker.
Special care should be taken when applying the adaptation control features based on re-
ignition detection. This is the case if the shunt reactor can be energized from two sides
and where the CT’s are common. An example of such an installation is shown in Figure
100.
In this case current will always be measured by the CT as long as any of the two breakers
CB1 or CB2 is conducting. The only way to use the re-ignition detection function here is
to arrange a fixed switching order and to set the function disabled for controller of the
first shunt reactor breaker to open.
147
Figure 100 Shunt reactor installation in a CB and a half scheme substation.
148
10 QUESTIONNAIRE ON EXISTING PRACTICES OF
SHUNT REACTORS PROTECTION, MONITORING AND
CONTROL
10.1 Introduction
This chapter is based on the survey conducted by the working group B5-37 to investigate
the situation in the protection, control and monitoring of Shunt Reactors.
The work group received responses from following countries.
Canada (CA) Hydro Quebec

Brazil (BR) Copel
South Africa (ZA) ESKOM
France (FR) RTE
Spain (ES) RED Electrica
Romania (RO) Transelectrica (TEL) & SMART
Sweden (SE) Swedish Power Grid (Svenska Kraftnät)
Norway (NO) Statnett
Finland (FI) Fingrid
Saudi Arabia (GS) Saudi Electricity Company SEC
China (CN) State Grid Corporation of China (SGPC)
India (IN) Power Grid
United Kingdom (UK) National Grid
Australia (AU) Power link
New Zealand (NZ) Transpower
The objective of the survey was to get some details of the practices followed in various
countries regarding control, monitoring and protection of shunt reactors that are directly
connected to EHV system and those that are connected to the EHV system through
tertiary of transformer. The areas on which information was sought are listed below.
Application and Design of Shunt Reactors
 Information on the voltage levels used

 Types of connections to the power system
149
 Grounding methods
 Reactor design
Protection and Redundancy
 Types of electrical protections

 Types of non electrical protections used
 Redundancy
 Degree of redundancy
Fault types and protection limitations
 Types of faults experienced

 Protection performance
 New Solutions
Control and Monitoring of Shunt Reactors
 Types of control (Automatic/ manual)

 Frequency of Switching
 Point on Wave Switching
 Reactor parameters monitored
10.2 Application and design of Shunt Reactor
10.2.1 Question: In what voltage levels the Shunt Reactors

are being used?
Several countries use Shunt Reactors and these are at voltage levels ranging from 10kV
to 1000kV. Table 11 below gives voltage levels furnished by responding utilities having
shunt reactors installed in their systems.
Table 11 Voltage levels

10-66 100-160 220-330 380-500 735-765 1000-1200
kV kV kV kV kV kV
CA 161 315 735
BR 13.8
ZA 11/15/22 132 220/275 400 765
150
10-66 100-160 220-330 380-500 735-765 1000-1200
kV kV kV kV kV kV
FR 20 225 400
ES 30 220 400
RO 35 400
SE 220 400
NO 22/ 66 132 300 420
FI 10/20
GS 13.8/ 33 132 380
CN 10/35/66 220/330 500 750 1000
IN 33 400 765
UK 13 275 400
SCT 33
AU
18 275,330
NZ 11
10.2.2 Question: What is the type of connection to the

power system? (E.g. direct connection to the line, bus or
transformer tertiary)
Shunt Reactors installed at voltage level 10-66kV are mainly connected to transformer
tertiary. Shunt Reactors installed at voltage level 100-160 kV and 220-330 kV are mostly
bus connected. Shunt Reactors installed at voltage level 380-500kV and 735-765 kV and
1000-1200 kV are either bus or line connected.
Table 12 below gives details furnished by responding utilities having shunt reactors
connected to bus, line and/or to tertiary winding of transformer.
Table 12 Shunt reactor connection to the power system

Line Connected Bus Connected Tertiary Connected
(kV) (kV) (kV)
CA 735 161/315/735
BR 13.8
ZA 400/765 220/400/765 66
FR 225/400 20
ES 220/400 30
RO 400 35
151
Line Connected Bus Connected Tertiary Connected
(kV) (kV) (kV)
SE 220/400 ?
NO 420 300/420 22
FI 10/20
GS 380 132/380 13.8/33
CN 500/750/1000 10/35/66
IN 400/765 400/765 33
UK 275/400 275/400 13
SCT 33
AU 275/330 18
NZ 11
10.2.3 Question: What grounding methods are being

used? (e.g. direct grounding, reactance grounding...)
Shunt Reactors installed at voltage level 380-500kV and 735-765 kV and 1000-1200 kV
are either bus or line connected. Bus connected reactors are always directly grounded.
Line connected reactors are grounded either directly or through a neutral reactor. ZA,
CN , IN and AU ground some of the line connected reactors through a neutral
reactor.AU are considering removal of neutral reactors as they are facing some problems
with 330kV circuit breakers.
All countries that have tertiary connected reactors are normally ungrounded. CN have
some tertiary connected reactors that are grounded.
10.2.4 Question: How are your Shunt Reactors designed?

(e.g. Oil or Dry, 3 leg or 5 leg, 3 phase or 1 phase, one or
split winding per phase, with or without auxiliary
winding,...)
Directly connected reactors up to 500 kV are normally of 3Ph, 3 limb or 3 ph, 5 limb
construction. In 735/765 kV and 1000kV level they are of single phase construction. All
these reactors are oil filled.
Shunt reactors designed with split winding or auxiliary winding per phase do not seem to
be common. None of the responding utilities report having these types of reactors. FR,
NO, GS and UK have specifically answered that they do not use any auxiliary winding.
NO has specifically confirmed that they do not use split winding.
Table 13 below gives different types of shunt reactor designs among the responding
utilities.
152
Table 13 Shunt reactor design
3ph,3 leg , oil 3ph, 5 leg , oil
1ph , oil immersed
immersed immersed
CA X
ZA X
X (765kV)
FR X X
ES X
RO X
SE X
NO X X
GS X
CN X X
X (220/330/750/1000 kV)
IN X
X (765kV)
UK X
SCT
AU
X
Tertiary connected reactors used in SA,FR,RO,NO,FN,IN,AU are normally of dry type,

single phase in construction and are air cooled. BR, SCT use 3ph, 3 leg oil filled
reactors. FN and NZ also have some 3ph, oil filled reactors.
10.3 Protection and redundancy
10.3.1 Question: What types of protection are being used

for: e.g. Phase to Phase fault, Phase to Ground fault,
Inter-turn fault?
Following electrical protections are generally used for directly connected shunt reactors.
Differential for ph-ph, restricted earth fault for ph-g, impedance /Buchholz for inter turn
faults, ph over current (instantaneous and time delayed, ground overcurrent
(instantaneous and time delayed) as back up.
Following electrical protections are generally used for tertiary connected shunt reactors:
Ph over current (instantaneous and time delayed) for ph-ph faults, ground overvoltage for
ph-g and inter turn faults.
Table 14 and Table 15 give some details furnished by responding utilities about the
electrical protections used for directly connected and tertiary connected reactors.
153
Table 14 Electrical protections - directly connected reactors
:instantaneous
:instantaneous
Restricted E/F
: time delayed
time delayed
delayed)
overcurrent
overcurrent
overcurrent
overcurrent
Differential
Impedance
Ground
Ground
Others
Phase
Phase
me
X X
CA X X X
ZA X X X X X X
See
note1
FR Neutral ground relay, tank
leakage protection
ES X X X X X X
RO X X X X X X
SE X X X X X
NO X X X X X
GS X X X X X X
X Adaptive directional power
CN X X See X X X X
note2
IN X X X
UK X X X X X X
AU
X
Note1) Zero sequence distance protection
Note 2) Zero sequence distance protection
Table 15 Electrical protection used for tertiary connected reactors

instantan
Overvolt
overcurr
overcurr
Ground
delayed
Others
Phase
Phase
time
eous
age
ent
ent
BR X X X
Sensitive compensated
unbalance relay and
FI X X backup non
X compensated
unbalance relay
154
instantan
Overvolt
overcurr
overcurr
Ground
delayed
Others
Phase
Phase
time
eous
age
ent
ent
SCT X X
Differential and
NZ directional earth fault,
overcurrent and earth
fault.
10.3.2 Question: What types of non-Electrical protection

are being used? (e.g. Oil Temperature, Winding
Temperature, Sudden Pressure, Buchholz relay, Oil level
etc.)
For directly connected reactors following nonelectrical protections are generally used: Oil
temperature, Oil Pressure, Winding Temperature, Oil level, Buchholz, Fire protection.
For Oil filled tertiary connected reactors following are used: Oil temperature, Oil level,
Buchholz.
Table 16 and Table 17 give some details furnished by responding utilities about non-
electrical protection functions used for shunt reactors.
Table 16 Non electrical protections- directly connected reactors

temperature
temperature
Oil pressure
protection
Buch holz
Winding
Oil level
Others
Fire
Oil
CA X X X X X
ZA X X X X
FR X X X X X
ES X X X X
RO X X X X
Extreme voltage
SE X X X X automatic relay and
cooling equipment
NO X X X X
155
temperature
temperature
Oil pressure
protection
Buch holz
Winding
Oil level
Others
Fire
Oil
GS X X X X X
CN X X X X X Cooling intervention
IN X X X X X X
UK X X X X
AU
X X X X
Table 17 Non electrical protections - tertiary connected reactors
Buchholz
temperat
temperat
Winding
Oil level
ure
ure
Oil
BR X X X
FI X X X
SCT X X
NZ X
10.3.3 Question: In what Reactor Voltage Level do you use

2 Protection Groups (e.g. Main A and B or Main and
back-up)? Describe how the protection functions are
organized in each group? (e.g. differential in group A
and overcurrent in group B)
Most of the responding utilities use two groups of protections for directly connected
reactors but they are not fully duplicated. One utility provides full duplication for
electrical protections but use only one set of mechanical protections. There is another
utility that use only one group of protections.
For tertiary connected reactors use of single group of protections is common. One utility
uses two groups of protections for tertiary connected reactors too.
156
Table 18 and Table 19 gives details of manner in which redundancy is provided in the
protection system by the responding utilities in case of directly connected and in tertiary
connected reactors.
Table 18 Redundancy for directly connected reactors

Group A protections Group B protections
CA - Group A - Group B
- Differential
- Differential
- Restricted earth fault
ZA - Restricted earth fault
- Overcurrent (inst+time
- Overcurrent (inst+time delayed)
delayed)
- Earth fault (time delayed)
FR - Only one group -
- Differential
- Differential
ES - Restricted earth fault
- Overcurrent (inst+time delayed)
delayed)
- Differential
RO - WTI
delayed)
- Oil temperature
- Earthfault (inst+timedelayed)
- Oil pressure
- Buchholz
delayed)
SE - Differential
- Neutral point overcurrent relay
- Buchholz
- WTI
- Oil temperature
delayed)
NO - Differential
- Mechanical protections
GS - Group A - Group B
- Differential
- Differential - Restricted earth fault
- Restricted earth fault - Adaptive zero sequence power
- Adaptive zero sequence power protection
protection - Adaptive zero sequence
CN - Adaptive zero sequence distance protection
distance protection - Overcurrent (inst+time
- Overcurrent (inst+time delayed) delayed)
- Earthfault (inst+timedelayed) - Earthfault (inst+timedelayed)
- Neutral point overcurrent - Neutral point overcurrent
- Mechanical protections
IN - Differential - Restricted earth fault
157
- Backup impedance - Buchholz 1)
- Oil temperature - WTI
- Oil pressure - Oil level
- Fire protection
UK - Overcurrent (inst+time
- Differential
delayed)
AU - Differential - Differential
- Mechanical protections - Mechanical protections
NZ - -
1) For 765 kV reactors two Buchholz protections are connected in series
Table 19 Redundancy for tertiary connected reactors

BR
- Earth fault (inst+timedelayed)
FI - Overcurrent (inst+timedelayed)
- No compensated unbalance
- Compensated unbalance relay
relay
SCT - One group
NZ - One group
10.3.4 Question: What degree of redundancy is applied?

(e.g. 2 DC supplies, different tripping coils, 2 secondary
windings of CT, …)
For directly connected reactors all the utilities

(CA,ZA,ES,RO,SE,NO,GS,CN,IN,UK,AU) where two groups of protections are
provided use separate DC supplies , different trip coils and different CT cores for main
and back up protections. FR where only one group of protections are provided use only
single DC supply, single trip coil and single core for main and backup protections.
In case of tertiary connected reactors where mostly single set of protections are provided
single DC , single trip coil and single set of CT cores is used. FN where two groups of
158
protections are provided uses two DC supplies, different trip coils and two secondary
windings of CTs.
10.4 Fault types and protection limitations
10.4.1 Question: What types of fault are experienced and

how often do they occur? (e.g. percentage of the
occurrence of various faults with respect to the total
number of faults per year).
Types of faults reported in case of directly connected reactors are winding faults (Ph-g,
Ph-Ph, turn to turn), faults in bushings and auxiliary equipment, overload, oil
temperature/winding temperature high, mechanical defects and problems in CB. The
reasons for faults are moisture ingress in auxiliaries and bushings, over voltage, oil level
falling, dielectric failures, thermal, mechanical failures. The faults range frequency, from
0.5 per year to 5.6 per year.
In case of tertiary connected reactors one utility has reported inter turn faults, faults
related to auxiliary apparatus and erroneous trips.
Fault types, reasons and failure rates reported by responding utilities are given in Table
20 and Table 21 below.
Table 20 Types of faults on directly connected shunt reactors
Type of faults , reasons and how often they occur

CA
- Both phase to ground and 3 phase faults are experienced.
- Faults are mainly as a result of moisture ingress in the auxiliaries and
bushings. This results in insulation breakdown which can create inter-
ZA turn faults as well as phase to phase faults between windings.
- Some faults have been caused by animals being electrocuted.
- Frequency of faults is limited to about 3 or 4 a year.
- 2007 recorded seven reactor failures.
FR - Very few due to the small number of reactors.

ES - Problems in the shunt reactor circuit breaker that remained with two
poles closed and one pole open
RO - In the last 10 years 3 faults in 400 kV shunt reactors all located on the
high voltage bushings
- Total number of grid faults 1118 for year 2000 – 2007
SE
- Total number of shunt reactor faults so far is 45 which come to an
average of 5.6 faults/year. Out of these 39 are caused by cooling fan
159
Type of faults , reasons and how often they occur
problems.
- No major reactor faults experienced during 1984-2008

- No phase to phase or phase to ground faults.
NO - Some faults on the auxiliary equipment such as failure in the winding
temperature measurement.
- Approximate failure rate less than 0.5 fault /year (9 reactors)
- 4 times CB failures experienced when disconnecting the reactor.
GS - Winding and bushing breakdown : 3 %
- Faults experienced in reactors include earth fault in reactor windings,

turn-to-turn fault, phase-to-phase fault and earth fault of down-lead.
CN - Overload due to overvoltage, falling of oil level, high oil temperature,
high winding temperature and failure of cooling
- Turn-to-turn fault can cover about 70% of internal faults.
- Most of the failures are in line reactors
- Failure rates: Single phase units -0.63% and 3phase units – 0.9%
IN - Out of the above bushing failures are 40%, winding failures are 32.5%,
magnetic circuit failures are 15% and faults in terminals are 12.5%.
- Origin of faults: dielectric 80%, thermal 10%, mechanical 7.5% and
unknown 2.5%.
UK - Winding faults (inter-turn), less than 1% per year
- No reactor faults for many years now, notably after older 3 limb core
type units were replaced with 5 limb units.
AU
- One case of failure of one phase of the small tertiary connected air
cored unit due to an earth fault. In general performance in the past 10
years or so has been excellent.
Table 21 Types of faults on tertiary connected shunt reactors
Type of fault, Reasons for fault and how often they occur
BR Not available
- Most common fault is short-circuit (interturn/separate windings)

- These kind of faults have occurred 14 times in the period of
FI 1980...2007, when total amount of major faults in the same time period
is 30 (faults that relate to auxiliary apparatus e.g support insulators are
not included)
- In addition in 7 cases protection have done an erroneous trip
SCT - None of the SSE reactors have ever failed in all voltage levels
NZ - No recorded faults however BRY reactor is old and leaking. Going to
be removed in the short term.
160
10.4.2 Question: What percentage of each category of
these faults is not cleared correctly due to the
protection functions limitation?
In case of directly connected reactors mal operations due to mismatch in CT cores used
for REF and malfunctioning of back up impedance relays during switching and
unsuccessful auto reclosing due to improper tuning of neutral ground reactors are some of
the problems reported by some utilities. Some utilities have said there have been no
maloperations.
In case of tertiary connected reactors one country has reported some instances of
erroneous trip.
Table 22 and Table 23 below give summary of the responses received from different
responding utilities.
Table 22 Protection performance- directly connected shunt reactors

Reasons for failure of fault clearance
CA
- There have been experiences with CT mismatches for restricted earth
ZA fault protection as a result of different classes of CTs being used in
phase and in the neutral.
- This has caused incorrect protection operation.
FR ?
ES ?
RO - No failures of fault clearance so far
SE - No failures of fault clearance so far
NO - No failures of fault clearance so far
GS - No failures of fault clearance so far
CN - No formal statistic data showing the percentage.
- Sometimes CT ratios of cores used for restricted earth fault protection
become an issue when one is bushing CT and other is from the main
CT.
IN - In about 10% of the cases auto reclosing has been unsuccessful. The
reasons are not clear.
- There have been cases when backup impedance protection has
maloperated under some switching conditions.
UK - No statistics, almost all the faults were cleared correctly by Buchholz
protection.
- Recalling the failures of 3 limb core type units well over 10 years ago,
AU
these often did result in destruction of the reactor though protections
operated optimally. They have been replaced with 5 limb units
161
Table 23 Protection performance- tertiary connected shunt reactors
Reasons for failure of fault clearance
BR
- The most common fault is a short-circuit fault (interturn/separate
windings). These kind of faults have occurred 14 times in the time
FI period of 1980...2007, when total amount of major faults in the same
time period is 30 (faults that relate to auxiliary apparatus e.g support
insulators are not included).
- 7 cases protection have done an erroneous trip.
SCT ?
NZ ?
10.4.3 Questionnaire: Are you looking at new solutions to

resolve the above-mentioned limitations? If yes, please
describe.
Careful attention to specifications of CTs in new installations, proper tuning of earthing
reactors are some of the actions being taken by utilities that have faced problems of
restricted earth fault protection maloperations and unsuccessful auto reclosing. Some
utilities have reported that after they have started using numerical protections their built
in DR is helping them in better analysis. Table 24 below gives summary of the responses
received from responding utilities for directly connected reactors.
Table 24 New solutions for directly connected shunt reactors

New solutions to solve protection function limitations
CA
- Specifications for CTs are being carefully issued for new installations
to meet a requirement such as a class X or similar
ZA
- There have been experiences of reclosing onto faults as a result of
incorrectly tuned neutral earthing reactors. These are being tuned now
to the line capacitance
FR - No ongoing studies on this topic.
ES - Possibility to set the relays more sensitive is being studied.
RO - Not applicable
SE - Not applicable
NO - Not applicable
GS - Numerical relays for their added benefits.
162
New solutions to solve protection function limitations
CN - No
IN - After introduction of numerical relays with in-built disturbance
recorders analysis is simpler
UK - No limitations, second Buchholz relay is being considered.
AU ?
10.5 Control and monitoring of shunt reactors
10.5.1 Question: What types of control are being used for

switching in and out the shunt reactors? (e.g. Manual or
Automatic) If Automatic what philosophy is used?
The comments from responding utilities on the philosophy used for automatic switching
are given in Table 25 below. In case of directly connected reactors both manual and
automatic switching is practiced in various countries. Some explanation of the philosophy
used when automatic switching is used is also given. Automatic control in this context
means that the switching decisions are made for instance by voltage regulators.
Table 25 Types of control - directly connected shunt reactors

Manual Automatic Philosophy used if automatic
CA X ?
Automatic switching is only allowed where the device
ZA X (SVCs being switched can be completely compensated, for e.g. if
X
only) a 100MVA reactor is being switched out, the SVC must
have 100MVAR capacitance availability.
FR X ?
ES A device (Synchronized switching) that controls the
X
best instant to close the circuit breaker is used.
On increase of voltage on the bus bar command to
close the shunt reactor CB is issued automatically.
On decrease of voltage command to open the shunt
reactor CB is issued automatically. This is realized
RO with over voltage and under voltage relays
X
respectively, with independent time delay having
accurate and adequate pick-up/drop-off ratio.
The duration of the commands is limited, and after a

command, automation is blocked for an adjustable
163
Manual Automatic Philosophy used if automatic
time duration. The automation logic is provided with
lock-out against successive commands, and when
following protections operate: the shunt reactor
protection, busbar protection or transformer
protection operation. Unblocking has to be done by
operating personnel.
SE Manually switched in normal operation but

X X
automatic control at lower / higher voltage level
NO Manually switched in normal operation but
X X
automatic control at lower / higher voltage level
GS In most cases manual. Switchsync relay used in two
X
380kV substations
CN Only line reactors are that are permanently connected to
X
the line used
IN Switching is done as advised by regional load dispatch
X
center
Automatic Reactive Switching (ARS) system is used
to switching shunt reactors (and MSCs) at all voltage
levels to control HV voltage (275, 400 kV)
UK X Pre-fault (fine control): fine tune voltage, maximize
dynamic MVar reserves
Post-fault (Coarse control): quickly recover voltage

Normal control is manual.
There are some “very wide” settings that enable
AU
X automatic emergency operation (e.g. switch in at
extremely and dangerously high system voltage, or
trip at extremely low system voltage.
In case of tertiary connected shunt reactors in BR, SCT, NZ the switching is done
manually as per the advice from the control room. In FN manual local, manual remote,
automatic local (voltage regulator) and, in two special cases, automatic remote (regional
regulation) is used.
10.5.2 Questionnaire: How often the shunt reactors are

being switched in and out?
The frequency with which shunt reactors are switched in and out depends on loading
conditions. Seasons also have an effect on this. In most cases switching in and out of
reactors is done daily and in some cases once in a while. Table 26 and Table 27 gives
164
summary of the responses received for this question for directly connected and tertiary
connected shunt reactors.
Table 26 Frequency of switching - directly connected shunt reactors

Frequency of switching
CA ?
Most reactors are not switched regularly (< once a week). A few reactors
are switched to increase power transfer capability of the network and are
ZA typically switched twice a day. At the stations where automatic switching
takes place, reactors are switched 10 – 15 times a day (assuming normal
operating conditions)
FR ?
ES Reactors are switched in and out daily.
RO 130 switching per year.
SE Once or twice every day
NO Daily
GS In general during winter this is more frequent for bus reactors.
CN Shunt reactors are directly connected to transmission line without any circuit
breaker.
Almost daily according to system voltage condition as advised by regional
IN load dispatch center. Usually bus reactors are switched two or three times
each day. These switching’s are also seasonal.
UK On an average each reactor is switched once per day.
AU Typically, daily.
Table 27 Frequency of switching - tertiary connected shunt reactors

Frequency of switching
BR Daily
FI Depends on voltage stability, many of the reactors are switched in / out daily
SCT Once a day, tend to be in the night.
NZ Rarely; only when there is light loading.
165
10.5.3 B5.3 Questionnaire: Do you use synchronized
switching (point on wave switching)? How are they
applied?
In case of directly connected reactors many utilities use synchronized switching mostly
for opening of the CB. In some cases they are also used for closing of the CB. Some
utilities who do not use them at present are considering there usage due to increased
failure rates observed in the breakers. Table 28 below gives summary of the responses
received from responding utilities for directly connected shunt reactors.
Table 28 Point on wave switching - directly connected shunt reactors

Use of Synchronised switching
CA Synchronized Switching is used
ZA Applied to 765kV reactors. Switching done on opening of the breakers.
FR Synchronized switching used on 400kV and 225kV shunt reactors
ES Synchronised switching used in independent pole mode..
Synchronized switching is used for manual and automatic closing and
RO opening.
SE Synchronised switching is used. Adaptive or fixed to current zero crossing

Synchronized switching is applied on most reactors . The function was
NO originally used for operator controlled deenergizing of the reactor. Now
synchronised switching is also used when energising.
GS Synchronised switching used.
CN Synchronized switching is usually not used in normal transmission lines, but
it is used more and more in double circuit lines in recent years.
IN There have been failures of CB during operation. Synchronised switching
isC is being considered to avoid these failures.
UK Synchronised switching is used.
AU Synchronised switching (POW) is used. POW relay is set to prevent re-
strike on open and close operations.
In case of tertiary connected reactors no utility has reported use of synchronized

switching. BR, FI, SCB and NZ have specifically stated that they do not use
synchronized switching for tertiary connected shunt reactors.
166
10.5.4 B5.4 Question: What parameters are monitored
(e.g. I, U, Q, Oil and winding temperature, dissolved gas,
bushing residual current, fire detectors,...) and what
actions are taken?
In case of directly connected reactors most countries monitor I, V, Q, oil temperature,
winding temperature, dissolved gas, bushing residual current. Actions taken are providing
alarm and trip.
Table 29 and Table 30 given below gives summary of responses received for this
question both for directly connected and tertiary connected shunt reactors.
Table 29 Reactor parameters monitored - direct connected shunt reactors

Parameters monitored and actions taken
CA
- Tripping of the reactor
- Winding temperature – alarm at 95deg, trip at 105deg
- Oil temperature - alarm at 95deg, trip at 105deg
- DGA – New reactors will be provided with online DGA analyzers.
Existing ones will be retrofitted with DGA analyzers.
- Fibre optic sensors – New units have these to measure core temperature
ZA as well as oil leaving the top of the winding
- Bushing monitoring – Two new units have on line tan delta and
capacitance monitoring as these have new bushings of which there is no
history
- Fire detectors – Pilot project testing out fire detectors on a transformers
going on. If accepted, it may be adopted on reactors too.
- Oil temperature - reactor delayed trip
FR - Neutral current - reactor instantaneous trip
- Tank current - reactor instantaneous trip
- Phases currents (20kV) - transformer instantaneous trip
ES - I, U (for the synchronized switching device)
- Oil and winding temperature
- Monitoring oil temperature (alarm/ trip)
- Monitoring windings temperature (alarm/trip)
- Monitoring oil level (alarm)
- Monitoring of the cooling system (pumps, fans)
- Monitoring electrical parameters: U (per phase), I (per phase), active
RO power (P), reactive power (Q)
- Monitoring the dissolved gas and water ( H2, H2O) – alarm
- Monitoring bushings (change of capacitance) – alarm
The monitoring system provides information about main isolation condition.

(humidity, operating time, hot-spot temperature, operation life time)
SE - I , oil and winding temperature
167
- I, Q
NO - Oil temperature and winding temperature
- Monitoring the dissolved gas and water (H2 , H2 O)-alarm
GS - All parameters monitored and actions taken according to the nature of

the fault.
- Voltage, current, temperature of oil or/and winding, residual current
flowing through neutral reactor are common parameters that are
monitored.
- For signals such as high temperature of oil and windings and dissolved
CN gas, which may lead to damage to the reactor alarm is issued and at the
same time the circuit breaker is tripped.
- For some signals, such as overload of neutral point reactor, only alarm is
issued. The operator has to take steps according to predefined
instructions.
- All above parameters except Bushing Residual Current are monitored
- DGA of oil from reactor and bushings is being done. Based on the
IN results trend in DGA, a national & regional transformer committee
comprising of expert engineers recommend the test plan, test frequency
etc. On-line dry-out systems are used. DOMINO test is conducted on
reactors every year. SFRA tests are conducted upon requirement to
compare traces with base signatures.
- I, U, Q, oil level, WTI, all the alarms and indications generated by
UK protection functions.
- WTI switches cooling system on/off, and trip reactors if winding
temperature is detected to be higher than a set value.
AU - U & I, with manual investigation of any abnormalities.
Table 30 Reactor parameters monitored - tertiary connected shunt reactors

BR Tripping of shunt reactors
- Q for reactor
FI - U for 20 kV switchgear busbar
- I for feeding main transformer bay
No special actions required.
SCT - Winding temperature and usual protection devices/relays.
- Tripping of the reactor
NZ ?
168
APPENDIX
A. EXAMPLE OF CONTROLLED SWITCHING OF A 500KV
SHUNT REACTOR IN JAPAN
A.1 Introduction
It is known that when the switchgear interrupts a shunt reactor current, it generates re-
ignition, causing high overvoltages [ref 25] ~[ref 30]. Tokyo Electric Power Company
has constructed a 500kV underground substation in an urban area. Having long-distance
cables connected, the substation is equipped with 500kV shunt reactors. 550kV one-break
circuit breaker is used to switch the shunt reactor current. The controlled switching is
employed for this 500kV shunt reactor switching. Full-scale shunt reactor current
interruption tests of the 550kV gas circuit breaker were carried out for this study at a
high-power laboratory.
A.2 Reactor current interruption test circuit

Figure 101 shows a test circuit of a single phase 500kV reactor current interruption.
2.65H was interrupted by a 550kV one-break gas circuit breaker (CB). By taking account
of the actual Gas Insulated Switchgear (GIS), with a small load side capacitance CL, tests
were carried out at 1.8 kHz and 2.5 kHz. The capacitance CS at the power supply side
was determined to be about 10 times CL, so that high overvoltages would be generated at
the reactor terminal in the event of re-ignition [ref 31].
The test conditions are summarized in Table 31.
Figure 101 Single phase test circuit for interrupting 500kV reactor current.
169
Table 31 Test conditions for single phase circuit for interrupting 500kV reactor current.
No. CB conditions Circuit conditions
Operating hydraulic SF6 gas Applied Interrupting Load side
pressure pressure voltage current frequency
(MPa) (MPa) (kV) (A) (kHz)
1 0.6 (rated) 318 (500/√3) 381 1.80
2 31.5 (rated) 0.6 (rated) 318 (500/√3) 381 2.50
3 0.6 (rated) 289 (500/√3) 347 1.80
4 0.55 (alarm) 318 (500/√3) 381 1.80
A.3 Relation between opening phase angle and re-ignition

Figure 102 shows an example of an oscillogram obtained in the interruption tests
mentioned above. The circuit breaker started contact separation at point Q. At the first
current zero crossing R, the reactor current was temporarily interrupted and directly
thereafter, a re-ignition took place, which was followed by a flow of power-supply-
frequency reactor current.
Finally, interruption was completed at point S. As described above, the re-ignition
generation is closely related to time T0 between contact separation point Q and the first
reactor current zero crossing R.
Figure 102 Oscillograms obtained in interruption test.

Thus, with the test circuit in Figure 101, a series of interruption tests were carried out
with different values of T0. The results are shown in Figure 103.
170
Now, the longer T0 is, the longer the distance is between the circuit breaker contacts at
the first reactor current zero crossing. At the longer distance between the contacts, the re-
ignition takes place under a higher voltage between contacts;
Thus, re-ignition overvoltage is higher. Tests were carried out to obtain the relation
between time T0 and the voltage between the contacts upon re-ignition. The results are
shown in Figure 104.
Figure 103 Relation between probability of re-ignition and T0.
Figure 104 Voltage between contacts at re-ignition.
171
A.4 Application of controlled switching
Figure 105 illustrates how the controlled switching in the 550kV breaker acts to suppress
re-ignition overvoltages. Figure 105(a) shows the scheme of the controlled switching.
Figure 105(b) shows the time chart of the controlled switching. The control unit outputs
the contact separation command to the circuit breaker at an appropriate point of time.
Figure 105 Illustration of re-ignition suppression by controlled switching.

The results showed the following:
(1) Number of operations and temperature variation, showed no significant influence
on the opening time.
(2) The opening time variations with control voltage fluctuations can be compensated
with the control system (after establishing the relationship between the control
voltage level and the separation time)
172
(3) Even if the opening time variations with hydraulic pressure variations taken into
account, the contact separation points are in a certain range of dispersion.
This dispersion of the contact separation points on the 550kV one-break circuit
breaker was limited within ±1.5ms.
Using the test results in Chapter 15.3 and the results of the investigation on the
dispersion of contact separation points described above, a controlled switching
was achieved for the 550kV one-break gas circuit breaker.
It is shown in Figure 106, which contains the test results for the relation between
the contact separation point and the voltage between the contacts at re-ignition
shown in Figure 104.
Figure 106 shows followings:
a) The target arcing time, which is between contact separation target point C
and the interruption point E is 11ms;
b) The latest point D in the contact separation point dispersion range Ts is
not in the contact separation range =9.0ms, in which re-ignition takes
place. So that the interruption is completed without re-ignition at the point E.
c) At the earliest point B in the contact separation point range Ts, the voltage
between the contacts at the re-ignition is 0.2 p.u.
The overvoltage at this re-ignition is estimated to be approximately below
1.2p.u (overvoltage = terminal voltage + voltage between contacts = 1.0 +
0.2 = 1.2pu). This is a sufficiently safe value in terms of equipment
insulation.
d) In the contact separation point dispersion range Ts of B to D, overvoltages
are suppressed to approximately below 1.2p.u. Therefore, the controlled
switching is quite effective.
173
Figure 106 Opening phase control achieved on 550kV one-break circuit breaker.
A.5 Conclusion
1. The relation was obtained between the contact separation point and the probability of
re-ignition as well as the voltages between the CB contacts which produce re-ignition.
2. Even with the dispersion of circuit breaker operations taken into account, there was a
contact separation point that did not cause high re-ignition overvoltages. Thus, by setting
a contact separation at this point, high re-ignition overvoltages could be suppressed.
3. It was proven that even with re-ignition; there was no high-frequency arc extinction,
that is, no voltage escalation due to multi-re-ignition. Overvoltage’s due to current
chopping where at a safe level in terms of equipment insulation.
The investigation shows that the reactor current could be interrupted safely by
suppressing re-ignition surges by employing controlled switching.
174
Figure 107 Shunt reactor installed in the substation.
175
B. PROTECTION OF 800KV 3 PHASE SHUNT REACTOR
PROVIDED WITH SPARE SINGLE PHASE REACTOR
765 kV transmission line generates large capacitive charging MVAR and, therefore, to
limit over voltages under lightly loaded condition and also to limit switching
overvoltages.
Three 80/110 MVAR, 765 kV line/bus type shunt reactors with a spare reactor are used
in the 765 kV network.
Figure 108 shows a single line diagram of a typical 765 kV substation, showing lines
with reactors. An advantages of the configuration is that the spare reactor can be used for
any one of the reactors, whichever is connected to same bus irrespective of, if it is
switchable or non-switchable i.e. it is provided with a circuit breaker or not.
176
Figure 108 Single line diagram 765 kV substation
177
Figure 109 shows the three line diagram of two banks of 765 kV line reactors three 80
MVar reactors connected with a 765kV auxiliary bus to which the spare single phase
reactor is also connected.
The protection of the EHV shunt reactor in this case is achieved by using CT switching
arrangement for all the single phase reactors including the spare reactor. Since the relays
have limited analogue inputs for current and voltage the switching for the analogue inputs
is done externally when spare shunt reactor is placed into service.
178
Figure 109 Three line diagram 765 kV substation
179
C. PROTECTION OF SHUNT REACTOR AUXILIARY WINDING
Figure 110 and Figure 111 show two different auxiliary winding applications, see [ref 8]
and [ref 34]. It is very important to keep the shunt reactors in service so it is very
important that the design minimizes faults on the low-voltage distribution. Segregated
cables can be used from the reactor auxiliary winding to the low-voltage distribution to
prevent faults from occurring. The low-voltage distribution equipment may have
additional design measures because of this application. Faults that do occur must be
detected and cleared in a coordinated manner. It should be noted that the auxiliary power
winding, although drawing power from the main winding, does not affect the differential
function 87.
87
CB-HV
52
49
94
62 CB-LV
50N
63
50 51 50N 51N
94
CB-HV
CB-LV 52
50N
50/51
51N
DEVICE DESIGNATIONS
87 DIFFERENTIAL RELAY 52 52 52 52
49 THERMAL RELAY
63 SUDDEN PRESSURE RELAY
50 INST. OVERCURRENT RELAY
TO DISTRIBUTION
51 TIME OVERCURRENT RELAY
62 TIME DELAY RELAY
94 AUX. TRIPPING RELAY
Figure 110 Grounded-wye Auxiliary Power Winding for Low-voltage Distribution
Figure 110 shows the auxiliary power winding in a grounded-wye configuration. In

Figure 110, the instantaneous overcurrent 50 and 50N functions at CB-LV are time-
delayed (62) long enough to allow the downstream supply feeders to operate first for
faults on the feeders or beyond. Even with this delay, these functions initiate fast tripping
of CB-LV for a failure to clear a downstream fault. The time-overcurrent functions 51
180
and 51N at CB-LV are also coordinated with downstream protection but act as a backup
by tripping CB-HV and removing the entire shunt reactor from service.
87
CB-HV
52
59N ALARM
51
CB-LV 52
52 52 52 52
TO DISTRIBUTION
Figure 111 Ungrounded Auxiliary Power Winding for Low-voltage Distribution
In Figure 111, the auxiliary power winding is shown in an ungrounded configuration. In
this case, it may have been determined that keeping the reactor in service for a ground
fault on the auxiliary winding is more important than clearing the fault. A broken-delta
VT with an overvoltage relay (59) scheme can be used to detect the ground fault and
alarm to a dispatcher with intent to repair the problem as soon as the reactor can be
removed from service in a planned manner. Note that unlike Figure 110, there is no
circuit breaker in the reactor auxiliary power winding output and there is also an
intermediate distribution transformer. Some shunt reactor designs cannot tolerate an
open auxiliary winding much like a CT cannot tolerate an open secondary. The
intermediate transformer acts as a protective burden to the auxiliary power winding. A
time-overcurrent 51 function trips the entire reactor off line for a problem in the
intermediate transformer and beyond. For the scheme in Figure 111, power fuses protect
the area between the reactor and the 51 relays. Fuses are used because they are less prone
to be opened accidentally causing an unintentional open circuit condition. Downstream
distribution protection must still be coordinated with the fuses and phase 51 relays on the
source side of the intermediate transformer.
181
D. APPLICATION OF 500 KV SHUNT REACTOR WITH
AUXILIARY WINDING SYSTEM IN SWITCHING STATION
D.1 Introduction
It is known that the charging current is proportional with the degree of the voltage and the
length of the transmission line. For this reason, overvoltage always emerges at the end of
the long unloaded line, mainly due to the large amount of capacity accumulated along the
line. Therefore, in order to limit the overvoltage within a certain range, it is necessary to
build adequate numbers of switching stations equipped with reactors along the
transmission lines. But if the switching station is located at such a remote area without
reliable power supply, and also it is far from the load centre, then a reliable source of
service power need to be supplied.
An example hydropower station possesses an installed capacity of 3300 MW. The

electrical power is delivered to area A through three 500 kV transmission lines with
length of 470 km each, and then the power is forwarded through two 500 kV transmission
lines to other places.
To regulate the voltage level, a switching station has been established between generation
and area A, 470 km away from the generation.
There is no other reliable power supply in this region.
D.2 Solutions to the Problem
After comprehensive investigation, the following electrical configuration arrangements

for the service power of the switching station were proposed.
 Introduce one 35 kV line from the regional power grid, as the main backup power
supply of the switching station, and the capacity of station transformer is 630
kVA.
 Install a set of diesel generators which is started manually, as the assisted backup
power supply, and its capacity is 40 kVA.
 Extract electrical power energy through the auxiliary winding system of the two
groups of 500 kV high voltage shunt reactors, as the primary power supply for the
switching station, and the capacity of each shunt reactor is 60 MVAr.
D.3 The main structure and technical parameters of the

shunt reactor
182
There are two groups of ordinary high voltage shunt reactors, and two groups with
auxiliary winding system through which the power energy is extracted to feed the
switching station.
Reactor structure
The structure of the reactor is the 3-pole form, consisting of a main pole core with air gap
and two secondary pole cores. The 550 kV winding is wrapped around the main pole
core, composed of two winding modules with series connection, and at the axial
intermediate point of the two modules there is a third terminal. The energy extraction
winding consists of the two windings wrapped around the two secondary pole cores with
parallel connection, whose voltage depends on not only the turns ratio of the main
winding and secondary winding but also the magnetic flux through the two secondary
windings. The three-phase terminals of the energy extraction windings connect to three
independent power cables, with a load break switch.
The installation
Breakers, grounding switches, fuses, a 6.5-kV intermediate (distribution) transformer,
load switches, lightning arresters and all the other necessary protection, control and
metering devices are installed in a sealed concrete component box. To avoid phase to
phase fault, three independent cables are used to interconnect the energy extraction
windings of the three shunt reactors with the high-voltage buses of the service power
distribution box. After entering the distribution box, the power energy extracted goes
forward through lightning arrester, grounding switch, SF6 circuit breaker, fuses and then
reaches the intermediate transformer, in which the voltage is stepped down to 400/231V
from 6.5 kV. Passing the intermediate transformer and then a main load switch, the
power energy flows into the low-voltage buses of the service power distribution box.
Through three separated power cables, the power energy is forwarded to the electricity
distribution room of the control building. To ensure quality of secondary voltage, a
transformer with on-load tap changer is selected as the intermediate transformer, and the
range of regulatory voltage is ± 2 × 2.5%.
Main equipment parameters
Shunt Reactors
Rated capacity: 60 MVAr
Rated power of the auxiliary winding: 167 kVA
Rated voltage of the main winding: 550 kV
Rated voltage of auxiliary winding: 6.5 kV
Rated current of the main winding: 189 A
Rated current of auxiliary winding: 44.5 A
Weight: 78000kg
Intermediate Transformer
Rated capacity: 500 kVA
Rated voltage of the primary winding: 6.5 kV
Reactance: 1681 Ω
183
D.4 Measures for protection
As the high-voltage shunt reactors are attached to the 500kV line directly, this leads to a
disadvantage that the shunt reactors can only be isolated by tripping the circuit breakers
at the two ends of the line, which may disturb the operation of power grids. Therefore it
is extremely important to ensure reliable operation of the reactor.
Besides the ordinary structure, the shunt reactor has an additional auxiliary winding
system, which increases exposure to reactor faults. To maintain the operation of this
special reactor, the manufacturer should be requested to take effective measures to avoid
the internal faults of the reactor, especially when a fault occurs on the auxiliary winding
system, it should not spread to the main winding system.
The faults on the external devices attached to the auxiliary winding system should be
avoided outside the reactor. Three independent cables are used to join the auxiliary
winding systems of each reactor with the external devices. And these cables are separated
with a distance aimed at eliminating the short-circuit fault between these cables ahead of
the SF6 breaker.
To eliminate the impact on the reactor in case a fault happened on the downstream device
of the load switch, overcurrent and grounding protective devices have been installed, and
a high-voltage fuse is also attached to the load switch as another overcurrent device;
Thermal and overcurrent protection have been installed on the main breaker of the
secondary circuit of this intermediate transformer. Moreover, fuses have been set to
isolate the faults on the 400V feeder circuit. For protection of the high-voltage reactor,
preventing the overvoltage of the main winding from entering energy extraction and
power distribution system, a lightning arrester is installed in the cable of auxiliary
winding system.
D.5 Operating history

During the elapsed operating time only the breaker of feeder circuit connecting to the
secondary winding of the intermediate transformer has been tripped once because of the
failure in electrical motor of the station water pump. The power energy extracted from
500 kV power grid system by two groups of auxiliary winding systems guarantee the
reliability of service power of the switching station.
D.6 Conclusions
Due to the low reliability of regional power grid, through the 35 kV cables, the power
flowing into the power service transformer is often interrupted, so the reliability of the
power service of the switching station cannot be maintained only by depending on the
regional power grid. Locating in remote areas, without reliable power supply sources,
184
these switching stations should make use of shunt reactor with auxiliary winding system
to obtain reliable power supply.
With enhanced energy extraction capacity, the additional power energy can be distributed
to the adjacent areas.
During the system design it is necessary to ensure the reliable operation of the high-
voltage side of the power grid system even when a fault occurs on either the energy
extraction winding system or the downstream distribution system. Therefore, on one
hand, the manufactory should ensure the reliability of internal elements of the reactor in
the design and manufacturing procedure, that is, the high-voltage part of the reactor
should not be damaged by the fault either on energy extraction winding system or on
downstream distribution system. On the other hand, separated (segregated) power cable
needs to be adopted to avoid faults happening on cables of the energy extraction.
Interrupting and the protection devices should be appropriately equipped to restrict the
fault on distribution system within the downstream area of the circuit breaker.
E. AUTOMATIC REACTIVE SWITCHING IN UK
Automatic Reactive Switching (ARS) system, see Figure 112, is used to switching shunt
reactors and mechanical switching capacitors (MSCs) at all voltage levels to control HV
voltage (275, 400 kV), [ref 38]. Pre-fault (fine control): fine tune voltage, maximize
dynamic MVAr reserves Post-fault (coarse control): quickly recover voltage.
In the UK, the Automatic Reactive Switching (ARS) system shall monitor and control the
400/275 kV system voltage by switching shunt reactive plant at a substation.
The ARS system shall be capable of controlling up to 4 HV (high Voltage) reactive

plants and an unspecified quantity of other reactive plants at a site.
The ARS system shall include the functions of HV-ARS and LV-ARS. The HV-ARS and
the LV-ARS shall control the HV and LV reactive plant respectively.
The ARS system shall be capable of handling connection paths containing up to eight
switchgears in series. Any given switchgear shall be capable of being in more than one
connection path. A connection path is the link between a plant item and an associated
busbar. A connection path can be formed through switchgear and transformers as well as
other plant items.
There shall be 3 main control points for an ARS system. These are the local control point
(LCP), Substation Control Point (SCP) and Remote Control Point (RCP). A selection
facility shall be provided to enable control to be switched from the LCP to the SCP/RCP.
It shall be possible to select to one of the five pre-set voltage targets for the ARS system.
The target voltages shall be settable in the range of ±10% of HV nominal system voltage.
When voltage target changes, the new target shall ramp to the new target at a preset rate.
However, the indication of the voltage target change shall take effect immediately at all
185
the control points. This ramping function is to avoid the fast reactive switching due to the
voltage target change.
For Coarse Voltage Control (disturbed operation), ARS provides fast switching of HV
and LV reactive plant to correct large HV and LV voltage variations.
For Fine Voltage Control (normal operation), ARS provides slow switching of LV
reactive plant to maintain the voltage within set limits of user selected target voltage.
The power system voltage inputs to the ARS system shall be supplied conventional
voltage transformers with an output of 63.5 V rated secondary. Only single-phase voltage
measurement shall be used for the ARS system. A balanced load and voltage shall be
assumed at all times.
All voltage measuring transducers (if applicable) shall be full range 4 - 20 mA true rms
The average of all the HV group voltages shall be displayed as HV voltage on the local
control panel with an accuracy of at least ±0.5 kV and a resolution of 0.1 kV.
Fine Voltage Control mode shall monitor the HV voltage and carry out slow switching of
LV reactive plant to keep the HV voltage within a narrow dead-band around a selected
voltage target.
The thresholds and dead-bands associated with this control mode are detailed as follows:
186
Figure 112 Automatic Reactive Switching (ARS), relationship of thresholds and dead-bands for Coarse
and Fine Voltage Control Modes of the combined HV/LVARS
MHVVS : Measured HV Voltage Step by LV reactive plant

HVSM : HV Safety Margin
HHVM : High HV Margin
LHVM: Low HV Margin
VLHVM: Very Low HV Margin
FC: Fine Control

CC: Coarse Control
Note that VHHV and EHHV are fixed value, not linked to the Target Voltage
187
Fine Voltage Control Mode - Fairly Low HV
Refer to Figure 112, when the HV falls below the Fairly Low HV threshold “FLHV(FC)”
no LV MSCs/Reactors shall be switched out/in by the Fine Control Mode. LHV(FC)
shall be set to VT - LHVM + MHVVS1 + HVSM.
Fine Voltage Control Mode - Low HV

When the HV stays below the Low HV (LHV) threshold for a preset period of time "Low
V Initial Switching Delay (LHVIS)", a LV or tertiary reactor/MSC shall be switched
out/in. The threshold for the LHV shall be set to VT - LHVM.
If the voltage remains below the LHV, another LV or tertiary reactor/MSC shall be
switched out/in at intervals of Low HV Subsequent Switching Delay (LHVSS seconds)
subject to flicker inhibit times until the voltage rises above the threshold.
Fine Voltage Control Mode - Fairly High HV

When the HV exceeds this Fairly High HV setting “FHHV(FC)”, no LV SCs/Reactors
shall be switched in/out by the Fine Control Mode. HHV(FC) shall be set to VT +
HHVM - MHVVS1 - HVSM.
Fine Voltage Control Mode - High HV

When the HV stays above the High HV (HHV) threshold for a preset period of time
"High V Initial Switching Delay (HHVIS)", a LV or tertiary MSC/reactor shall be
switched out/in. The threshold for the HHV, shall be set to VT + HHVM.
If the voltage remains above the threshold HHV, another LV or tertiary MSC/reactor
shall be switched out/in at intervals of High HV Subsequent Switching Delay (HHVSS
seconds) subject to the flicker inhibit times until the voltage falls below the threshold.
If the voltage has been outside the Low or High HV thresholds for the required delay
time, but no switching can take place due to the large value of MHVVS1, an indication,
"Fine Control Not Possible" shall be raised. The indication shall be cleared when the
switching action subsequently takes place or is no longer required, or if Fine Control is
deselected.
Coarse Voltage Control Mode

LVARS shall be involved in the Coarse Voltage Control Mode of the ARS system by
switching LV Reactive Plant with a preset time delays to keep the HV within preset dead-
bands around a selected target. The coarse voltage control usually has wider dead-bands
and shorter time delays than the fine voltage control.
Coarse Voltage Control Mode – LVARS Fairly Low HV

When the HV falls below the LVARS Coarse Control Fairly Low HV threshold
“FLHV(CC)1” no LV MSCs/Reactors shall be switched out/in by the coarse control
mode. FLHV(CC)1 shall be set to VT - VLHVM + MHVVS1 + HVSM.
188
Coarse Voltage Control Mode – Very Low HV (VLHV)
When the HV stays below the Very Low HV threshold “VLHV” for a preset period of
time "LV Plant Switching Delay (LVPSD)", LV reactors/MSCs shall be switched out/in
at intervals of a preset period of time "LV Plant Subsequent Switching Delay (LVPSS)",
until the HV rises above the threshold. The threshold for VLHV, shall be set to VT -
VLHVM.
Coarse Voltage Control Mode – LVARS Fairly High HV Inhibit

When the HV exceeds the Coarse Control Fairly High HV threshold “FHHV(CC)1” no
LV MSCs/reactors shall be switched in/out by the Coarse Control Mode. FHHV(CC)1
shall be set to VHHV – MHVVS1 - HVSM.
Coarse Voltage Control Mode – Very High HV

When the HV stays above this Very High HV (VHHV) threshold for a preset period of
time "Very high HV Initial Switching Delay (VHHVIS)", LV or tertiary MSCs/reactors
shall be switched out/in at intervals of a preset period of time "Very High HV Subsequent
Switching Delay (VHHVSS)", until the voltage falls below the threshold. The threshold
for this VHHV shall be a fixed value.
Coarse Voltage Control Mode – Extra High HV

When the HV stays above this Extra High HV (EHHV) threshold for a preset period of
time "Extra High HV Initial Switching Delay (EHHVIS)", LV or tertiary MSCs/reactors
shall be switched out/in at intervals of a preset period of time "Extra High HV
Subsequent Switching Delay (EHHVSS)", until the HV falls below the threshold. The
threshold for EHHV shall be a fixed value.
Coarse Voltage Control Mode – HVARS Fairly Low HV

When the System HV falls below the HVARS Coarse Control Fairly Low HV threshold
“FLHV(CC)2” no HV MSCs/Reactors shall be switched out/in by the Coarse Control.
FLHV(CC)2 shall be set to VT - VLHVM + MHVVS2 + HVSM.
Coarse Voltage Control Mode - Very Low HV (VLHV)

If the System HV is below Very Low HV threshold (VLHV) for a period of Very Low
HV Switching Delay "VLHVSD" the HV reactor/MSC shall be switched out/in. The
threshold for VLHV shall be set as VT - VLHVM.
Coarse Voltage Control Mode – HVARS Fairly High HV

When the System HV exceeds the HVARS Coarse Control Fairly high HV threshold
“FHHV(CC)2” no HV MSCs/Reactors shall be switched in/out by EMS/Manual Control
Mode. FHHV(CC)2 shall be set to VHHV - MHVVS2 - HVSM.
189
F. ABBREVIATIONS
ANSI Numbers and abbreviations in Table 32 and Table 33, see [ref 12] and [ref 13].
Table 32 Function descriptions with ANSI, IEC61850-5, IEC61850-7-4 and IEC Symbols
ANSI Code IEC 61850-5 IEC 61850-7-4 IEC Symbols Function description
IEEEC37.2 Code Code
21P PDIS PDIS/PSCH Phase distance
21G PDIS PDIS/PSCH Ground distance
25 RSYN RSYN Synchrocheck
27P PTUV PTUV Phase Undervoltage
27X PTUV PTUV Auxiliary Undervoltage
32 PDPR PDOP/PDUP Directional power
protection
37 PUCP PTUC/PDUP Undercurrent protection
40 PUEX PDUP/PDIS Field loss protection
46 PPBR PTOC Reversal phase or phase
balance current
protection
47 PPBV PTOV Phase sequence voltage
protection
49 PROL/PSOL PTTR Thermal Image
(overload)
50P PIOC PIOC Phase Instantaneous
Overcurrent
50G PIOC PIOC Ground Instantaneous
Overcurrent
50hr PIOC PIOC Instantaneous
overcurrent with
harmonics restrain
50Q PIOC PIOC Negative Sequence
Instantaneous
Overcurrent
50N PIOC PIOC Neutral Instantaneous
Overcurrent
50P BF RBRF RBRF Phase Instantaneous
Overcurrent, Breaker
Failure
50N BF RBRF RBRF Neutral Instantaneous
Overcurrent, Breaker
Failure
190
50EZ PIOC PIOC End Zone
50HS PIOC PIOC Switch On To Fault
50NU/51NU PIOC PIOC Wye neutral current
unbalance (capacitors)
50STUB PIOC PIOC Stub protection
50T PIOC PIOC Transformer tank
protection
51 PTOC PTOC Phase Time Overcurrent
51N PTOC PTOC Neutral Time
Overcurrent
59P PDOV PTOV Phase Overvoltage
59G PDOV PTOV Ground Overvoltage
59Q PDOV PTOV Negative Sequence
Overvoltage
59X PDOV PTOV Auxiliary Overvoltage
66 PSMU PMRI Excessive number of
start-ups protection
67P PDOC PTOC Phase Directional
Overcurrent
67N PDEF PTOC Neutral Directional
Overcurrent
67G PDOC PTOC Ground Directional
Overcurrent
68 RPSB RPSB Power Swing Blocking
78 PPAM PPAM Out-of-Step Tripping
79 RREC RREC Autorecloser
81M PFRQ PTOF Overfrequency
81m PFRQ PTUF Underfrequency
81df/dt PFRQ PFRC Frequency rate of change
85 RCPW PSCH Teleprotection / Direct
intertrip
87L PLDF PDIF Restrained Line
Differential
87N PNDF PDIF Restricted earth fault
87P PPDF PDIF Phase comparison
87R PTDF PDIF/PHAR Restrained Reactor
Differential
87T PTDF PDIF/PHAR Restrained Transformer
Differential
87BB PBDF PDIF/PDIR Restrained Bus
191
Differential
50/87B PBDF PDIF/PDIR Unrestrained Bus
Differential
OSC RDRE RDRE Disturbance recording
ER RDRS RDRS Events recording
FL RFLO RFLO Fault locator
46 PPBR PTOC Current Unbalance
46BC PTOC PTOC Broken conductor detection
50TCL YLTC YLTC Tap changer lock
64 PHIZ PHIZ Zero sequence overvoltage
52 a,b XCBR XCBR Circuit breaker (CB) control
50DD PIOC PIOC Current Disturbance
Detector
50/74 PIOC PIOC CT Trouble
CILO CILO Interlocking
CSWI CSWI Control function
MET MMXU MMXU Measuring functions (I,
U, f, P,Q)
MET MMTR MMTR Metering function
CPOW CPOW Point-on-wave breaker
controller
Table 33 Abbreviations used

Abreviations Description
AC Alternating Current
ARS Automatic Reactive Switching
BBP Busbar Protection
BCU Bay Controller Unit
BFP Breaker Failure Protection
C Current
CB Circuit Breaker
CT Current Transformer
DC Direct current
EHV Extra High Voltage
G Ground
GIS Gas Insulated Switchgear
192
GOOSE Generic Object Oriented Substation Event
GPS Geographic Position System
GUI Graphical User Interface
H High
HV High Voltage
HW Hardware
HMI Human Machine Interface
I Current
IDL Instantaneous differential currents
IDMT Inverse Definite Mean Time
IED Intelligent Electronic Device
L Low
MV Medium voltage
MSC Mechanically Switched Capacitors
MTBF Mean Time between failures
MTTR Mean Time to Repair
N Neutral
OC Over current
OGST Oil gas separation tank
OLTC On load tap changer
PD Partial Discharge
PC Personal Computer
PLC Power line carrier
POW Point On Wave
REF Restricted Earth Fault
TCS Trip Circuit Supervision
TRV Transient Recovery Voltage
RTU Remote Terminal Unit
SCADA Supervisory Control and Data Acquisition
SW Software
U Voltage
193
V Voltage
VSR Variable shunt reactor
VT Voltage transformer
SO System Operators
194
G. REFERENCES
G.1 Books
[ref 1] B. Stenborg, Elkraftsystem del 1, Systemet och dess lugndriftstillstånd
[ref 2] B. Stenborg, Elkraftsystem del 2, Analys av onormala tillstånd
[ref 3] Numeriska metoder, Lärobok 1, Peter Pohl, Gerd Eriksson, Germund Dahlquist William D.
Stevenson, Jr., Elements of power system analysis, Fourth Edition
[ref 4] Kompendium I ELKRAFTTEKNIK av Fredrik Gustavsson m.fl., KTH 1991
[ref 5] Elkraftsystem, Systemet och dess Analys, Gassam Batal, Institutionen för elektronik,
Mälardalen University 2003
[ref 6] Gerhard Ziegler, Numerical Differential Protection, Principles and Applications, March 2005,
ISBN 3 89578-234-3
G.2 Technical Papers
[ref 7] HV SHUNT REACTOR SECRETS FOR PROTECTION ENGINEERS, 30th Western

Protective Relaying Conference, Spokane, Washington, October 21-23, 2003 by Zoran Gajic,
Birger Hillström, Fahrudin Mekic
[ref 8] IEEE Std C37.109-2006, IEEE Guide for the Protection of Shunt Reactors
[ref 9] CIGRE WG report 34.01 “ Reliable fault clearance and back-up protection”IEEE Std
C37.109-2006, IEEE Guide for the Protection of Shunt Reactors
[ref 10] International IEC, Standard 60270, High-voltage test techniques – Partial discharge
measurements
[ref 11] IEEE Guide for the Detection and Location of Acoustic Emissions from Partial Discharges
in Oil-Immersed Power Transformers and Reactors, C57.127-2007
[ref 12] IEC 61850-5, First edition 2003-07
195
[ref 13] IEC 61850-7-4, Edition 2.0, 2010-03
[ref 14] IEEE C62.92-1989, IEEE Guide for the application of Neutral Grounding in Electrical Utility
Systems. Part II – Grounding of Synchronous Generator Systems
[ref 15] CIGRE B5.05 Report: Modern Techniques for Protecting Transformers
[ref 16] IEC 60599, Mineral Oil Impregnated Electrical Equipment in Service- Guide to the Interpretation
of dissolved and free gases analysis.
[ref 17] IEC 60422, Mineral insulating oils in electrical equipment - Supervision and maintenance
guidance.
[ref 18] IEEE C57.104, Guide for the Interpretation of gases dissolved in oil.
[ref 19] IEEE C57.106, Guide for Acceptance and Maintenance of Insulating Oil.
[ref 20] IEEE C37.015-2009, Guide for the Application of Shunt Reactor Switching.
[ref 21] CIGRE Wg 13.07, Controlled switching of HVAC circuit breaker – guide for application
lines, reactors, capacitors, transformers (1st and 2nd part), Electra nr.183, april 1999 and
Electra nr. 185, august 1999.
[ref 22] CIGRE WG B5.05 Modern Techniques for protecting, controlling and monitoring power
transformers, June 2011.
G.3 Other Documents
[ref 23] CIGRE WG B5.37 Questionnaire on Existing Practices of Shunt Reactors Protection,
Monitoring and Control 2007/2008
[ref 24] Carlson, Å., “Shunt Reactor Manual”, ABB Power Technology Products/Transformers,
Ludvika-Sweden, 2002-08-20
[ref 25] CIGRE Working Group 13.02, “Interruption of Small Inductive Current 1 & 2”, ELECTRA,
No.72, p.73 1980
[ref 26] H.Kawada, K.Goto, M.Hanamura, S.Yanabu, S.Nishiwaki, T.Ueda, Y.Murayama,

“Switching Surge of Shunt Reactor caused by SF6 Circuit
[ref 27] E.Zaima, S.Okabe, S.Nishiwaki, M. Ishikawa, T.Nakamoto,Y.Murayama, Y.Kawaguchi,

“Reignition Surges at Reactor Current Interruption in Cable-system GIS” IEEE Trans. on
Power Delivery Vol.5, No.2,pp.947-954 1990
196
[ref 28] S.Okabe, M.Kosakada, H.Toda, K.Suzuki, M. Ishikawa, “Investigation of Multiple
Reignition Phenomena and Protection Scheme of Shunt Reactor Current Interruption in GIS
Substations” IEEE Trans. on Power Delivery, Vol.8 No.1 1993, pp197-202
[ref 29] Z. Ma, C.A.Bliss, A.R.Penfold, A.F.W.Harris, S.B.Tennakoon, “An Investigation of

Transient Overvoltage Generation when Switching High Voltage Shunt Reactors by SF6 Circuit
Breaker” IEEE Trans. on Power Delivery, Vol.13, No.2, 1998, pp472-479
[ref 30] D.F.Peelo, E.M.Ruoss, “A New IEEE Application Guide for Shunt Reactor Switching” IEEE
PES Summer Meeting 95SM607-2 PWRD 1995
[ref 31] IEC1233 Technical Report, “High-voltage alternating current circuitbreakers Inductive load
switching” 1994
[ref 32] E.Zaima, S.Okabe, S.Nishiwaki, M.Ishikawa, K.Suzuki, H.Toda, “Application of Dynamic
Arc Equations to High-frequency Arc Extinctions in SF6 Gas Circuit Breakers” IEEE Trans.
on Power Delivery Vol.8, No.3, pp.1199-1205, 1993
[ref 33] T.Kobayashi, S.Tsukao, I.Ohno, T.Koshizuka, S.Nishiwaki, N.Miyake, K.Matsushita, T.Saida,
"Application of Controlled Switching to 500kV Shunt Reactor Current Interruption" IEEE Trans.
on Power Delivery, Vol. 18, 2003, pp480-486
[ref 34] LAI, Loi Lei, et al, Application of 500 kV Shunt Reactor with Auxiliary Winding System in
Switching Station in China, January 19, 2009
[ref 35] ELECTRA. An International Survey on Failures in Large Power Transformers in Service. Paris:
CIGRE, Ref. no. 88, 1983.
[ref 36] APPLICATION OF NUMERICAL RELAYS FOR HV SHUNT REACTOR PROTECTION (Z.
Gajić, B. Hillstrőm, M. Kockott) ABB Automation Technologies AB Vasteras, Sweden
[ref 37] Transformer PD Diagnosis using Acoustic Emission Technique, Tord Bengtsson, Håkan Kols
ABB Corporate Research, Bengt Jönsson ABB Transformers AB, ISH ’97
August 25-29, 1997, Montréal, Canada
[ref 38] Ray Zhang, Automatic Reactive Switching for HV and LV Shunt Reactive Plant, National Grid
Report, UK, 2001
197
[ref 39] Adaptive Differential Protection for Generators and Shunt Reactors, Relay Protection and
Substation Automation of Modern EHV Power Systems, Moscow – Cheboksary, September 10-
12, 2007, I. Brncic, Z. Gajic and S. Roxenborg.
198

Protection, Monitoring and Control of Shunt Reactors

Caricato da

Informazioni sul documento

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

Protection, Monitoring and Control of Shunt Reactors

Caricato da

Copyright:

Formati disponibili

546

Protection, Monitoring and Control

2.1 Design of shunt reactors

Design and operation

Figure 1 Three-leg shunt reactor core

Figure 2 Five-leg shunt reactor core with three wound limbs

Medium voltage reactors, connected to tertiary windings of transformers, in most cases

Oil or Dry type shunt reactors

Three phase up to: 250 Mvar

Single Phase reactors

Three phase (3-leg or 5-leg shunt reactors)

Shunt reactors equipped with auxiliary power windings

Shunt reactor in the neutral

Variable shunt reactor (VSR), with tap-changer

Shell type core

Figure 3 Reactor construction, Core type (A), Shell type (B).

The electrical characteristics of a shunt reactor that needs to be studied are:

2.2.1 Air-gap of the shunt reactor core

2.2.2.1 Switching in air core reactors

Figure 4 Simple shunt reactor model

The instantaneous voltage value when the breaker is closed.

The derived short current obtained through the limiting values

Shunt reactor Shunt reactor

top value at time I0

Shunt reactor Shunt reactor

Figure 5 Aircore reactor

The magnetizing characteristic of a shunt reactor

Shunt Reactor Characteristic

Figure 6 Typical magnetizing characteristic of a gapped core shunt reactor

Switching in a shunt reactor

The differential equation when a shunt reactor is switched in follows

k  1  k    t  u L k  eq. 25

Calculate the magnetizing current iL :

Figure 7 Test result of inrush

Figure 8 Bus connected, 300kV, 150MVAr, oil immersed shunt reactor

2.2.3 Shunt reactor disconnection

However, the primary current chopping causes a transient, an exponentially decaying dc

Figure 12 Idealized inrush currents for an air core reactor

Figure 13 Idealized inrush currents for an iron core reactor

Ipeak /Ir [pu]

2.2.4.1 Zero Harmonic (DC)

2.2.4.2 2nd Harmonic

Figure 17 Shunt reactor current for operation with over voltage

Figure 18 Harmonics in the reactor current in case of over voltage

Their major applications are:

3.1 Connection to the Power System and Grounding Methods

3.1.1 Line and Bus connected reactors

Figure 19 Solidly grounded three phase reactor directly connected to line

3.1.2 Tertiary winding connected reactors

3.2 Effects of Shunt Reactors on Transmission Line Voltage

Figure 23 Nominal-π circuit of a transmission line

Vs  AVr  BIr eq. 34

Now we need to calculate Vs and A.

4.1 Fault types in Dry-type reactors

4.2 Fault types in oil immersed reactors

1. High current phase-to-phase and phase-to-ground faults.

4.3 Failure rates of shunt reactors

The definition of failure rates yields;

4.4 Turn to turn faults

The main principal hazards to a shunt reactor are:

General protection requirements

Additionally, for oil immersed reactors:

Shunt reactor arrangements

Figure 26 Typical shunt reactor connections